There’s no reason for non-recourse

We don’t know exactly what Timothy Geithner has in mind for the “Public-Private Investment Fund”. But we do have a few hints. First, we know that among its purposes is that it

allows private sector buyers to determine the price for current troubled and previously illiquid assets.

And we also know, that on the very day Mr. Geithner offered his outline of a financial stability plan, the Federal Reserve announced its intention to expand its Term Asset-Backed Securities Lending Facility, or “TALF” to up to a trillion dollars, coincidentally the round number that Geithner suggested the “PPIF” might expand to. Hmmm. What is the TALF again?

Under the TALF, the Federal Reserve Bank of New York will provide non-recourse funding to any eligible borrower owning eligible collateral… As the loan is non-recourse, if the borrower does not repay the loan, the New York Fed will enforce its rights in the collateral and sell the collateral to a special purpose vehicle (SPV) established specifically for the purpose of managing such assets… The TALF loan is non-recourse except for breaches of representations, warranties and covenants, as further specified in the MLSA.

Does your head spin, acronym upon acronym, non-recourse, warranties, and covenants? Well, unspin it. The New York Fed is telling us, in plain and simple legalese, that it is planning to make a very generous gift to investors that participate in this program (and indirectly to the banks that sell assets to them). A non-recourse loan bundles an ordinary loan with an option to “put” the collateral back to the lender instead of paying off the loan. Sometimes this is not much of a gift: When a pawnbroker lends you half of what your Fender Stratocaster is worth, and the fact that you can surrender the guitar rather than pay off the loan is cold comfort. But if someone fronts you substantially all of what an asset is worth, and the value of that asset is uncertain and volatile, then the put option bundled into the “loan” becomes extraordinarily valuable. If the asset appreciates, you take the profits and “ka-ching!”. If the asset falls in value, the lender takes the trash and eats the loss.

A near-the-money option is itself a valuable asset. Offering non-recourse loans to participants in the PPIF would directly contradict the program’s goal of “allow[ing] private sector buyers to determine the price for… troubled… assets.” Private sector buyers would not be pricing the assets themselves: they would be pricing a portfolio containing a troubled asset and a free, three-year put option, courtesy of the Fed. Depending on how much of the transaction the government is willing to finance, the value of the put option could represent a substantial fraction of the value of the asset being priced. This is a subsidy, that would be incorporated in the sales price of the asset and split by banks and private investors. It amounts to the government bribing investors to certify banks as more solvent than they are, by overvaluing bank assets in subsidized purchases.

John Hempton wrote a very brilliant essay on what it means for a bank to be solvent. If you haven’t read it, go do. Hempton’s definitions 2 and 3 of bank solvency — current accounting value (which implies mark-to-market valuation for many assets) and economic value as an ongoing enterprise — diverge because the cost of funding for investors in risky bank assets is unusually high. Under these condition, Hempton reasonably suggests, private fund managers will be unable bid assets up to their best estimates of “hold-to-maturity value”, less a “normal” risk premium, because investors are desperately unwilling to hold anything other than government guaranteed securities. Definition 3 is a very generous view of what it means for a bank to be solvent, because it implies that the actual market risk premium is wrong, that an estimate of hold-to-maturity asset values by a reasonable analyst, even accounting for risk, would put those values above current market bids. But in evaluating bank solvency we should be generous: Since an insolvent bank must be nationalized (reorganized, received, conserved, preprivatized, whatever), we should try to avoid declaring as insolvent banks that do have positive economic value, since that would amount to a capricious expropriation of private property.

But generosity in evaluation is distinct from a generous cash gifts from taxpayers to banks and investment funds. What is required to get a generous but still accurate evaluation of bank solvency is inexpensive funding, so that analysts willing to bet on what a “toxic asset” is worth can borrow the funds they need to back their spreadsheets with shekels without giving away all the upside to nervous lenders. What is not needed, what is in fact positively counterproductive, is to give investors a special bonus in the form of a free option if they buy the asset. This guarantees that assets will not be accurately priced (they will be overpriced), and reduces analyst incentives to value assets carefully and generate reliable market prices.

I actually think having the government offer cheap, full-recourse loans on a maturity-matched basis to investors willing to bear the risk of holding currently disfavored assets is a clever idea. (“Maturity-matched” means investors don’t have to worry about margin calls: as long as they get the long-term values right, they can ride out any tempests in mark-to-maket prices.) We do need a market in these assets, and if it is true that funds availability for people willing and able to bear the risk of ownership is preventing such a market from arising, then by all means, that’s a “market failure” the government can correct. But the key point is that a market price is the price at which private parties are willing to bear that risk. If funds are provided non-recourse, much or all of the risk of ownership is absorbed by the lender. Any prices that result from “private” purchases by investors funded at high-leverage on a non-recourse basis are not market prices at all. Such prices would be sham prices, smoke-and-mirror prices, sneaky off-balance sheet public subsidy prices.

We are all tired of the lies, Mr. Geithner. By all means, let nationalization be a last resort, and do all you can to offer liquidity to private parties willing to take both the upside and downside of speculating in questionable paper. But if you keep nationalizing the downside and privatizing the upside, it will not be very long at all before the public concludes that stress tests and market prices are just a sleight-of-hand for Davos man while he picks our pockets, again. Act fairly, and you may end up nationalizing the worst few of the larger banks. Keep up the games, and we will insist that you nationalize them all. It is getting hard to believe that there is a banker in the land who has not already robbed us. Eventually we will tire of drawing fine distinctions.

Afterthought: There’s another way to generate price transparency and liquidity for all the alphabet soup assets buried on bank balance sheets that would require no government lending or taxpayer risk-taking at all. Take all the ABS and CDOs and whatchamahaveyous, divvy all tranches into $100 par value claims, put all extant information about the securities on a website, give ’em a ticker symbol, and put ’em on an exchange. I know it’s out of fashion in a world ruined by hedge funds and 401-Ks and the unbearable orthodoxy of index investing. But I have a great deal of respect for that much maligned and nearly extinct species, the individual investor actively managing her own account. Individual investors screw up, but they are never too big to fail. When things go wrong, they take their lumps and move along. And despite everything the professionals tell you, a lot of smart and interested amateurs could build portfolios that match or beat the managers upon whose conflicted hands they have been persuaded to rely. Nothing generates a market price like a sea of independent minds making thousands of small trades, back and forth and back and forth.


149 Responses to “There’s no reason for non-recourse”

  1. Steve:

    Unless I am eligible for a non-recourse loan, I say no dice. Let Citigroup file bankruptcy. Who will get these non-recourse loans? This program is corruption at the highest levels. Tell you what. You and I will form an investment partnership. We’ll each put in say $10,000. Now we go to Geithner for our $20 million non-recourse loan. Ain’t gonna happen. No deal.

  2. JKH writes:

    I’m a bit surprised at the general b-sphere response to Hempton’s article, but probably shouldn’t be. It’s brilliant, but I wish it were slightly clearer. I wonder how many understand what he is saying (including me).

    E.g. it seems to me the critical characteristic of his test 3 is the inclusion of an assumption for future accrual book earnings over some specified time horizon for the measurement of solvency. Tests 1 and 2 are based on existing book capital measures, representing various hurdle points for the definition of solvency on that basis (e.g. full regulatory requirement; > 0 capital; 1/3 regulatory requirement).

    But unlike test 3, in no way do his test 1 and 2 capital measures include an assumption for future accrual book earnings, or equivalently, the present value of future accrual book earnings as an incremental valuation component. This is the biggest conceptual difference in his various interpretations of solvency.

    The incorporation of FE or PVFE allows for the bank to recapitalize itself over time, which is fundamentally why it is a more generous measure. The easiest example is that of a bank whose current equity, if properly measured, is wiped out by a properly measured trading book mark-down, but which can be recapitalized on the basis of a currently properly measured accrual book generating earnings over the next 5 years. That recovery would not be captured by test 1 or test 2 but would be by test 3.

    In that sense, because test 3 includes the contribution of FE or PVFE, it is a dynamic scenario capital model, whereas tests 1 and 2 are static capital models. Being dynamic, it must assume ongoing viability, which is where the assumption of reasonable funding costs comes in, but only indirectly as the required assumption (for consistency) underlying an income statement approach (including periodic accrual FE) rather than a balance sheet approach (instantaneous book and market value view).

    Hempton doesn’t seem to fully emphasize or highlight this FE or PVFE component as such, when it is the most fundamental distinction in the comparative discussion across all of his models. (The other dimensions are loss expectation and discount rates for loss expectation, which can affect the value attributed to both accrual and trading books). But his post is generally brilliant.

    The following was my admittedly weaker and less robust effort at tackling what is essentially the same subject matter, but from the vantage point of the effect of mark to market on crisis response engineering, as opposed to the interpretation of solvency per se:

    …. (join)…


  3. Nemo writes:

    Another good article, Steve. I like to call this the “Geithner Put”.

    Note that even if the loans are full-recourse, a private equity firm could set up multiple shell corporations to take on independent loans, thus making them effectively non-recourse again. A private lender would not be fooled, of course, but would the government really apply the same discipline?

    But I do not think they will bother. The loans will be non-recourse. It is the only way to gift hundreds of billions of dollars to the banks under this framework.

  4. Kevin Kleen writes:

    It’s not a free put – there are haircuts required which vary for each asset class and the term of the facility. The issue is whether or not the haircuts are large enough ( for example, I have no idea if a 10% haircut on the AAA tranche of a 5 year prime auto loan facility is big enough).

  5. Addendum to your afterthought. Slice it down to individual houses. Although I’m told the paperwork to do this with existing CDOs could take years.. :(

  6. Gordon Gekko writes:

    Unfortunately all of this is now just rearranging deck chairs on the sinking Titanic. It’s all useless discussions. The old order’s expiry date has arrived and no amount of machinations – no matter how brilliant or well intentioned – can save it. It’s just nature at work.

  7. RueTheDay writes:

    Steve – Agree on the non-recourse loans. Disagree on the gov’t providing them cheaply (e.g., at below market rates) being a good thing. But my biggest concern wasn’t mentioned here – the PPIF plans I saw provided some sort of asset floor (typically a loss sharing arrangement). I think you really need to take all three components together – gov’t providing cheap, non-recourse loans together with asset price floors to see the big picture – yet another example of privatizing profit and socializing losses. If that’s the ONLY way to breathe some tentative life into this market, then I reach a different conclusion – the securitization model itself is dead.

  8. beezer writes:

    In think the TALF program horse has already left the barn. So that socialism for the wealthy program is up and running. Read the FAQs on this program. It specifically allows for the formation of investment spinoffs, as long as the parent company is US based.

    Once the public finally starts to understand the underlying principal at work here, socialism providing a moat for wealth, all heck is going to break lose. The shame is that when this happens, and it will happen believe me, Obama’s terrific programs for tax reform, energy reform, universal health care, and education will be flushed.

  9. writes:

    Thomas Barker – Not a bad idea, try

  10. writes:

    Same as the comment above but with a link that works –

  11. David Pearson writes:

    I agree that the TALF is non-recourse and will stay that way. Recourse is a good suggestion, but its too late.

    The essence of the non-recourse model is that the NY Fed does not need credit officers to examine risk. That allows virtually any and all comers to seek TALF funding. Those comers include, primarily, hedge funds (private equity, in this case, would be acting as if it were a hedge fund).

    So TALF is a time machine: enter in March 2009, emerge eighteen months earlier, in a time when prime brokers offered hedge funds unlimited leverage with few restrictions.

    Every action the government takes is intended to hide the true cost of stabilizing the financial system (such will be the case with insurance provided on the public/private assets). As long as this is the case, there will likely, in the end, be a bull market in pitchforks.

  12. JD writes:

    Please please please know the difference between a put and a call. I realize by put/call parity that the true option value, ie time value, is the same but you are looking at this wrong. The FED is selling a free call, not a put. Here is a simple example.

    Neglect transaction costs, interest rates, and ‘haircuts’ or margin capital for the sake of simplicity. Assume BofA owns an asset that happens to be Citi stock (could be any asset, but this clearly shows my point) currently priced at $2.00 per share. If I understand non-recource correctly, BofA delivers the Citi stock to the FED in exchange for $2.00 per share. Then over the next three years, BofA has the RIGHT but NOT THE OBLIGATION to repay the $2.00 and receive the shares for repayment of the loan. This is a call, BofA has the right to purchase the Citi stock for $2.00. This P/L profile is EXACTLY the same as selling the shares to the FED for $2.00 and receiving a free three year CALL with strike of 2.

    FYI: Citi Jan 2011 2.5 calls are 1.01 bid

  13. Toxic credit cultures produce toxic assets.

    At some point, to clean up the mess, President Obama is going to have to make character judgments and stop doing business with the same financial tricksters that got us into the mess (Rubin, Summers, and Geithner for starters). But who is his closest advisor? Rahn Emanual, who got rich quick with a hedge fund in between political gigs.

    Franklin Roosevelt, in the Depression, confident in his place in the artistocracy yet having his own principles, was content to fire the managements of bad banks and make any number of other judgment calls required to restore some sense of propriety and ethical integrity to the financial system. Many people loved him for it, although not many Republicans.

    Does the current President have the judgment, courage and confidence (all three are required) to clean up the financial system? Not so far. He will need to clean his own house before the American people are going to buy into any Treasury plan.

    Half of Americans opposed the [extraordinarily ill-conceived] stimulus bill, according to the surveys I’ve seen. A majority of Americans oppose giving the banks any more money.

    My working hypothesis, borrowed from Strauss and Howe, is that the American social contract is broken, and that the crisis gets worse until Americans feel they’ve got a fair system again.

    Geithner’s proposals are laughable. They’re like a jobs act for unemployed investment bankers, and yet another opportunity for massive wealth extraction from tax-paying Americans by the plutocrats (aka oligarchs) in New York and their friends.

    Toxic credit cultures produce toxic assets. The President needs to see this, and do the right thing. Get some fresh blood in to advise him before these guys try to blind us all with science again.

  14. JD — This is an old discussion, and the crucial point is what you allude to, put-call parity. The implicit option in a non-recourse or limited liability leverage arrangement can be viewed as both a put or a call, specifically you can view the levered equity holder as long assets and a put option or as long money and a call option. Write that as an equality…

    long assets + long put = long cash + long call

    and rearrange by subtracting the long options (equivalently adding short options) to both sides of the equation and you get

    long assets + short call = long cash + short put

    which is the canonical form of put/call parity:

    covered call = fiduciary put

    (I’m neglecting the exponential factor next to the cash indicating the time value of money. That’s quantitatively crucial, but qualitatively unimportant here.)

    I find the “put or call” question to be very much like looking at one of those optical illusions: is it two faces, or a vase? an old lady or a young maiden? Sometimes I find it more intuitive to think of the limited liability levered equity-holder / non-recourse-financed asset holder as a dude with cash and a call option, at other times more intuitive of her as a broad with assets and a put option. I wrote it as a put option, because I think that’s more intuitive for most readers, since in terms of the superficialities of ownership, the borrower “owns” the assets but can “put” them back. But in economic substance, it is equally accurate to view the situation the way that you have described, and since the assets are pledged as collateral, you might find that more intuitive. Again, is it a vase, or two faces?

    All — Sorry, I don’t have time right now to participate extensively in the comments thread. I felt I had to respond to an accusation of not knowing the difference between a put and a call option, lest my finance testicles shrivel away to nothing. But I won’t be able to give much in the give and take.

  15. OK. I’m really not supposed to be doing this. But apropos the last comment and to give commenter JD a bit more of his due, consider what the first equation above looks like if no one is holding any cash. There is cash involved, a loan was advanced, but the recipient used the cash to buy the asset. So subtract cash from both side.

    Here’s the original equation, just a rearrangement of put/call parity.

    long asset + long put = long cash + long call

    But what happens if in order to get to this position (either version), you had to shell out the cash value of the asset? In other words, in order to get the option and the asset, you had to shell out cash? Then subtract cash from both sides:

    long asset + long put – cash = long call

    So, this is the position of someone who receives a nonrecourse loan to completely finance an asset. You can view it “naturalistically” as an ordinary financing arrangement — long asset, short cash — plus a put option. Or you can view the _entire arrangement_ as a synthetic call option. This, I think, is the sense in which JD prefers to think of nonrecourse financing as a call option. Either view is “right” — a portfolio including a put or a standalone synthetic call — but JD’s view does have an advantage, which he exploits in his example: The value of the whole arrangement can be summed up by the price of a call option, which may be observable.

    As JD suggests, if you take Citi itself as a toxic hard-to-value asset, then the free option associated with 100% 3-year nonrecourse financing is equivalent to a 3-year call option on Citi. According to JD, a two-year call goes for about $1, and Citi itself goes for $2, so 100% nonrecourse financing amounts to giving takers around 50% of their investment’s value. Of course, the financing wouldn’t be 100%, and these relationships are rough, but that’s a nice way of getting a rough sense of the magnitude of the subsidy. If investors paid “fair value” for Citi under these terms, they’d walk away with 50% expected profit on the notional value of shares purchased (notional in the sense that their actual cash outlay is de minimis, actual profit on capital committed wold be much larger). Obviously, competitive investors would bid up the notional price of Citi until the value of the call option is included. That is, they’d end up paying close to 50% more than the real value of Citi in order to capture the option value of the arrangement. So Citi’s “market value” would be inflated by 50%. If instead of Citi itself, the market were valuing “toxic assets” that Citi holds under these terms, and the option value of these were similar to that of Citi as a whole (which needn’t be the case), then Citi assets would appear correctly valued at current book as long as they aren’t overstated by more than 50%! That seems like a pretty low bar for a “stress test”.

    I was perhaps a bit too prickly in responding to JD the first time around: he added an important thought experiment and means of quantifying the subsidy to the discussion.

    (It remains true that you can view non-recourse financing “naturalistically” as a portfolio that includes a put option or “synthetically” as a cobbled-together call option, and I’d not change the original piece because I think the synthetic view would confuse most readers.)

  16. CapVandal writes:

    As far as the test 3 is concerned, it is dynamic in one sense, can be thought of as valuing the assets and liabilities at market, assuming that you cover the entire balance sheet.

    That is, deposits are cheap funding and are worth more then face value. You could, in theory, sell deposits for more then their book value with the excess considered either an asset or contra liability. In addition, the debt is worth less then par, which would also increase capital.

    The idea of a real m2m accounting world with all parts of the balance sheet marked consistently is unworkable. A perfectly healthy business *could* melt down (or get a windfall) if the market perception of risk changes significantly, even in the absence of any other factor. If your only business is derivatives and you mess up on volatility, then ok — that makes sense. Real businesses need stability and most balance sheet items don’t have anything approximating efficient markets.

    Fortunately, banks don’t mark whole loans to market. They book them at amortized cost less a provision for loan losses. These (loan losses) are under stated, since they don’t fully reflect the known lags in a business cycle. However, they also don’t pretend to know the inflection point. Further, the direction of the error is pretty well known and thats one reason they hold capital — to use it as a buffer. If economic capital can’t drop under stress, then why even hold it?

    Regular banks don’t hold the alphabet soup of assets, outside of investment banks or those with capital markets operations.

    The plan “could” work but only if the assets are already booked way below par. And if the current booked values seem attractive if they can be bought with leverage. As long as private money takes the first x% of the losses, they might be able to use enough leverage to make the entire idea work. I’m not holding my breath.

  17. JKH writes:

    So as I understand it, a non-recourse loan from the lender’s perspective is equivalent to the combination of an asset and an option, consisting of either:

    a) Long the collateral, short a call on the collateral (JD)

    b) Long the risk free asset, short a put on the collateral (SRW)

    And from the borrower’s perspective, the non-recourse loan is equivalent to either:

    a) Short the collateral, long a call on the collateral (JD)

    b) Short the risk free asset, long a put on the collateral (SRW)

    I guess I don’t see the natural/synthetic distinction so clearly here, as the construction seems fairly symmetric.

  18. Looks like a “gift” from the taxpayers to the investors to me.

  19. MethodMan writes:

    Excellent post.

  20. ccm writes:

    If the PPI is not the same as TALF, but instead a structured deal, couldn’t it result in fair prices even with non-recourse financing? I’m thinking specifically of the following case: The private investor puts up at least 25% of the price (determined by the private investor) and takes the first loss position. The government provides 75% non-recourse financing and only incurs losses after the private investor’s stake is wiped out. This would, of course, mean that the private investor could receive no income or principle from the investment until the government was fully compensated.

    While I don’t know that such a plan would do much to establish bank-friendly prices, it might serve to silence the people who claim that the only problem is lack of liquidity for private investors.

  21. JKH — “Synthetic” only in this sense:

    long asset + long put – cash = long call

    The accounts of the borrower look pretty much like the left hand side of the equation above: they show ownership of an asset, and a liability to the lender. The put is not entered as an asset on the accounts, but neither is the RHS call option. To a naive reader of accounts, the borrower owns the asset and shows a debt, but has the option of erasing both entries if their sum is negative.

    The sensitivity of cash flows to asset values, summing all three components, constitute a call option, as JD suggests. But per the accounts, the situation of the borrower more closely resembles that of someone who has synthesized a call option than that of someone who has purchased a simple contract. And presuming the borrower considers herself the owner of the purchased asset, the act of exercise takes the form of a put (although again, I am appealing to intuitions about investor perceptions), exchanging ownership of the asset in exchange for a cash payment (in the form of debt extinguishment).

    Really, we’re back to the two-faces-vs-a-vase thing. This is actually my favorite example of put-call parity. (Not the asset-specific nonrecourse loan, but the optionality of a leveraged, limited-liability firm.) Try to explain how a covered call must be the same as a fiduciary put by arbitrage arguments and eyes glaze. But tell two stories about a firm — in one the owners skip town and stiff the bank with crappy assets instead of paying their loan, in another the pawnshop/bank holds the assets in hock but equity holders have the right to take ’em for a price — and it’s easy for people to see that the economic substance is the same, but one version describes exercising a put, while the other involves exercising a call.

    But, when I’m not explaining put-call parity, I usually tell the put version, because most people’s intuition is that equity owners own the firm but might abandon, not that banks own the firm but might be forced to surrender. There’s a disadvantage — the call version describes the net sensitivity of the parties’ positions to asset values more elegantly. But it seems more “natural”.

    Again, this is a distinction with little difference, and per JD’s exercise, the concision of the call view simplifies valuing the net position in a very nice way.

    ccm — the badness of non-recourse finance for the purpose of finding a market price is related to the degree of leverage he government is willing to fund on a non-recourse basis. any non-recourse financing implies a subsidy of some option value, but the value of that subsidy becomes negligibly small if the option is way “out of the money”, that is if it’s only worth exercising on a very substantial loss. i’d probably shut up about this if the gov’t provided nonrecourse financing on a 50% basis (one dollar loaned for every private dollar invested) or less. requiring 25% down would be better than requiring 5% down, but still implies a pretty big subsidy, given the uncertainty surrounding the valuation of these assets.

  22. JIMB writes:

    No one would sell to the Fed unless the Fed were overpaying. This is just an explicit subsidy to those that have brought us to this point. And the accounting gimmicks are part of the problem, not the solution. Things should be booked at lower of cost or market at all times … and this would have been avoided because the “capital” wouldn’t have (fictitiously) appeared for the additional lending. I must also say that the market fired a lot of these “managers” (public and private) in 1987, 1992, 1998, 2001, and now in 2007. It’s the Congress and the Fed providing bailouts and keeping this game going against all of us, not the Congress trying to “save” the system that was co-opted by evil aliens. Congress does this because they want a market to fund their overspending. The Fed does it for the connected financial insiders. The Solution? Deposits are only 60% of obligations for Citi and JPM (70% for WFC). Convert the non-deposit bondholders to equity and go from there. Just to give the reader an idea of what has transpired so far: roughly 10 trillion of private debt was converted to public debt … there is only 12 trillion in total residential mortgages!

  23. mencius writes:

    Hempton’s post is a good one in many ways, but I am puzzled by it for a slightly different reason than JKH.

    It seems to me that the interesting problem is Hempton’s test 5 (“liquidity solvency”). Apparently Hempton believes that a banking system which passes this test, at least without massive butt injections of Fedianabol, is an almost cosmically unreachable goal – like Alpha Centauri, or world peace, or a good diet soda. Whereas I think a liquidity-solvent banking system is, or at least ought to be, normal. Probably this is because Hempton is a banker and I’m not.

    Hempton does not describe test 5 with much precision, which makes sense if he believes it’s unrealistic. I don’t, so let me clarify a bit.

    A liquidity-solvent bank is one which can fulfill all its obligations without incurring any new obligations. This definition is only useful if “new obligations” (ie, lending to the bank) is defined to include both (a) loan rollovers, and (b) asset sales.

    These restrictions are perfectly reasonable. A rollover is a new loan from an existing customer. A sale of a loan is a new loan whose terms happen exactly cancel an old loan. Moreover, a “demand deposit” can be seen as a continuously rolled-over epsilon-term loan.

    In other words, a liquidity-solvent bank is one that can pay all its creditors without borrowing from new creditors. Duh. Is this kind of stability really too much for us to expect, in the 21st century, from a financial system?

    And note that liquidity-solvency is the question that creditors, in a free market, want to ask. They want to ask: can this institution make its payments without borrowing more money? If the answer is “no,” it should not be able to borrow more money, and it must be restructured in a way that treats all its creditors fairly. Ie: bankruptcy.

    While it is incorrect to describe non-liquidity-solvent banks as Ponzi schemes, the dependence on a continuous inflow of money is the same. It is probably wishful thinking to expect the phrase “Bagehot scheme” to catch on, but it is certainly unsurprising that a reversal of the credit flow would expose Ponzis as well as destroying banks.

    There are a lot of reasons we don’t have liquidity-solvent banks. But IMHO they are all, as we say in my line of work (programming), “historical.”

    One of the main obstacles is that there is no way to get from standard “T” accounting to liquidity solvency. You cannot calculate liquidity solvency by summing on a bottom line. It is useless to talk of the “value” of an asset. A liquidity-solvent financial intermediary must match every pledged outgoing payment with an expected incoming payment, on or before its date. Is this difficult? In the 21st century? No, it’s just different. Change is good.

    I should also note that maturity matching, contra an impression I get from reading the above, is orthogonal to leverage ratio. There is absolutely no technical problem with running a maturity-matched, highly-levered book. In fact, high leverage is much safer in a maturity-matched environment, because there are no maturity crises to blow up your risk calculations.

  24. mencius writes:

    Rereading Steve’s post, I think he is actually saying the same thing I’m saying about maturity matching.

    In which case, we agree that matched maturities do not prevent you from getting a margin call if you’re levered and your assets go south! Maturity matching may make prudent lending stable and profitable, but there is no formula for prudence. (Despite many rumors to the contrary.)

  25. babar writes:

    i don’t actually see the big issue with non-recourse, and i don’t see how it works any better with recourse loans. a few reasons:

    — the fed has to monitor the credit of the debtor continuously to make sure that they have the resources to pay back the loan in case of default, even if the collateral goes to zero. this means that the fed is employing credit analysts, which is one of the things they say they cannot do well.

    — potential buyers will just set up their own legal entities to get around this restriction. if i was going to set up a fund to invest in TALF and solicit funds from qualified investors, they would not be liable for losses over 100%. if they were, or if there were requirements for investors to post additional collateral upon defaults or downgrades, this facility would have the same problems that we’ve seen with these securities in the past — inconsistent ability to lever leading to market participant failure leading to market failure.

    — there are other “lighter” versions of the above — these securities become more valuable to investors who can figure out a firewall than to those who can’t.

    in my mind the problem is not “recourse” versus “non-recourse”. it’s the percentage of money coming from government sources — this is supposed to be more than 80% the last i heard. if you make this percentage much less — thereby lowering the strike of the option — you are making it a much fairer game. and on the other hand, if the fed/treasury is coming up with 80% of the money i think they should just buy the securities outright. if they need a free market to assign prices, they should run one in-house.

  26. babar writes:

    mencius> One of the main obstacles is that there is no way to get from standard “T” accounting to liquidity solvency. You cannot calculate liquidity solvency by summing on a bottom line. It is useless to talk of the “value” of an asset. A liquidity-solvent financial intermediary must match every pledged outgoing payment with an expected incoming payment, on or before its date. Is this difficult? In the 21st century? No, it’s just different. Change is good.

    Looking at Hempton’s post, what he is saying is that banks the way they are set up now can’t survive a bank run because their reserve capital comes from deposits and deposits are not locked in — i think you are saying that it should be possible for a bank to remain solvent even if there _is_ a bank run and deposits go to $0.

    I don’t want to hijack the thread and make it about this idea, but I don’t think this is simple. On one hand, a lot of products don’t fit into this — you could have no products with contingent payments or draws — no callable bonds and no guaranteed lines of credit for example. You couldn’t make any assumptions at all about currency movements — you would have to run completely separate accounts in each currency. On the other hand, you would still have to set aside a separate reserve for credit losses, which are indeterminate, and so you couldn’t make any strict guarantees about indefinite future solvency. So it is difficult for me to see what you gain by this.

    At this point in time a lot of what banking does is provide semi-fungibility (through the medium of future money) between varying assets. I think what you are saying is “don’t do that — if they are not fungible you have to keep them in separate accounts.” in that case it’s difficult to see what the function of the banking system is.

  27. JKH writes:


    I didn’t think about it before, but Hempton did spend surprisingly little time examining the relationship between maturity mismatching and solvency. The conventional view is still embedded – the usual test of acceptable maturity transformation risk is not whether it is zero, but whether it is sufficiently close to zero so as not to constitute an imprudent level of mismatch. The definition of “sufficiently close” or “imprudent” is not clear. The conventional view also is that commercial retail based banks are closer to zero MT risk or balanced in this sense than wholesale or investment banks. This is partly but probably not entirely due to FDIC protection on small deposits.

    I come from the school where solvency is considered to be a balance sheet equity issue rather than a liquidity issue, but that’s just a matter of definition.

  28. JKH writes:

    re above:

    “sufficiently close” is probably a sloppy way to put it, but you get my point, I think

  29. mencius writes:


    You have it right – there are no bank runs in a liquidity-solvent banking system, for precisely this reason. Since the last ones out do not get screwed, no one has any reason to run for the exits.

    Maturity matching does not preclude contingent instruments. It just requires the bank to assume, for purposes of accounting, that the contingency is always exercised.

    It is true that this may make some instruments undesirable – it is hard to see a role for the demand deposit, for example, in a liquidity-solvent bank. There are no productive investments at zero term. Those who want to hold cash should hold cash.

    As for currency risk, banks should absolutely run separate balance sheets in each currency. Presumably the phrase “Eastern Europe” means something to you. Currency risk, at least in our present financial environment, has no relationship to the normal distribution and often (as in Eastern Europe) is simply hiding our good old friend, liquidity risk.

    The analogy is an excellent one: an asset in Latvian lats should not be used, under any standard of sane accounting, to balance a liability in US dollars. Just as a 2019 dollar and a 2009 dollar are not equivalent goods, a 2009 dollar and a 2009 Latvian lat are not equivalent goods. (At least not without some kind of hedge that eliminates the currency risk.) Frankly, it appalls me that anyone ever thought this was a good idea.

    I would say the purpose of a (healthy, liquidity-solvent) banking system is to connect borrowers with lenders, while diversifying across default risks. Eg: Joe wants to lend money across 30 years. Perhaps he is buying an annuity. Sarah wants to borrow money across 30 years. Perhaps she needs a mortgage. The task of banking is considerably more complicated than saying “Joe, meet Sarah,” but that’s the idea.

    A liquidity-insolvent banking system functions differently. It finds Stanley, someone who really just needs a secure mattress to put his Krugerrands under, and convinces him that his Krugerrands are “in” the bank. Then it lends them to Sarah for 30 years. On what planet is this OK? Ours, apparently,

    The liquidity-insolvent bank is therefore in the business of creating the illusion that there exists more present money than there actually is. When operated on a gold or other hard-money standard, this is extremely dangerous for reasons that, while not obvious to everyone,

    Such was the so-called “classical gold standard” of the 19th century, the British system if you will, which in practice was heavily watered with paper. The rule Bagehot set down was that 10% rates would “draw gold out of the ground,” stopping any attempt to collapse the pyramid. But after WWI, the debt was too big, and the pound’s gold cover was simply too low to live.

    (The correct move would have been a substantial devaluation, but for some reason this was not politically acceptable. Historians have few nice things to say about British politicians of the early 20th century. Similarly, FDR probably should have devalued the dollar to $100 or even $200 an ounce, or whatever level would have allowed the US to remain on gold.)

    But when a liquidity-insolvent financial system is operated on a fiat standard, it can always be bailed out by the State. Even when the bailout does not need to happen, the lender of last resort is effectively issuing free options to banks. Just on account of because they’re banks.

    Of course, babar, since you don’t have a problem with the non-recourse loans either, no one can say your position is inconsistent!

    However, the absurdity of retaining, in the 21st century, a system that was originally designed in the pre-democratic era (the Bank of England has been transforming maturities since 1694) to allow sovereigns to effectively debase a gold-standard currency without actually clipping coins, strikes me as considerable. Especially since we don’t even have a gold standard! Talk about throwing out the baby and keeping the bathwater.

  30. mencius writes:


    Precisely. The conventional view is that maturity transformation is like the temperature of Goldilocks’ soup – you can have too little liquidity risk, too much liquidity risk, or just the right amount of liquidity risk. Which is true, of course, for default risk. So it’s easy to see how the idea persists.

    Whereas from the Austrian perspective, liquidity risk is more like the level of fecal coliform in Goldilocks’ soup. There may be some acceptable level, but it is not a high level. And less, surely, is always better.

    As you note, however, it is impossible to construct a quantitative (or even qualitative) metric for what “the right amount” of liquidity risk might be, whether for an institution or for an entire financial system (it’s perfectly safe to transform maturities if you’re the only one doing it). To me this seems like a good tip that something is not quite right with the whole design. Note that there is no difficulty in crunching the numbers for default risk.

  31. mencius writes:

    “Ours, apparently,” should be followed by a period. “Not obvious to everyone,” should be followed by “but obvious to you.”

  32. JKH writes:


    Non-recourse financing

    = Asset + put + liability

    = (Cash + call – put) + put + liability

    = Cash + call + liability

    = Call

    = Synthetic Call

  33. winterspeak writes:

    Enjoying the discussion here, and don’t want to throw things off. Just a few observations:

    SRW: Excellent explanation of the embedded option in the latest Geither plan. It’s appalling that he’s been working on this for two years and we are no farther along than this. Based on the comments he, and other Obama administration officials have made, is that they really do see this entire recession as an aberration, a liquidity crises, and not a crises of solvency. I would also love to know who the bondholders are, as the degree of political protection they have exceeds that enjoyed by Obama’s daughters.

    JKH: I was similarly half-plussed (is that a word?) at the John Hempton article. But he put forward definitions of solvency when those were needed, so that’s all well and good.

    Personally, I found Hempton’s analysis of capital requirements most interesting. The degree to which they are pro-cyclical is as useful, in a crises, as a matrix of covariants that never goes to 1. If you calculated Tier 1 capital I’m confident that, at any reasonable mark to market, banks are far below required regulatory levels, and may even be negative. Their ability to lend depends upon their ability to get funding from the Government. Geither’s unwillingness to nationalize this Potemkin market is appalling.

    More importantly, if you accept that banks are pro-cyclical, it also points to the best course of action to improve the system — improve household balance sheets so they start borrowing again. Obama has slated $400 (not a misprint) for some workers, an amount that is laughable. The committee today to slash the deficit (which will gut HHs, and kill AD) is leading us to Japan style stagnation with European style unemployment levels. The good news is that once AD stabilizes at its lower level, even with high unemployment, expect to see financials improve earnings quickly, especially now that labor has been well and truly subdued. Is it time now to go long Financials?

    To sum, the focus on supply (banks, interest rates) ignores the real constraints that exist on the demand side (HH balance sheets, prices). Repairing the HHs will fix the banks. Giving money to the banks is simply a transfer from HHs that save, to bank employees, equity holders, and bond holders. Not helpful. An example of this is ZIRP — right now it’s doing nothing to goose borrowing but has removed considerable interest income from the private sector. The private sector will continue to save until it has enough. The Obama administration’s actions inspire the private sector to save more and more and, at the same time, make it harder and harder for it to do so. This will not have a positive outcome.

    MENCIUS: You have convinced me of the benefits of no-MT. That said, you can run MT on a fiat currency, and I don’t think the benefits of a gold standard outweigh what a competent Govt can do with its own paper. What we have now is a fiat system run by people who, deep down, believe we’re still on a gold standard world. JKH, who I respect deeply (sorry to pick on you) believes that a balance sheet entity for a combined Federal Government would have a positive equity value. But that is a different discussion and I’m really not going to derail things here. But we are left with the worst of both worlds — no sound control over (fiat) money supply with all of the operational constraints of a gold standard.

    Therefore the “solution” is to transfer additional money to financiers, in the hope that somehow, over-indebted households will get further in debt.

  34. mencius writes:


    I’m glad someone is convinced!

    I would rather see the fiat system flattened, stabilized, and de-MTized, because I think fiat currency is perfectly fine and in fact slightly more optimal than gold (the money supply can be fixed). However, I do not expect this to happen. It would require a government several orders of magnitude more thoughtful and effective than the one we have.

    What I expect is one of three outcomes – curiously, all desirable. In the long run. I list them in order of descending desirability.

    (a): Present conditions continue indefinitely, until the world’s savings flee, slowly at first and rapidly in the end, to gold. Eventually, not even governments can resist the trend, and all fiat assets have to be revalued (at a much lower price than could be attained at by exchanging them now) in gold. New mechanisms and institutions for gold finance are created. Civilization resumes its forward progress with a sound monetary system.

    If (a) happens, it happens because the fiat system can no longer create fiat investments that generate a positive gold-on-gold return. Thus there is no force to suck money out of gold and into paper, as there was in the ’80s with Volckerific 20% rates. There is certainly a force for capital flight from gold to paper – although the result remains minor, at present.

    (b): (a) starts to happen, and capital controls on PMs are imposed. Honestly, I don’t think a government that doesn’t have the stones to nationalize a bankrupt bank has the stones to do this. But I could be wrong.

    The result: a terrifying descent into panic, chaos and mayhem, as people realize their monetary system is sustained not by faith, but by force. The military steps in, suspends the Constitution and restores order.

    The outcome: order, which is good. Another desirable outcome.

    (c): by some miracle, Geithner and company get the debt engine started again. Possibly by filling the entire chassis of the vehicle with monetary gasoline, and lighting it with a giant rock of pure fiscal crack. Burning with a horrible green flame, the FIRE economy begins to roll again, lending, spending and building with new, demonic speed. The debt-to-GDP ratio, which had become dangerously stagnant and even threatened to decline, resumes its healthy and vigorous quest for infinity.

    In other words: recovery. So this is a good outcome, as well.

    The only trouble, however, is that there is no investment which does well in all three good outcomes…

  35. babar writes:

    > You have it right – there are no bank runs in a liquidity-solvent banking system, for precisely this reason. Since the last ones out do not get screwed, no one has any reason to run for the exits.

    Why? If you have a higher level of credit loss than you have provisioned for, you can still be undercapitalized.

    I get your idea of a bank as a big “matching engine” and it makes some sense. (Lots of NP-complete problems!) But I don’t think instruments with contingent claims will exist if banks are not allowed to treat claims as contingent. If they need to be managed as non-contingent claims they will have to be priced accordingly.

    For the record, I do have a problem with non-recourse-ness of the loans but it is shades of gray to me. It’s a subsidy to the financial sector. It socializes losses. I have a problem with these things. But I consider socialization of losses a political inevitability. I’d like to see the losses minimized. If Geithner’s scheme works, this will be true. But I have strong doubts about whether it will work.

    In my mind the big “moral hazard” issues are not at the “organization” scale but at the individual scale. Why have there been no serious proposals for compensation reform? Compensation incentives caused the crisis IMHO. Why have there been so few prosecutions for loan fraud, even when the FDIC has seized assets and presumably records of failed banks?

  36. mencius writes:


    Why? If you have a higher level of credit loss than you have provisioned for, you can still be undercapitalized.

    Yes, of course. The confusion is my fault. I meant “bank run” in the sense of a systemic, self-reinforcing collapse, caused by the redemption of contingent instruments.

    A bank run in the sense you mean is just a decline in the value of the instruments issued by an institution whose financial condition is not as expected. This is no different from a decline in the stocks and bonds of IBM, if IBM reports bad results. If these instruments are negotiable securities – all instruments should be negotiable securities – their market price will decline. Possibly to zero, if IBM has screwed up badly enough.

    But this is an endogenous problem, ie, one coming out of planning mistakes in the real productive economy. This form of pain will always exist, and someone will always have to eat it. What we don’t need is pain caused by the financial structures themselves.

    Compensation incentives caused the crisis IMHO.

    Many things caused the crisis IMHO, including this thing.

    However, I would say the main cause was the collapse of a time pyramid in which $100 trillion of present-value financial assets were built on less than $1 trillion of actual present dollars. What should be surprising is not that it collapsed, but that it stayed up so long, and that such large components of it remain standing albeit as ruins.

    Logically, this structure could only stand with government support, ie, the presumed existence of trillions of dollars of Federal loan guarantees, which were presumed to be standing even behind the “shadow banking system” due to the “too-big-to-fail” delusion.

    USG probably should have bit the bullet, printed all those dollars and saved the shadow banking system. But in retrospect, I see little chance that it could have. USG is much bigger than anyone thinks, but much weaker, too. Wall Street was leaning on an extremely broken reed.

    It is easy to think that continuous debt expansion is normal, healthy and sane, because it has been going on for our entire lives. Both logic and history, however, tell us that this is an incorrect assumption.

    I get your idea of a bank as a big “matching engine” and it makes some sense. (Lots of NP-complete problems!)

    Very funny, babar. And true. We obviously have had some of the same training.

    It would certainly take quite a bit of coding to adapt high-powered, high-geared computational finance to a liquidity-solvent financial system. But our present system of computational finance, or perhaps I should say our legacy system, needs to be rewritten, because it was constructed on the basis of assumptions that turned out not to be true.

    What we are witnessing is not the death of free-market finance, because free-market finance is the truth and the truth cannot be killed. You can throw out Nature with a pitchfork, but she always crawls back in.

    What we are witnessing is the death of “free-market finance,” ie, the cargo-cult zombie version of the real thing. Free-market finance: Mises. “Free-market finance”: Friedman.

    Perhaps an appropriate comparison is to Lenin’s NEP. It was simply delusional to think that this activity could continue indefinitely under a system of government now rooted firmly in the principle of state socialism, in which the American people have finally come to believe.

    Most importantly, turning “free-market finance” into free-market finance is not at all a straightforward operation. Old software does not become young again. I fear that real free-market finance will have to grow out of some tiny, hidden sprout, rather than a patch to today’s broken OS.

    Which means that NP-complete problems aren’t really the issue. Before financial engineering comes basic finance: lending and borrowing, stocks and bonds. 1.0 before 4.0. At present, we don’t even have an 0.1.

  37. Menicius,

    Toward an 0.1, see

    Fits the description of a disruptive innovation, as per Clayton Christensen’s framework. A so-called “new-market disruption,” to be precise. (See footnote [10] for background on CC’s framework.)

    Thoughts? Questions?


  38. Menicius,

    I’m reading part 1 of your intro to UR. Nice overlap of subject matter with


  39. Zombie banks it is, gentlemen, as I read Bernanke’s policy statement du jour. Doesn’t he know (could our monetary and fiscal policy any more spastic?)…

    It’s the debt, stupid.

  40. RueTheDay writes:

    mencius said:

    “What we are witnessing is not the death of free-market finance, because free-market finance is the truth and the truth cannot be killed. You can throw out Nature with a pitchfork, but she always crawls back in.

    What we are witnessing is the death of “free-market finance,” ie, the cargo-cult zombie version of the real thing. Free-market finance: Mises. “Free-market finance”: Friedman. ”

    What in the hell is free-market finance, minus the scare quotes? We have always had boom/busts and financial panics as long as we’ve had a financial system, for the reasons Minsky noted – debt financing of production + uncertain future = endogenous instability. This occurs under a gold standard or fiat money, central bank or no central bank, government intervention or no government intervention. It is historical fact.

    And why did you bring “Nature” into this? As Bentham noted over 200 years ago, all talk of natural laws and natural rights amounts to “nonsense upon stilts”. There are no universal laws governing human behavior in all contexts, and natural rights are a social contruct created by people (even Mises admitted this much).

  41. JKH writes:


    Felix Rohatyn was interviewed this morning on CNBC. One of his central points was that it’s a fundamental mistake to classify government outlays on infrastructure as expenditure rather than investment. Maybe you’re related?

    The government needs a balance sheet to force transparency.

    Better yet, a balance sheet that consolidates its central bank.

  42. mencius writes:

    A balance sheet! JKH, we can only dream.


    You are believing the libertarian version of history and assuming that something like free-market finance existed in the past. It did not. It sort of did, but only by default. The free market isn’t something that just happens. It needs to be designed and invented. The past is at best an inspiration.

    In particular, not even the classical gold standard of the 19th century was a true hard-money standard. Again, the Bank of England has been watering the gold supply with paper, using FRB/MT, since 1694. Not to mention the various American follies. Even in the English-speaking world, America has been notorious for paper money since the 17th century.

    The last hard-money bank (metals warehouse, basically) was the Bank of Amsterdam in the Dutch golden age. I think there was also a hard silver bank in Hamburg. Only small, isolated cases exist in the English-speaking world.

    But two of these are described by Condy Raguet, here: Gibraltar and Havana, in the early 19th century. Raguet’s description of Gibraltar is especially provocative:

    Such being the theory of this branch of my subject, I have the satisfaction to state in regard to the practice under it, upon the testimony of a respectable American merchant, who resided and carried on extensive operations for near twenty years at Gibraltar, where there has never been any but a metallic currency, that he never knew during that whole period, such a thing as a general pressure for money. He has known individuals fail from incautious speculations, or indiscreet advances, or expensive living; but he never saw a time that money was not readily obtainable, at the ordinary rate of interest, by any merchant in good credit. He assured me, that no such thing as a general rise or fall in the prices of commodities, or property was known there; and that so satisfied were the inhabitants of the advantages they enjoyed from a metallic currency, although attended by the inconvenience of keeping in iron chests, and of counting large sums in Spanish dollars and doubloons, that several attempts to establish a bank there were put down by almost common consent…

  43. mencius writes:

    Note especially Raguet’s terminology: the market for lending and borrowing in Gibraltar is healthy and stable, at least if you believe both him and his respectable merchant – but there is no such thing as a “bank.” Ie, an entity that backs claims to current money with future loans.

    This usage seems to still be followed by most bankers today. Obviously it is an obstacle to my modest proposal. I will maintain, however, not as a matter of finance but one of English, that any institution that borrows and lends is entitled to the humble name of “bank.”

  44. winterspeak writes:

    JKH: “Felix Rohatyn was interviewed this morning on CNBC. One of his central points was that it’s a fundamental mistake to classify government outlays on infrastructure as expenditure rather than investment. Maybe you’re related?”

    What does it matter? Government spending is Government spending, and is fundamentally different from private sector (or microeconomic) spending.

    The Government must run persistent deficits in order for the private sector to net save. This is true as a matter of accounting.

    That spending might be good or bad, but you certainly will not get the answer to that question by classifying it under one entry on a balance sheet vs another.

    If the Government outlays were reclassified as “investment” or as “blueberry pancakes” it would make no difference as to whether they were wise. The first question is: is the Government funding the private sector’s demand to save as quickly as that demand is increasing, or is the private sector having to save out of aggregate demand?

    The second question is: are the Government expenditures which exist, wise? (Either for reasons of economics, or charity).

    A balance sheet treatment does not answer either of these. Even worse, a standard, private sector balance sheet treatment would be aghast at the huge, negative entry that is “equity and retained earnings” and trigger actively harmful activity, like cutting the deficit.

    A standard balance sheet treatment also generates unhelpful ideas like “intergenerational transfer of wealth” or “the public debt will need to be paid down someday” or “who will finance the deficit”. etc. etc.

    Accounting, for a currency issuer, whose purpose is “public good” (not profit maximization) is profoundly different from accounting as we know it. I am a friend to accounting, but a Government balance sheet would, and should, look profoundly different from a private balance sheet, and honestly, is less important that the cash flow statement.

  45. JKH writes:


    It matters in this case because the comment was addressed to Steve, relating to a specific post he had done earlier.

  46. JKH writes:


    Why is it important that the private sector “net save”?

    The private sector saves, even when “net saving” = zero.

    What is so special about the condition of “net saving” being greater than zero?

  47. winterspeak writes:

    JKH: Sorry, missed the reference to SRW.

    As for net saving:

    An individual private sector entity (household or firm) can deficit spend to the extent that someone will lend the money. Each liability must be matched by an asset. Therefore, the sum of all private sector balance sheets must net out to zero, as a matter of accounting. Therefore the private sector as a whole cannot run a net deficit (or a surplus) unless some other sector is will to lend (or borrow) to them.

    It is a fact that, usually, the private sector wants to net save. That is, overall, individual private sector entities want to spend less than they earn, and bank the difference as savings. The only way the private sector can do this, as a matter of accounting, is if another sector, the public sector, runs a deficit. Moreover, sovereign currency issuers are not constrained in the same way as private entities as to the level of deficit they can run.

    It is a reasonable goal for a private entity to have a large and positive equity entry. Therefore, any consolidated public sector balance sheet has to have a negative entry as equity. And probably a very large one. A consolidated public sector balance sheet should look very different from a private sector balance sheet, and may even look unhealthy if read as if it were a private sector balance sheet.

  48. JKH writes:


    I disagree.

    In a closed (e.g. global) economy, saving equals investment.

    The level of both global saving and global investment is roughly 23 per cent of global GDP.

    Public deficits constitute negative saving by the public sector, by definition.

    Thus, the global private sector saves 23 per cent of global GDP, PLUS its financing of the global government budget deficit.

    The PLUS factor is your definition of net private sector saving.

    In fact, this PLUS factor has nothing to do with global investment or global saving, since government dissaving offsets net private sector saving according to your definition.

    So the question remains, what is so special about the PLUS factor, which is a net wash against government dissaving?

  49. winterspeak writes:

    JKH: The PLUS factor is important, because when it changes, it can drive down aggregate demand and create a recession that cannot be handled by lower interest rates if public dissaving does not change as well.

    In my definition, an increase in net private sector savings, which is not matched by an increase in public deficit (through higher G, or better yet, lower T) reduces AD, increases unemployment etc. Yes, this higher private savings rate does mean higher investment (as per the identity) but that identity counts inventory as investment. We see stockpiling if inventory at every major dock today, and yes that’s investment as per the identity, but not the sort of investment that anyone thinks is positive.

    It does not have to be this way. The PLUS must be met by higher Government deficits, but this can be at many levels of unemployment, and AD, and the deficit can be reached quickly by active fiscal policy, or slowly (with higher unemployment) by passive fiscal policy (which we are seeing now. Tax revenues fall in recession. Government spending automatically ticks up.)

    So, from a policy perspective, understanding that actively managing the deficit to help the private sector save, and that private sector savings is stimulative, is important because it leads to an economy with lower unemployment and higher AD. We’ve just seen a “stimulus” that only gives HHs $400 and will be a policy disaster.

    From a conceptual perspective, it’s important because otherwise you start wanting the public sector to run a zero deficit, or even worse, a surplus, which misunderstands the point of public deficits entirely. You also see all this nonsense about “the deficit is too big” and “we’re passing on a huge burden to our children” etc. etc. which just are not true.

    S=I, but “I” can be containers of extra cars in Long Beach or loans to promising businesses. private savings = public deficit, but that can be at 15% unemployment, with the deficit driven by low tax revenues and high unemployment payouts, or it can be at 5% unemployment with the deficit drive by a payroll tax holiday. The differences are material.

  50. JKH writes:


    That was quite a good exposition. I agree with much of it, but it still doesn’t answer my question.

    I have no problem with the idea that the private sector has a greater demand for net private savings in an environment of rapidly imploding aggregate demand and deflation such as the one we’re in. That to me is another way of saying that there is good reason for Keynesian type fiscal stimulus and running larger government deficits in such an environment. It’s easy to connect those two ideas as two sides of the same coin.

    But that’s not my question. My impression from what you’ve written on a number of occasions is that you believe the private sector has a demand for net private savings in a structural sense. You’ve described here the rationale for changes in net private savings at the second derivative level, which I agree with. But my question is why do you think (G – t) must be greater than 0 as an ongoing structural phenomenon – not why you think (G – t) must grow in size in an environment like this.

    Second, you’ve noted the fact that compositional issues are important in the context of both I = S and (G – t) = NPS. Of course, compositional issues are important. They’re important whether (G – t) is greater than zero, zero, or less than zero. But the fact that compositional issues are important in any context doesn’t address the question of why it is important that (G – t) be greater than 0 in the sense that the private sector has a demand for net private savings as I’m assuming you claim it does at a structural level.

    And, by the way, I’m sympathetic to the view that (G – t) should structurally be greater than zero. I just haven’t heard a convincing explanation of why this should be the case.

  51. mencius writes:

    I think the trouble here is the bizarre Keynesian definition of “saving.” If there is even any such definition. A reading of Hazlitt’s Failure of the New Economics might prove informative, and certainly entertaining.

    The distinction between the US’s “private” and “public” sector strikes me as arbitrary and unimportant. Ownership means control; many decisions taken in the “private” sector are in fact “regulatory,” ie, public. This is especially true, obviously, for the banking system – which I believe should be consolidated onto USG’s balance sheet.

    (If there is a qualitative economic difference between USG and private actors, it is that USG can print dollars and no one else can. Of course, IBM can issue IBM shares and no one else can, which is why I prefer to view Federal Reserve Notes as equity rather than debt – since they are freely issued and carry no obligation.)

    Of course, we can (at least logically) draw an arbitrary line between any two sets of economic entities, consolidate the balance sheet of each set, and observe the flow between them. For example, we can consolidate the balance sheets of all the households whose head has a last name beginning with A-M, the same for N-Z, and observe the monetary flow between them, just as we can observe the monetary flow between the US and China.

    If we think of Winterspeak’s statement in these terms, he defines “saving” as lending by private citizens to USG. This equals borrowing by USG from private citizens, and (G – t) must indeed exceed 0. Thus, for the private sector to “save,” the public must borrow. But (as I’m sure WS would agree) this is an accounting identity, not an observation.

    Rothbard, referring to the “equation of exchange” (MV = PQ) had perhaps the best analogy for the Keynesian-Fisherian fondness for such identities. “The amount of rain that falls from the sky,” he said, “is the same as the amount of water that hits the ground.” Also true. But a weather forecast it is not.

    I find most uses of the word “saving” meretricious. For me, saving is a state, not an action. Everyone who stores dollars (or, as Keynes would say, “hoards” them) is saving dollars. Thus the quantity of dollar savings in the world is the number of dollars in the world. Again, for this number to increase, USG must create new dollars – our accounting identity.

    But why must this number increase? This goes to the heart of our chronic inflation dependency. By increasing the number of dollars, USG is certainly not increasing the market capitalization of the dollar – in whatever unit you choose to denominate that. For example, printing more dollars is not an effective way to increase the total value of all dollars in gold. Or in potatoes, wheat, houses, euros, etc, etc.

    Therefore, when (G – t) exceeds 0 and USG runs a deficit (most simply, by printing money to make the accounts balance), USG is simply diluting each dollar held by each dollar-holder. If USG prints dollars and spends them, it is practicing “forced spending” – it is effectively ransacking the vaults of those who have stored dollars, in order to create artificial demand and make the accounts of what would otherwise appear an unprofitable economic system balance. Indeed our GDP is extremely dependent on these injections of ersatz demand. Oops.

    And of course USG is doing the same when it prints its free loan guarantees, which account for most of the price of most financial assets today. (Eg, without the FDIC your bank deposits would be worth less than par – quite a bit less, in fact.)

    Note that this usage – anyone who has dollars is “storing” them – makes the distinction between “savings” and “investment” make slightly more sense. You can do three things with a dollar: you can spend it (exchange it for goods and services), store it (just keep it), or lend it (exchange it for a promise of future money). The distinction between spending and lending is slightly fuzzy. The distinction between storing and lending is not. And every dollar in the world, at every instant of time, is being stored by someone.

    When USG finances its deficit not by issuing present dollars, but by issuing future dollars which it exchanges for present dollars (ie, by selling Treasury bonds), the situation does not change in its essentials. It is simply sterilizing, ie, moving present dollar purchasing power into the future. As demand for long future dollars dwindles, USG’s deficit is financed by increasingly shorter obligations. The limit of this is plain and simple money-printing. Present money is a special case of future money.

    Transitioning to an economic system which does not continuously dilute its currency is a task whose difficulty cannot be overstated. It really is quite comparable to the conversion of communism into capitalism. For example, the trivial algorithm for a transition is for the Fed to break its printing press and repudiate all informal loan guarantees. A dollar is then either (a) the piece of green paper, or (b) an entry in a Fed account (high-powered money).

    It is a simple exercise to imagine the devastation that this algorithm (deflation to M0) would wreak on America. Landing a plane is not the same thing as crashing it. Again, the problem is hard, hard, hard.

  52. JKH writes:


    Yes, I’ve been assuming the definition of saving as per Keynes.

    Also, I believe Keynes defines saving as a “flow” (i.e. saving from income) and savings as a “stock” (cumulative saving). The former is action; the latter state.

    Discussions about particular points do become more challenging when the definition of things is allowed to change. Nothing wrong with changing the definition of things, but zeroing in on what a person means exactly by something can become more difficult, I think. So my question here depends more on the maintenance of definitions than definitions being Keynesian.

    In this case, I might acknowledge that net private saving should normally be positive to allow for the fact that monetary base growth is normally positive and the monetary base normally funds government expenditure. But this falls apart if the rest of the budget is in surplus to the same degree.

    What I’m really getting at is that there is a slightly obscured implication in the assumption that “net private saving” must be structurally positive, which is that government under this set of definitions can’t or shouldn’t run a surplus from time to time. That’s certainly wrong as a fact. But still my question is should governments aim toward structural deficit as the norm,and I probably mean structural deficit beyond the funding of the monetary base, although I’m not sure. Depending on this, the premise of the question may conflict with a view that the monetary base should not be expanding in the first place.

  53. winterspeak writes:

    JKH: You are correct, I believe that the private sector has a structural demand for NPS, but this isn’t a matter of accounting or identity. I will posit a few reasons here, you tell me what you think of them.

    1. I think the most important initial reason is that the public sector is a currency issuer, and therefore has to generate demand for its currency in the private sector. It does this by levying taxes, and then it needs to spend to inject the private sector with the currency it needs to pay those taxes. The usual coordination story then spreads the currency throughout the entire sector. This is a good reason for why the initial condition has to be public sector in deficit, private sector in surplus, but it does not explain why that has to structurally continue.

    2. The public sector is capable of maintaining deficits in a way that the private sector cannot as it is a currency issuer and has access to monetary and fiscal tools that the private sector cannot. To the extent that one sector wants to be in deficit and the other in surplus, it is natural that the deficit falls on the one that can carry it most easily.

    3. In a fiat world, the public sector carries inflation risk, and the private sector default risk. It is easier to bear inflation risk, particularly if you are a currency issuer, than default risk if you are a currency user. Therefore, the deficit lies in the public sector, surplus in private sector.

    4. Currency issuers have no use for “savings” or “surplus”. Why hoard money for future use if you can just create as much as you need? Currency users have a clear use to hoard money for future use, so they do. This is the common sense explanation.

    5. Empirics. Government is usually in deficit. Private sector is usually in surplus. Why is the sky blue?

    I am not 100% satisfied with any of these explanations, although I believe as an observation I’m on pretty safe ground assigning a desire to save to the private sector, and a desire to run deficits to the public sector. I’m interested in your thoughts here.

    MENCIUS: Thank you for your long and thoughtful response. Always good to have goldbug/austrians in the mix, they will challenge your assumptions if nothing else!

    I don’t think a public/private distinction is unimportant, but I do agree that many nominally private entities have been revealed as extensions of USG. The ability to print currency is a *big* difference — I would not discard it so easily.

    I don’t think that currency=equity is a good analogy. Currency is much less intrinsically valuable than equity, and you do equity a disservice with that analogy.

    I do not define “savings” as private entities lending to USG. USG has no need to private actors to “lend” to it, as it can create its own money. “Savings” is the residual left after the private sector spends, invests, and pays taxes. It must be mathematically equivalent to the Federal deficit. (G-T). I’m happy to use the word “hoard” as people on this board object to the positive associations with the word “save”.

    As for “saved dollars” being the only “real dollars”, in some ways they are, in some ways they are not. As far as prices are concerned, saved dollars needn’t exist at all! Having no impact on the real world is not a typical test *for* existence!

    I’m sympathetic to the merits of a world of mild deflation. However, the realities of nominal debt loads means that, practically, de-leveraging is far more unpleasant that leveraging (although excess leverage is not great either). The goal of USG is not to maintain, increase, or maximize the value of its currency. Currency is not that valuable to USG.

    I would, instead, approach the problem as trying to find a mechanism for private sector actors to defer consumption/save/hoard. We agree that fiat currency is a lousy vehicle for that for the dilution reasons you suggest.

  54. mencius writes:


    Au contraire – equity has no intrinsic value. If I sell 100,000 shares in my daughter’s nonexistent lemonade stand, what I have is 100,000 pieces of paper. Nonetheless, all these pieces of paper obey the rules of equity: each share has the same value (zero),

    FRNs are special in a couple of ways. While FRNs have never paid a dividend and never will, USG creates a vast amount of demand for FRNs in three ways which are very foreign to conventional accounting.

    One, it standardizes them as a medium of exchange. Two, it demands them in exchange for taxes. Three, one and two combine to make the FRN the winner of the Nash equilibrium game for the standardization of savings.

    And, of course, USG is sovereign. Otherwise, its scrip is just scrip. There is no magic about it. If a government is a sovereign corporation, sovereign accounting is corporate accounting. Of course, most corporations do not denominate their own accounts in their own shares! But one could imagine a world in which they did, terrifying and awful though it would be.

    As far as prices are concerned, I would argue, stored dollars most certainly exist: they set the demand schedule for all the markets in which one can exchange dollars for something else. (More precisely, this demand curve is a function of stored dollars plus dollar substitutes, ie, savings in the normal, English-language sense of the word.)

    Eg: if no one has any stored dollars, the dollar price of everything is zero. If anyone can print dollars, the dollar price of anything is infinity. A currency lives on the curve between these endpoints, and price schedules in that currency are entirely a function of its stock and distribution.

    As for lending: when I buy a promise of future money, or goods, or sexual favors, or anything else, issued by X, I feel it is appropriate to state that I am lending to X. This is true whether the purchase is directly from X, or there is some middleman involved, or I buy the bond on the open market.

    Thus, when I buy a USG-issued T-bill, I am lending to USG. In my vocabulary. You see how tricky these words can be.

    USG can create its own money, but it does not want to create its own money – or rather, it prefers to create future rather than present money. Among other things, this teleports the “inflationary” consequences of its fiscal indiscretion from the present to the future. Although maturity transformation can teleport them right back.

    I agree that USG, unlike any normal corporation, is not terribly motivated to maintain the purchasing power of its own “shares.” This is one of the many unsolvable financial and management problems with USG – an entity I feel is in need of comprehensive restructuring.

  55. mencius writes:


    You might enjoy reading Hazlitt. He goes through the General Theory quite literally line by line. To make a long story short, Keynes (and note that Keynes is not always on the same page with the Keynesians) is very slippery with his terminology.

    Besides, Hazlitt was an NYT reporter (amazingly, considering his views), and his elucidation is quite, um, lucid. The Failure (which he wrote with Rothbard’s assistance) is really an unrecognized classic of the economics literature. Not even the Misesians pay enough attention to it, IMHO.

    As far as the use of “savings” today, I see people use it indiscriminately to mean (a) hoarding currency; (b) lending currency; or (c) both. This ambiguity is sufficient to disqualify it from the English language for my purposes, regardless of what Keynes did or did not mean.

  56. mencius writes:

    Above, read “each share has the same value, zero, and the value of all the shares together is the value of the nonexistent lemonade stand – zero.”

  57. JKH writes:


    Quick responses to your points:

    1. Steady growth in the monetary base is the closest thing to a structural rationale that I can think of. An interesting exception is the case where the central bank intermediates the base to private sector financial assets, such as is occurring with the Fed now. More generally, a case might be made that the base should be backed by some kind of investment rather than used for non-investment expenditure. But in either of these latter cases, interestingly, the result is not necessarily current net private saving. This is because it is a pure balance sheet transaction, a swap of assets. E.g. private sector sells an existing asset to the CB in exchange for base money. This results in no dissaving for the government and therefore no net saving for the private sector.

    2. This pretty much boils down to 1. I’m not sure the “one or the other” argument holds.

    3. That’s an interesting one. It amounts to using the monetary base and perhaps more than that as a risk transformation function – which is essentially what is happening now with various Fed and Treasury programs. It’s a financial intermediation function rather than an expenditure one. Again, however, the effect of this type of financial intermediation does not necessarily show up as current net private saving.

    4. Having no use for surplus is not quite the same as wanting a deficit.

    5. Yes. Government is usually in deficit. I don’t know that translates directly to a true demand for net private saving, although the result is an identity.

    Colour me open. I need to think more about my own question, although I’m in no rush to prove it one way or the other. The risk transformation aspect is important, but it is primarily a balance sheet transaction rather than an expenditure and deficit one. And it ramps up in Keynesian stimulus times. Maybe there’s a corresponding function on the expenditure side, but I’m not sure this demands that there naturally be a deficit to fund it.

  58. JKH writes:


    Thanks. I’ve bookmarked Hazlitt.

  59. winterspeak writes:

    MENCIUS: Perhaps I have the same reaction to “equity share” that others have to “savings”. There is a sense of intrinsic worth that simply does not exist for FRN. “Point” or “token” gives you better intuitions for what currency really is than “equity” (you can try to define associations away, but why not pick something more accurate to begin with?)

    USG creates money whenever it spends. It uncreates money whenever it taxes. The money it creates and uncreates is present money.

    Think carefully about your assertion that buying a FRN is lending the USG money. Why would USG want its own scrip? Also, if USG printed $1B and gave it to me, and I put it under my mattress (hoarded it), it would not show up in any CPI number. If a tree falls in the forest etc. etc. China is not funding the US deficit, the US is funding the Chinese demand for savings. You are not lending the US money. The US is funding your demand for savings.

    When USG runs a surplus it is not “hoarding” money for future use, it is draining the private sector of savings. Eventually, that will cause demand destruction if the private sector decides to try and replenish its savings through cutting back on I and C.

    I think the issue is that, deep down, you believe that fiat money has some intrinsic store of value. You certainly believe that money *should* have some intrinsic value. I will not speak to that normative judgement call (although I am sympathetic to it), but in a positive sense, fiat money does not have any intrinsic value. The confusion is totally understandable, our financial system is littered with gold standard-esque arabesques, including fractional reserve banking, the idea that the government needs to tax in order to spend, that somehow you can have intergenerational transfer of wealth, the Government can run out of money etc. etc. This may be the crises that gets us to fully embrace fiat money, a change as dramatic as the move off the gold standard in the 30s.

    I don’t mean to assign motivations to you. It’s just that you dismiss MV=PQ, you think that dollars stashed under a mattress have some impact on dollars being actively circulated, and those things just aren’t true. If you think of dollars as points, it’s obvious why. If you think of dollars as having some, albeit slender, intrinsic value then you won’t agree.

    Embrace fiat and join me on the dark side. You will have Truth on your side. You can also keep no-MT.

    JKH: All good points. I still think there is a structural reason, but I cannot identify it, so maybe I’m wrong. Certainly we’re seeing a very strong demand for increased hoarding today, and I can see, in an inflationary environment, that demand decrease (and maybe turn negative).

  60. RueTheDay writes:

    Mencius – Again, what in your view constitutes true free market finance? A hard currency standard where all transactions are literally conducted with the physical transfer of gold and silver bullion?

  61. mencius writes:


    Simple: a free-market financial system is one with “separation of bank and state.” Ie, the state does not lend money or guarantee private loans. Its interaction with the financial system is limited to contract enforcement.

    (Obviously, this precludes the “lender of last resort” function, which precludes FRB/MT. For a free-market interest rate, private lending must equal private borrowing at every maturity. The market itself will enforce this regulation, although at least at first it can’t hurt to have a watchdog as well.)

    Note that this definition is orthogonal to the choice of currency. This could be gold, USG shares or other scrip, IBM shares, etc, etc. From the economic perspective, a fiat currency is exactly the same as a gold currency, if the latter is issued by a government who owns an infinite gold mine.

    If the first, you would expect a 21st century currency to consist of electronic allocations of gold. Google “allocated gold” or “GoldMoney” if this seems innovative – it is not.

  62. mencius writes:


    I don’t have a problem with fiat currency at all, actually. And I am happy to call FRNs “scrip” rather than “shares” if you want. The basic principle still applies: every dollar entitles the holder to the exact same benefits, whatever these may be.

    In the case of FRNs, the benefits USG provides to the holder are not dividends, but the right to defray taxes with FRNs. Different – but still quite valuable, if you consider the consequence of nonpayment.

    Imagine if Starbucks ran its entire accounting system in Starbucks stock. To buy an espresso, you would need to hand over a share (or whatever) of stock. This would go into the treasury and be destroyed. Starbucks could pay its contractors and employees in shares, of course – I’ve seen it done. And shareholders would not receive dividends, but they could exchange their shares for lattes (or sell them to people who had dollars and wanted lattes).

    This would be retarded, of course. From an accounting perspective, however, it could be made to function. And it is not at all dissimilar to what USG does with its “dollars.”

    No – a holder of an FRN is not lending to USG. He is holding USG’s scrip. However, if he exchanges present FRN for USG’s future FRN, ie buys a T-bond, he is most certainly lending to USG.

    Similarly, you could imagine Starbucks issuing “Starbucks bonds,” priced in present Starbucks shares, payable in future Starbucks shares. So you could exchange ten present Starbucks shares for eleven 2010 Starbucks shares. Same difference.

    I think the main difference between our perspectives is that I think even with a fiat currency, there is no legitimate business reason for the state to debase the currency – ie, issue more of it. A fiat currency should actually be harder than a gold standard, because the quantity of it can be permanently fixed.

    Why is there no rational reason for dilution? Because dilution amounts to nothing but redistribution. Anything you can do by diluting a fiat currency, you can do by redistributing it (and rewriting contracts in it). Neither of these are normal activities under a healthy, stable government.

    There are exceptions to this rule. USG at present is running a financial system with artificially high level of debt relative to the number of actual dollars. In order to wind down this system without massive liquidation, it is necessary to print dollars to buy the debt (or accomplish the equivalent through redistribution).

    But, like JKH, I fail to see the reason why G should chronically exceed t. For me, the difference between targeted, controlled dilution and chronic dilution is the difference between surgical anesthesia and morphine addiction. Surely you can’t deny that USG displays many symptoms of the latter.

  63. winterspeak writes:

    MENCIUS: “But, like JKH, I fail to see the reason why G should chronically exceed t”.

    Yup, and the key word there is “chronic”. I don’t know the answer. Does the private sector, chronically, need to increase its level of saving? There is a hard bound to it dissaving, but that does not get us to a full answer.

    If the private sector, as it grows, demands a higher and higher level of savings, then yes, G must chronically exceed t. But that’s a big “if”.

    — Other points

    If a entity buys a USG t-bill, they are lending to the US, but it’s better thought of as the US funding their demand for t-bills. China is not funding the US deficit, and they hold lots of t-bills. The USG is funding China’s demand for savings. You want to get the intuitions write, and holders of USG liabilities are best thought of as savers being funded by USG.

    Defraying taxes is totally valuable, but this value is determined by the tax rate. The “value” of the FRNs is seeded by how much in taxes you need to cough up. It is not intrinsic to the FRN. Your Starbuck example, if you also gave it a powerful army, is exactly what happens with USG.

    Government purpose is not “not debase” the currency — it’s to keep the economy fully employed in the short term, and productive in the long term. It really does not care what the currency is worth, remember, it has no value to it as USG can make as much as it wants. But it certainly needs to maintain a demand to save it, and high inflation reduces that demand, making deficits even more inflationary.

    The private sector wants to save a well managed currency, because they perceive it as a store of value. The more the private sector wants to save, the higher deficits the Government can (must) run. Governments like to spend. They don’t like to tax. This is a win win, and does not show up in CPI.

    I am *not* a redistributionist btw, and am sympathetic to your points about letting savers save.

  64. john c. halasz writes:

    As to why there might be a “ratchet” effect to government spending, private firms are actually loathe to commit to high-cost, long-run, uncertain fixed capital investment. If they enjoy a legacy of such investments, they tend to sit on them and manage the rents that accrue to them. If such rents deteriorate, such firms have severe difficulties with reducing their “operational leverage”, (as currently with GM), though other investments must take the place of such declining firms. But oligopoly, due to economies of/increasing returns to scale, and the high levels of productivity they induce in the economy, (such that oligopoly rents are not simply dysfunctional, but partly earned and serve to help manage long-run investments through a variety of business cycle conditions), is the predominant reality of developed industrial capitalist economies. Which fact gives the lie to pure “free market” competitive accounts of such economies, as well as to marginalist accounts of price-formation, (since decreasing, not increasing, “marginal” costs tend to dominate most of the oligopoly cost-curve and the management of rents entails that many prices are manipulated or administered, such that their “markets” don’t constantly or readily clear). Further, profits from successful productive investments need to be recycled into further investment opportunities, else an over-accumulation of profits will lead to a contraction of investment-spending levels and a corresponding contraction of aggregate demand, and a decline and disaggregation of the current networked or interlinked valuation of extant capital stocks. Hence government spending will serve, if of the right composition and policy mix, not just as a management of aggregate demand levels, but as a supplement and encouragement to real long-run productive capital investments, as well as, a facilitator of transitions and restructurings/inter-sectoral shifts from declining to emergent industries.

    The upshot then is not only that government fiscal deficits will and should vary counter-cyclically with the level of demand and productive activity in the overall economy, but that surpluses accumulated during boom times need not entirely counter-balance deficits during busts, provided, given the uncertainties involved in any future prospects, the rate of growth in public debt remains prudently less than the reasonably expectable prospective growth in future output/GDP of the economy as a whole.

  65. winterspeak writes:

    john c. halasz: Federal surpluses will certainly bring boom times to an end! Although I cannot get my head around why the Fed can, or would “accumulate” surpluses. It’s like an airline “accumulating” it’s own frequent flier miles. Curious.

  66. mencius writes:


    Yup, and the key word there is “chronic”. I don’t know the answer. Does the private sector, chronically, need to increase its level of saving? If the private sector, as it grows, demands a higher and higher level of savings, then yes, G must chronically exceed t.

    I disagree completely. Again the word “saving” is confusing you, I think. In my phrasing, the question is: “does the private sector, chronically, need to increase the number of dollars it stores?”

    When put in this form, I think the answer is clear: “no.”

    The question is exactly equivalent to asking: does IBM, chronically, need to increase the number of shares outstanding? Clearly no such change can affect the market capitalization of IBM, at least as denominated in anything but IBM shares. Similarly, increasing the number of dollars outstanding does not increase the aggregate purchasing power, in any currency or commodity, of all dollars outstanding.

    Intuitively, think of a currency (fiat or natural) as a sort of battery for purchasing power. Increasing the amount of energy in the battery involves charging the battery, not buying another empty battery. Similarly, savers may increase the purchasing power of a fixed-quantity currency, or decrease it, by deciding to be more thrifty or more frugal.

    Increasing thrift does not require dilution of the currency – it just pumps more monetary energy into the battery. Similarly, if citizens decide to become more prodigal, contraction of the currency is not required.

  67. RueTheDay writes:

    Mencius said: “Simple: a free-market financial system is one with “separation of bank and state.” Ie, the state does not lend money or guarantee private loans. Its interaction with the financial system is limited to contract enforcement.”

    It’s difficult to address this sort of statement without lanching into a “what is the proper role of the state?” digression, but I will try. One commonly agreed upon function of the state is to regulate activities that lead to or may lead to significant negative externalities. We do not have an unregulated, free-market system of nuclear power generation where anyone can build a nuclear reactor in their backyard and sell electricity to other parties because such a system would inevitably lead to radiation leaks and possible meltdowns that would negatively affect many people apart from the plant operator and his customers. The history of financial markets demonstrates them to have similar characteristics – when there is a financial crisis, it quickly spreads to other parties both within and outside of the financial sector; there are broad systemic impacts. In my experience, those who advocate for unregulated financial markets have placed the historical cart before the horse – we do not have financial crises because the government regulates the markets, we have regulated financial markets because past experience has shown them to have more severe and frequent crises in the absence of regulation.

    Mensius said: “(Obviously, this precludes the “lender of last resort” function, which precludes FRB/MT. For a free-market interest rate, private lending must equal private borrowing at every maturity. The market itself will enforce this regulation, although at least at first it can’t hurt to have a watchdog as well.)”

    This seems to smack of the libertarian view that we don’t need to have food safety regulations because the market will regulate itself – when restaurants serve tainted food and people die from it, no one will visit those restaurants anymore and thus they will go out of business leaving only restaurants that serve safe food. This view ignores the problem of asymmetric information (it also ignores the fact that such an approach is cold comfort for those that ate the food and died).

    Mencius said: “Note that this definition is orthogonal to the choice of currency. This could be gold, USG shares or other scrip, IBM shares, etc, etc. From the economic perspective, a fiat currency is exactly the same as a gold currency, if the latter is issued by a government who owns an infinite gold mine.”

    And what prevents it from being deposit account balances at fractional reserve banks? I can certainly understand a posteriori criticisms of the RESULTS of FRB, I cannot understand what, a priori, prevents such a system from arising under a free market.

  68. winterspeak writes:

    Mencius: Sure, you can rephrase it “does the private sector need to chronically increase the number of dollars it stores”, but I don’t think it’s obvious that the answer is “no”.

    Your IBM analogy is false — it’s more accurate to ask “does IBM need to chronically increase the number of dollars it stores”. IBM would certainly like to have some cash on hand, as would a typical household. If IBM, or a household, takes on debt, then they would naturally like to pay that debt down (to the degree, at least, it is not useful for shielding taxes). As IBM grows wealthier, it may like to have an even bigger cash hoard (at least in absolute terms). I think we both agree that it makes no sense for us to talk about IBM “hoarding” shares. Why bother hoarding something you can create, at will, out of nothing?

    Look at your own household. If you have a mortgage, don’t you want to pay it down, at least a little? Don’t you want to have more cash in the bank? Wouldn’t you like to grow your savings, so you could pass that on to family once you die? Or just have money as an insurance policy in case something bad happens?

    As people grow wealthier, it is an empiric fact that they save more (or, if you prefer, spend less of their income). As a society grows wealthier, it isn’t crazy to assume that it, as a whole, will save more too. (or, if you prefer, keep more dollars in a bank). This isn’t true as a matter of accounting, but it may very well be a matter of human nature, and human behavior matters!

    So, if decreasing marginal propensity to consume with wealth is human nature, then yes, as a society grows richer, it needs to run chronic deficits to fund the demand for more money in the bank. I don’t know if this is true, but it’s certainly plausible.

    You remain wedded to your conception of money as a store of value. You say you don’t, but I cannot interpret your battery analogy any other way. As a fact, you are wrong — fiat money is no such thing. It is also a fact that the monopoly issuer of said specie is not interested in maximizing its value, it has other fish to fry. I’m not disagreeing with you in a normative sense, but in a positive sense. Your mental model is giving you bad intuitions about how the world works, and decreasing your ability to predict actions and consequences. We should think about money how it is, as well as how it should be.

    When IBM issues more shares, existing shareholders actively feel the dilutive effect in a real sense. Earnings per share go down. Dividend yield goes down. Voting rights (snigger) go down. Share price should go down too. etc. This is true no matter whether those shares are stored in a brokerage account, or traded.

    When USG issues more dollars, you may see no effect at all if the dollars are stored. Everything costs the same. You could do as much with your hoard as you could before. I’m not saying that there hasn’t been an effect, but it’s clearly a very different kind of effect.

    One final thought. As debt is nominally denominated, real debt burdens grow larger in a deflationary environment, and therefore default risk is pro-cyclical. This means that any private sector (closed network of currency users) with debt has a risk of bank-run style collapses. A marginal loan encounters deflation, and tips it into default. A non-marginal loan now becomes a marginal loan, and defaults too, etc… This would be true in MT and no-MT.

    Bank liabilities are no place for market discipline. The same may be true for private sector debt.

  69. winterspeak writes:

    RTD: You really should ignore FRB. FRB has no impact on banks ability to lend. Banks make the loans they want, and then borrow the reserves they need to hit their requirements, either from other banks, or directly from the Fed. It’s just this weird old mechanism left over from gold standard days that is now used to set the Federal Funds rate.

    Capital requirements, and quality of borrowers are the real constraint on lending, not reserve requirements. FRB is a red herring.

  70. mencius writes:


    It’s difficult to address this sort of statement without lanching into a “what is the proper role of the state?” digression, but I will try. One commonly agreed upon function of the state is to regulate activities that lead to or may lead to significant negative externalities. We do not have an unregulated, free-market system of nuclear power generation where anyone can build a nuclear reactor in their backyard and sell electricity to other parties because such a system would inevitably lead to radiation leaks and possible meltdowns that would negatively affect many people apart from the plant operator and his customers.

    What you’re not quite getting, I think, is that the government has created the problem it is needed to solve. Not at all an isolated case of this phenomenon.

    Of course an FRB/MT system needs to be regulated. It needs to be so regulated that it is effectively a branch of the state. In fact, it is probably best regarded as a covert arm of government, meant to disguise the obvious: vote-buying through cheap loans. Unregulated, it is simply a means for privatizing gains and socializing losses, as of course we see.

    This is an almost universal pathology of democracy. Basically, when you net out all the accounting, FRB/MT is a way for the government to lend its voters money. If you read the Bancroft essay I linked to, you’ll see that this has been going on in the US since before there was a US.

    But all this aside: my entire point is that a free-market financial system, as described above, is intrinsically stable. As in other free markets, prices change only as individual supply and demand schedules change – meaning that your interest rates are as stable as your society.

    Therefore, it is not a nuclear reactor. It is a toaster-oven. It does not involve any such externalities. The iatrogenic disease of periodic banking crises is gone. And the regulators can find new employment in the lawn-care industry – if it will hire them.

  71. mencius writes:

    Sure, you can rephrase it “does the private sector need to chronically increase the number of dollars it stores”, but I don’t think it’s obvious that the answer is “no”.

    Sure it is. Let me expand on why. The proof does not require any empirical observation about mortgages, etc. It is a matter of deductive logic.

    Let’s simplify the dollar system and get rid of these pieces of green paper. Let’s also get rid of the banks. Flatten the system. Everyone has an account at the Fed. The number of dollars in that account is the number of dollars you have. You pay your bills with your cell phone, etc.

    Let’s also forget that there are any other financial assets in the world – instead, we will just have dollars. Perhaps USG nationalizes all traded assets at their current market price, replacing them with dollars. And we could have both present and future dollars (ie, current money and bonds), but let’s not – only present dollars for the moment.

    And let’s, in the same big Fed server, have a single database in which all contracts that involve payment in dollars are registered.

    Now: it is obvious that if we add a zero to both everyone’s dollar account, and everyone’s dollar obligations – a dime becomes a dollar, a ten becomes a hundred – we have changed nothing at all in the real state of the world. That is: aside from a little confusion, peoples’ behavior the day after the change will be the same as before – as it would not be if you performed some other change, eg, randomizing everyone’s balance.

    And the same is true (except for remainder pennies) if we divide by 10. Or by 7, or by 0.7, or by any other factor. We can express this clearly by performing our dollar accounting not in dollars, but in fractions of all dollars outstanding. What matters is the pattern in which the dollars are distributed across accounts – not the denominator.

    What does this tell us? It tells us that for any state of the world in which the Fed’s computer contains X dollars, there is an economically equivalent state in which it contains Y dollars. Thus, for any quantity of dollars Q in the Fed’s computer, any real-world state outside the Fed’s computer is possible.

    And thus, for the state of the world to change, there is no need for the quantity of dollars to change, which directly contradicts your argument.

    Now: how does this play out in the intuitive sense in which you are using the word “saving?” Simple – as I increase my hoard of stored dollars, those dollars are not coming fresh and hot off the Fed’s printing press. I receive them in exchange for goods and services, from someone else who had them before, but doesn’t now. In your terms, my “saving” is precisely balanced by his “dissaving.” Net “saving,” across the economy: zero.

    Suppose everyone wants to increase their hoard of stored dollars, as you postulate? But remember the fractional accounting. It is not mathematically possible for everyone’s balance to increase relative to everyone else’s balance. If everyone wants more money, some must gain and some must lose. Everyone does indeed want more money, and this is indeed how it works. Once again, logic matches reality.

    Basically, varying the quantity of money adds a free variable to a system that doesn’t need the free variable. All it contributes is instability. A fixed pool of money is like a string pulled tight; a variable poll is like one that is loose. There is only one way to be tight, but many ways to be loose.

    You can also see a monetary authority as a counterfeiter, of course. The intuitive feeling that counterfeiting is in some sense parasitic and criminal is one of great antiquity, and not to be cast aside lightly. A benevolent monetary authority is a benevolent counterfeiter. A contradiction in terms? I think so.

    Even if the counterfeiter does not merely spend his fresh bills, but instead lends them out. Nay – even if the counterfeiter does not even lend, but just uses his printing press to back free options that he gives away to his friends. Presumably in exchange for delicious, back-scratching favors.

    You remain wedded to your conception of money as a store of value. You say you don’t, but I cannot interpret your battery analogy any other way. As a fact, you are wrong — fiat money is no such thing.

    The battery analogy is an illustration, not an exact description. I actually detest the word “value” whenever I see it – it is usually hiding some kind of confusion. I don’t know exactly what you mean by “store of value,” and nor do I know exactly what I would mean if I said “store of value.” So I try not to. But I may slip up. If so, you are right to call me out.

    I will maintain, however, that there is one and only one motivation for holding money, whether fiat or metallic: the intent to make further purchases with it. (Okay, in a gold standard, it is possible that you want to melt down your Krugerrands and make a necklace. But this is an unusual case, even in a gold standard.)

    If you feel there is some other motivation, please say so! Otherwise, I will assume we agree, “store of value” or not.

  72. mencius writes:

    This seems to smack of the libertarian view that we don’t need to have food safety regulations because the market will regulate itself – when restaurants serve tainted food and people die from it, no one will visit those restaurants anymore and thus they will go out of business leaving only restaurants that serve safe food.

    Consider this, RTD: USG allows you to visit Morocco. USG does not regulate or inspect the restaurants in Morocco. USG does not require travelers to Morocco to bring their own PowerBars in sealed packages. How is this not, by your standards, a tragic lapse of USG’s tender paternal care?

    Of course, the US is not Morocco. It is impossible for me or you to know what would happen in the US of 2009 if its cities and states gave up on restaurant inspection. I can think of three possibilities, however.

    One, the number of food safety accidents would remain more or less the same as it is now, or even decrease. In which case we are correct in sending our regulators to work for Miracle-Gro.

    Two, the number of food safety accidents would increase, and customers would become concerned. This would create an excellent business opportunity for private inspection services, who could put their sticker in the restaurant’s window – just as with eco-certification today.

    Three, the number of food safety accidents would increase, and customers wouldn’t give a crap. In which case they probably deserve a good sharp case of salmonella.

    Are there any other cases I am missing here? If not, which of the outcomes do you disagree with?

  73. RueTheDay writes:

    Mencius – Of the three options you laid out, number 2 would be closest to reality, except the ultimate outcome would be very different. In fact, we already have a perfect example of “private inspection services” today, they’re called Ratings Agencies, and judging by how all of those AAA rated securities are doing right now, it’s safe to say that the concept is an abject failure.

  74. mencius writes:


    There is nothing even slightly private about Moody’s, S&P and Fitch. What do you think the “N” stands for in “NRSRO?” “Notorious?”

    You inadvertently illustrate the adaptive motivation for our Potemkin-capitalist financial system: plausible deniability. Here is what the NRSROs are in real life: regulatory agencies. And what happened? They failed. Just like the SEC. Just like every other alphabet-soup agency out there. These are the people you want to give more power to?

    You could merge the three big NRSROs into a Department of Ratings, and you wouldn’t have to change a damn thing about what the people there do all day. The big three were not providing personal opinions about the creditworthiness of bonds, as (say) a real ratings agency like Egan-Jones does. They were acting on the theory of scientific lending, which is inextricably embedded in our financial regulatory system today. Their bogus AAA ratings were produced by a formal mathematical process, as beloved by bureaucrats everywhere. Moreover, like bureaucrats everywhere, they had no skin in the game.

    I am not a libertarian. Like Communists everywhere, I support nationalization. Better honest Brezhnevism than Potemkin capitalism. At least, once it’s clear that Wall Street is no more than America’s Gosplan, we can start to think about how to privatize it properly.

    Any new free-market financial system has to be built from scratch. Our old one, the one we inherited from the 19th century, is as dead as the Holy Roman Empire. Attempts to restore it to life are misguided at best, sinister at worst. It was hardly perfect to begin with, anyway.

  75. john c. halasz writes:


    I was responding to your basic question as to why the government might tend to “chronically” run deficits, that is why government spending minus taxes might tend to be a recurrently positive number.

    Two preliminaries: 1) my answer is a a general, “in principle” one, and doesn’t directly address the current debacle, which may be beyond remedy, or, at least, we don’t know fully how severe the damage will be, except that the responses with far have been variously lagging, inadequate and inappropriate. 2) One can gain any clear view on the matter by focusing solely on the financial “economy” and the issues of money and credit to e exclusion of all else, since the financial “economy” is largely a superstructure that functions to intermediate the real productive economy, and the “values” of financial assets will tend endemically to deviate from and lag the value that is being generated in the real productive economy or not, such that the former is anything but a transparent reflection of the latter, at most a fun-house mirror, though, to be sure, the dysfunctions of the former can disrupt the operations of the former.

    That the sovereign government issues currency so that taxes might be paid is thus far nothing but a tautology, as if the government would be paying its own taxes. Obviously, the currency must circulate in mediating the activities and exchanges of he real productive economy, which the profits and wages realized thereby then partly serve to pay the taxes. The provision of a uniform and stable currency is one of the basic “services” the sovereign government is there to provide. (And, though this is a trivial point, the Fed does, in fact, “normally” run a surplus, due to interest from the bonds on its balance sheet, which it periodically remits to the Treasury, though nowadays it’s implicitly running a loss of tens, if not hundreds, of billions of $. And the Fed normally “prints” money simply by buying bonds from the public stock of debt to put on its balance sheet, thereby adding to base-money, though under deflation it might as well monetize the gov. debt by directly buying bonds newly issued by the Treasury. There is some limit, however, as to how far a central bank could go in monetizing debt before it bankrupts itself, both through running against the counter-productive effects of the inflation it would produce and through exceeding the ability of the Treasury to recapitalize it through any further issuance of debt, which Willem Buiter, if I understand him correctly, estimates at 40% of GDP).

    So my suggestion was, if you want to find a realistic reason why government spending “chronically” exceeds taxation, and thus government debt grows, in absolute, if not relative, terms, you won’t find it by focusing on a sheerly monetary definition of “savings”, but you need to look to the production/business cycle in the real economy, both medium-term and long-wave, (i.e. the cycle of such cycles). And, in my view, the production/business cycle is rooted in two basic asymmetries and lags: 1) productivity-enhancing. cost-reducing real capital investments that result in increases in the real distributable surplus-product, i.e. increased real output and increased real wealth, in any meaningful sense of “real wealth”, which spark a boom and yield high profits, result in a need to re-invest such profits, while opportunities for profitable reinvestments gradually start to yield diminishing returns and dry up. (Note that the need to reinvest high profits will spark an inflation in the “value” of financial “assets”, both based on rising expectations from prior realized profits and because realized profits will tend to get diverted into bidding up such “assets”, which will have the effect of raising costs for real capital investment, while lowering returns, which would occur regardless of any CB manipulation of interest rates, whether accurately judged and timed or not,- though I tend to take technocratic fantasies of perfect control with considerable grains of salt). 2) Profits from successful productivity-enhancing real capital investment will only partly and with a considerable lag be re-distributed to wage-based demand, whether through lower real prices or higher real employment and wages, such that eventually high real output will run up against deficient real effective demand, even as returns on investments have been diminishing. The result is a bust, in which both investment spending by firms and consumption spending by households will diminish and dry up, and both will attempt to increase “savings” by paying down debt, even as prior asset prices decline, further diminishing savings: the paradox of thrift. Government deficit spending will then automatically increase, both due to lowered tax receipts and due to increased spending on automatic stabilizers and relief measures, and, if the government attempts to decrease the deficit by raising taxes or cutting spending, the result would only be to exacerbate the downturn and increase the rate of contraction in a downwardly spiraling cycle. Hence public deficit spending/debt accumulation tends to rise automatically in downturns, in response to increased private “savings”,- (and the drop in financial assets prices and increases in private default rates tends to make gov. bonds relatively attractive vehicles for “risk-free” savings/investment, especially as the risks of private investments rise and opportunities diminish), just as public deficits and debt tends to diminish as private spending increases during booms. Further, additional increases in public spending and debt can serve to palliate the blows and lessen the extent of the downturn , while facilitating recovery, through recycling idled private resources, providing further new opportunities for private investment, and through redistributing income to prop up wage-based demand.

    The foregoing might serve to explain why government spending tends to “chronically” outrun taxes and hence public debt tends to grow over time, but you won’t get to such a realistic account by focusing solely on a monetary definition of “savings”, as if it were a matter of hoarding $ now to spend them later. Money standardly functions as a medium of exchange, unit of account, and store of “value”, but there is nothing to say that the temporary purchasing-power of money must endure sempiternally as a hoard withdraw from circulation: there is a reason why it is called “currency”. In fact, the “genius” of industrial capitalism largely consists in the way it constantly recycles surpluses to regenerate and increase surpluses, which contrasts with the hoarding of surpluses under feudalism. (I’m amused that enthusiasts for “free markets” often don’t seem to appreciate the key role of bankruptcy in promoting market “efficiency”, as it facilitates the recovery and recirculation of the remaining “value” of resources). And such a dynamic system is constantly redistributive, but there is nothing to guarantee that any temporary distribution is “natural”, optimal, or “right”. (Indeed, insofar as prior market outcomes, as a result of some distribution of probabilities, are always more a result of “luck” than inequal skill or effort, locking in such prior results into permanent market advantage is more likely to produce less effective investment and especially “investment” geared toward maintaining advantage rather than increasing aggregate real output, than periodically reshuffling the distributive deck to allow for new initiative by different agents, insofar as as new investment opportunities involve a “discovery” process, and require inter-sectoral shifts in labor and skills). Theoretically, the function of financial capital markets is to optimally allocate investment spending such that the rate-of-profit is equalized across all businesses and sectors as at optimal “equilibrium”, and, though there may be operative tendencies in that direction, due to changing productive ratios underlying the real productive economy, even leaving aside error rates and information asymmetries, such an “equilibrium” is
    never actually attained, while, equally, the effects of an investment boom will tend over time to shift financial “asset” prices away from any optimal “equilibrium”. The results of such cyclical readjustment processes are likely to show up as being counter-balanced by increases in public debt. But the overall upshot here is that there is no form of monetary “savings” that is not obversely a form of spending, by some agent or agency or another, whether “I” or “C” or “G”. One might save through buying a long-run durable good, such as a house, the equity in which might be recovered by later re-sale, or by buying a share in the legal claims to a set of capital goods, from which future revenues accrue, or by buying a debt-obligation such as a bond, or simply by putting one’s money in a bank. But there is nothing to guarantee that the future monetary “value” of one’s savings will equal or exceed its current “value”, though the least likely case is simply withdrawing a hoard of currency from circulation.

    There is nothing necessarily wrong with mild price deflation, as opposed to debt deflation, which increases the real burden of debts. But arguably, a regime of mild price inflation is preferable, as it encourages investment through the “scent” of perceived demand, through attenuating the risks of assuming debt, and through discouraging any hoarding of money. (One could simply “invest” in collectibles, such as stamps, which will increase in price a) with the general price level and b) with the general real expansion of the economy, and hence the available demand for such collectibles, an excellent “tax-free” way for small fry to save for retirement). Severe inflation would, of course, result in forms of hoarding and distortions in investment allocations. But the main point is that money, at bottom, is just a symbolic cypher, such that its “value” is not “absolutely” determined by some quantity, nor by any cost-of-production. There is nothing wrong with increasing the stock of base money to accommodate increases in real output, and arguably to fail to do so would have deflationary and contractionary effects. However, one should take account of Post-Keynesian accounts of the credit/finance system, such as Steve Keen’s, that expanding the monetary base does not cause and precede the expansion of credit, but rather the expansion of credit, endogenous to the production cycle, precedes and expands the monetary base. The equality of savings with investment is a short-run NIPA accounting identity, but the question as to whether savings produce investment or vice versa is chicken-and-egg. Insofar as productivity-enhancing, cost-reducing real capital investments increase the real distributable surplus-product, then they not only produce savings, qua cost reductions, but they increase the supply of loanable funds through the recycling of profits, even if some prior supply of savings is required to supply the basis of the credit system. But any system of lending and therefore of credit would involve the very same embedded leverage, as is involved in FRB, in whatever different guise, and would be “justified” by the realization of successful investment in the same way and to the same degree, even as it would absorb the losses of failed investments, just the same. Which is to say, that given the dangers and excesses involved, FRB or any other operative system of credit and finance requires careful and strong regulation. Which is why claims that our current woes are due to government regulation and intervention in the financial system, rather than the failure, evasion, and captive complicity of government regulation, are thoroughly bogus. (The proposed, or rather fantasied, alternatives to a regulated system are even more defective, if not entirely inoperable). But, even granted that the government, in issuing currency, doesn’t directly have to concern itself with its “value”, I disagree with you that it has no cause to do so. For one thing there is a dilemma between the desire to inflate away the accumulation of public debt and the increased interest and social costs of doing so. Insofar as the “private” sector has recourse to public debt to increase its “savings”, inflating away those “savings” and raising borrowing costs, nominally and possibly really, might be counter-productive, since it is the lowered prices of financial “assets” that reduce investment costs and encourage recovery. But, more importantly, the for-ex value of the currency is of importance to the sovereign government, or, at least, to those dominant “interests” that have captured and control the sovereign power. A high US$ might be favored by Wall St. and the MNCs, insofar as it advantages their terms-of-trade, and allows for cheaper FDI and for the buying up of foreign banks, companies and resources, to increase the scope of the rents that they draw off of the RoW. That it also increases the trade deficit, lowers domestic wages and increases household debt might be just added fringe benefits for such wealthy elites. Conversely, a lower for-ex value might be desirable to encourage export-led growth. (Though Sweden’s bank nationalization is often touted as a precedent, it’s little mentioned that the key to both economic and financial recovery was a 30% devaluation of the Krona, which kicked off exports. But, of course, nowadays, no country can count on devaluing for export growth, given the global collapse in trade). But it’s simply not the case, that, just because currency doesn’t have any “absolute” value, there are no policy interests conditioning its relative “value”.

    Austrian “economics” would apply, if at all, only to a steady-state, constant-reproduction economy. What is truly odd about it’s insistence on a “hard money” absolute standard of “value”, which supposedly could only be assured by a completely “free” market system, is not just how it misses the effects of technical improvements in real capital stocks and the radical alteration in (re-)production ratios that involves, which is the main source of the success and growth that capitalist “markets” provide, but rather in how it relies on an essentially feudal notion of hoarding, preventing the recirculation of surpluses, which is the very “secret” of capitalist markets. But what could one expect from a bunch of Austrian aristocrats and wannabes, nostalgic for the lost glories of their dying empire and the hierarchical “liberty” it allowed, and imagining it would all return to them in the utopian future, if only markets would just be “free” enough. Dear Mencius has just provided us with a restatement of “Say’s Law”, together with an exposition of the “neutrality of money” thesis, which even “New Keynesians” still adhere to in that ever elusive long-run. As if a monetized industrial production economy, in which increases in productivity and output underwrite increases in profits, could be reduced, if not to a primitive barter economy, then to commodity exchange in a petty producers’ economy, the artisan economy that accompanied the feudal agrarian economy, in which there is only a C-M-C circuit and no M-C-M’ circuit, i.e. in which increases in profits generate no asymmetries in exchange, by virtue of the fact that they do not, because can not, occur. Wonderful! Wunderbar!

  76. BSG writes:

    The assertion has been made (I think it was WS, perhaps RTD) that deflation leads to inevitable busts as more and more people become unable to pay their debts.

    This is usually used as an argument for retaining our current, inflation-based, financial system.

    It is misleading. Perhaps it is even a good example of using truth to deceive (or unintentionally mislead, if you prefer to soften it.)

    If we had a permanently fixed amount of currency units (no counterfeiting, fractional reserve banking or maturity transformation allowed,) it will be only the availability of goods and services that will determine prices, for all intents and purposes. During periods of prosperity, we’ll have deflation, typically mild, occasionally sharp as with major technological breakthroughs or major natural resource finds. During difficult times (e.g. large scale natural disaster, drought, war, social strife) we’ll have inflation. During some periods we’ll have steady prices.

    Those that lend and/or borrow will know that they must take account of this in setting interest rates and repayment terms. The only difference from the current system is that they won’t also have to take into account the capriciousness (sometimes disguised by fancy models) of bankers (central or otherwise) who can arbitrarily change the amount of currency units apparently available. Without the latter variable, while mistakes will still happen, they will be less likely and adjustments and improvements will be almost sure and quick.

    By contrast, the current system ensures the type of bust its defenders warn us of if we dare dump it. To illustrate this, consider what would happen if the Fed went all the way with the current bailout approach by making everyone whole with newly created money.

    Of course, we’ll have hyperinflation for a time, but once everyone was in the clear, that should end as there would be no need to create more money for bailouts. Right?

    Wrong. It doesn’t take much creativity to imagine that everyone with half a brain, or much less, will rush to borrow as much as too-big-to-fail financial institutions allow them too, secure in the knowledge that they will be bailed out if necessary. The lenders will feel no need to restrain themselves because they too will be bailed out. Before you know it, we’re back to hyperinflation and soon thereafter a complete currency collapse, which means the Fed no longer has the means to bail out anyone. Game over.

    What’s that you say? The Fed would never allow that. Once it was done with the initial bailout, the Fed would restrain lending, so no problem. Really?

    If the Fed was ready, willing and able to restrain lending, they would have done that after any one of the many busts we’ve experienced so we wouldn’t be facing the current crisis to begin with. Fercrissake, it hasn’t even been 20 years since the S &L crisis, not to mention the Japan implosion. Face it, the Fed exists to encourage lending and borrowing, or, more precisely, to enable private banks to maximize their unearned rents. The more the merrier, supposedly.

    Of course, they may try to slow things down, but that would only delay the next bust. Even centuries ago there were those (e.g. Thomas Jefferson) who knew that a system of engineered inflation inevitably leads to overindebtedness, which in turn leads to the very bust that brings about the cascade of bankruptcies that apologists for the current system warn us about. Maybe they should heed their own warnings.

    Considering the general prosperity we’ve been able to achieve (even accounting for the busted phony wealth) despite the corruption and inefficiency deeply embedded in our current financial system it’s that much more depressing to think about the much higher level of prosperity we’d likely attain as a society without being so weighed down.

    Racers of all types strive to lighten their load and reduce resistance as much as possible so they can do better. Here’s hoping that some day enough people to make a difference will figure out that we can overcome all of the propaganda and create a better future.

    Mencius makes a good observation about what may be democracy’s greatest failing – politicians, with legions of allies, use our money to bribe us into allowing them to take ever more from us in a seemingly never ending vicious cycle. Except it does end. Usually very badly. See, e.g., the French Revolution, the Weimar Republic, etc. and so forth. Beyond crying in the wilderness, as some of us here are wont to do, we can only hope that we can do better. It’s not much, but if anyone has some ideas – gold and guns offer no panacea – please share.

  77. JKH writes:

    It looks to me like the core issue is the intended full bailout of bank creditors, more so than the facilitating partial bailout by a thousand dilutions of bank equity holders. The repeated government capitalization of Citi seems designed to avoid the first outcome at all costs, including whatever residual value remains to the original equity holders as an unfortunate by-product.

    The nationalization debate is poisoned by at least implicit disagreement on the definition of nationalization. Nationalization is as much a virus of a given intensity and depth, starting with FDIC maturity transformation insurance, as an ultimate event per se.

    I’m still not sure I see a robust argument why (G – t) should be secularly positive, except for central bank monetization of natural currency demand. I can see the argument for a countercyclical tendency during times of bust. But I don’t see an argument against the opposite countercyclical tendency during boom times. Clinton did manage a surplus for several years, and Canada did it for a decade. I think the proof of (G – t) being optimally secularly positive, apart from demand determined CB currency monetization, could be linked to some sort of asymmetric Kondratieff correlation, but I don’t think I’ve seen such.


    “Basically, varying the quantity of money adds a free variable to a system that doesn’t need the free variable.”

    That strikes me as being at the heart of the truth, if it is true.

  78. RueTheDay writes:

    BSG said: “If we had a permanently fixed amount of currency units (no counterfeiting, fractional reserve banking or maturity transformation allowed,) it will be only the availability of goods and services that will determine prices, for all intents and purposes.”

    And if my aunt had balls, she’d be my uncle.

    Whether or not you realize it, what you are wishing for is a barter economy, dressed up as a “monetary economy” where money is perfectly neutral and the quantity of money is fixed. Unfortunately, a modern economy that relies on large scale industrial production has almost nothing in common with a barter economy, the quantity of money is largely endogenous rather than exogenous, and money is nowhere near neutral.

  79. BSG writes:

    RTD – If use of sound money is tantamount to barter, then maybe your aunt does have, uh, well, you know. Actually, even if she does, it isn’t. That is true practically by definition (look up “money” in a dictionary if you’re wondering what I mean.)

    The “modern economy” you apparently refer to is clearly intertwined with arbitrary manipulation of currency (whether fiat or faux gold standard) and that is its primary weakness, as said manipulation distorts price signals and inevitably leads to imbalances and overindebtedness that, as I said, cause the sort of bust our modern economy is currently in.

    You also state the obvious when you suggest that in an economy in which any quantity of the medium of exchange can be – and is – created out of thin air at will, the “quantity of money is largely endogenous.” That’s the problem!

    Obviously, every one will make up his/her own mind, especially here. I am offering a clearly unorthodox alternative analysis along with the suggestion that there has been much propaganda embedded in the conventional wisdom.

    We each have a filter through which we view the world. The more aware we are of that filter and the manner in which it distorts reality, the better able we will be to improve our lives (see, e.g., flat earth.)

  80. mencius writes:

    Austrian “economics” would apply, if at all, only to a steady-state, constant-reproduction economy.

    This claim is unequivocally false. Production of goods and services does not require production of currency. And this includes capital goods.

    Have you ever worked, for example, at a firm which produced goods and services? Did you notice that it was also producing currency? Have you ever worked at a firm whose stock price went up? Did it have an intaglio press in the back room?

    What is truly odd about it’s insistence on a “hard money” absolute standard of “value”, which supposedly could only be assured by a completely “free” market system, is not just how it misses the effects of technical improvements in real capital stocks and the radical alteration in (re-)production ratios that involves, which is the main source of the success and growth that capitalist “markets” provide, but rather in how it relies on an essentially feudal notion of hoarding, preventing the recirculation of surpluses, which is the very “secret” of capitalist markets. But what could one expect from a bunch of Austrian aristocrats and wannabes, nostalgic for the lost glories of their dying empire and the hierarchical “liberty” it allowed, and imagining it would all return to them in the utopian future, if only markets would just be “free” enough.

    Sez the man who believes that counterfeiting isn’t counterfeiting, if the government does it.

    John, are you curious at all about the history of the ideas that you yourself hold? You might enjoy looking into the biographies of Richard T. Ely, Irving Fisher and John Maynard Keynes.

    I believe you’ll find that the last was an aristocrat so dissolute that his likes can be found neither in the Austro-Hungarian nor even the Holy Roman Empire. You may have to go back to the household of Nero. As for the first two, I believe you’ll find that they were religious maniacs, so Christian that they make Pat Robertson look like Marilyn Manson.

    So you might have better luck engaging with the substantive arguments…

  81. winterspeak writes:

    MENCIUS: Thank you for your model. I think it’s excellent, and reduces the issue to its core. Let’s run with it.

    Say, in your system, we have four accounts. Let’s call them Sovereign, Investor, Entrepreneur, and Bystander. All begin at zero.

    — Scenario 1

    To fix money supply, let’s say that only one account, Sovereign, is allowed to carry an overdraft. It injects an initial amount of money into some combination of Investor, Entrepreneur, and Bystander, crediting their accounts and debiting its own. Let’s call this amount G-T, and let’s not care about how it’s distributed across the other three account. Our Sovereign, watchmaker like, now sits back and participates no further. Its account carries an overdraft of G-T.

    Sovereign, Entrepreneur, and Bystander carry, between then, a combined G-T on their asset side, and G-T on the liability side (as equity). Any of these three can shrink their balance sheets, so another can grow its balance sheets, but the balance sheet of the system as a whole cannot change. There would essentially be no debt financing, because even though one entity might choose to finance another via debt, since accounts cannot go into overdraft they would have to maintain at least a zero balance, so in essence it would be an equity investment or straight cash transfer.

    I find it difficult to imagine how this money system would support economic growth. It could bear deflation, as there would be no nominal debt contracts. Entrepreneur makes an equity investment in Investor, which is successful. This requires Bystanders balance sheet to shrink, and successful investments usually require buyers. Clearly this system is dramatically different from what we have now, and maybe it does not work because I lack imagination. Won’t be the first time.

    — Scenario 2

    Let’s keep the watchmaker Sovereign, as before, so Investor, Entrepreneur, and Bystander begin with balance sheets of G-T across them (cash of G-T on the asset side balanced by equity of G-T on the liability side), and the Sovereign carries an overdraft of G-T but acts no further.

    Let’s let *everyone* be allowed to carry an overdraft now too.

    Investor no longer needs to write down cash to make an investment in Entrepreneur. Investor can issue Entrepreneur a loan by crediting Entrepreneurs with the cash, and balancing that with a receivable. Entrepreneur accepts the loan as a credit, and balances that with an amount due as a liability. Entrepreneur and Investor now both have larger balance sheets, which they can use to buy things with. Bystander feels he is being diluted, just as he would if Sovereign gave another G-T slug to Entrepreneur and Investor. The point is that if you allow non-Sovereign accounts to run overdrafts, then you get dilution just as you would if the Sovereign prints money and issues it to a favored entity. Allowing debt enables dilution, no matter what the Sovereign does. If you’re measuring “CPI” you’ll even see an uptick.

    Of course, in this system, if any entity takes on more debt they can service, balance sheets will shrink, potentially in a bank-run, domino type effect until everyone who carried an overdraft repudiates their debts back to zero (and have no balance sheet, as their equity has gone to zero) and those who carried positive balances see those positive balances reduced down to whatever initial cash balance they had. The non-Sovereign sector is always in danger of collapsing back to a balance sheet size of G-T. This is true whether there is MT or no, but MT makes things much worse.

    It’s easy to see how this system support growth. Entrepreneur comes up with some new idea, which increases their income. This income enables them to service larger debt, so they can maintain a larger balance sheet. Entrepreneur wins, investor wins, and although Bystander is being “diluted” they’re probably better of as well because of whatever Entrepreneur came up with. You can think of “ideas” as reasons for the non-Sovereign entities to trade with each other. More ideas enable more trading, and thus higher measured GDP. But in order for this additional trading to enable higher consumption, you have to let the non-Sovereign sector, as a whole, increase its balance sheet, which means some non-Sovereign entities need to carry a negative balance (overdraft).

    A final observation: In the above, we now have a reason for the Sovereign to continue injecting money into the non-Sovereign sector — it enables the non-Sovereign sector to increase the size of its balance sheet while not taking on additional leverage. The higher the amount of leverage, the greater risk there is of the non-Sovereign sector to collapse its combined balance sheet down to the initial G-T. By leveraging up, the Sovereign enables the non-Sovereign to delever. If a larger balance sheet is necessary for economic growth (which it might be), then a chronic increase in G-T is valuable in making that growth less prone to collapse.

    JKH: Time to define “chronic”. There are certainly times when a Sovereign would want to run a surplus and destroy money in the private sector. During inflationary periods, for example. But if larger balance sheets are requirements for economic growth (big if) then, as an economy grows (with stable CPI) you would want to see larger deficits overall to limit the potentially destabalizing leverage underneath the whole system.

  82. mencius writes:


    Excellent! We are moving the ball down the field. You’ve accepted my model and I’ll accept yours –

    with one caveat, which is that I dislike the term “overdraft.” An overdraft is a financial facility engineered by your bank to simulate something that doesn’t actually exist: negative money.

    If I were building an accounting system from scratch, I would build it entirely with unsigned integers. The first goal of accounting is to speak the truth, and it must be built firmly on reality – in which all numbers are natural.

    An overdraft is of course a loan – of positive money – from your bank to you. Thus it is slightly odd to talk about a sovereign currency issuer running an “overdraft,” because loans to oneself cancel on the balance sheet. Yet the sovereign is not incurring any obligations to anyone else, either.

    This is why the definition of dollars as equity, shares or scrip, not obligations, makes much more sense to me. Economically, a fiat monetary system works exactly the same as a metallic monetary system in which the sovereign owns an infinite gold mine. (And indeed, the Fed’s printer has done much the same thing for the US that the treasures of Potosi did for Spain.) But when one mines gold, one does not incur an obligation to Pluto to put it back. Money is not debt.

    That said, let’s consider your scenarios.

    In scenario 1, your intuition is exactly correct: under a fixed money supply, the phenomenon you call “economic growth” is not to be expected. The question is: is this a good thing, or a bad thing?

    What is economic growth, anyway? An increase in a number called “GDP.” And what is “GDP?” There are a few ways to answer the question, but I like to think of the entity we call “the economy” as the consolidated balance sheets of all businesses. “GDP” is thus their consolidated net sales (ie, final sales).

    It’s easy to see why this number will be stable (ie, changing only in response to actual social trends) in a closed-loop monetary system. All statistics of monetary flow are stable in a closed-loop monetary system. Even capital prices (stock-market indexes) should be stable, thanks to the EMH.

    For example, we have no reason to expect that progress in science and technology should increase the ratio of annual business sales to the amount of money consumers have to spend. The goods sold will improve, presumably. Perhaps there is a slight downward trend in total sales with quality – goods become so good and so durable that everyone’s needs are permanently satisfied, and there are no sales at all. By definition, this is a desirable rather than an undesirable endpoint.

    At present, in our open-loop monetary system, business sales increase over time, for a simple and obvious reason: credit expansion creates money. 20th-century economists developed the rather absurd statistical tic of separating these regular increases into “real” and “nominal” growth. Eg, if the goods sold increase in quality at the same price, this growth is “real.”

    In a closed-loop monetary system, this artifact is not available as a method of measuring human progress. If human progress must indeed be measured, however, the wonks at the BLS are already in the business of quantifying the quality of goods. They may simply continue to do so, producing an index of Annual Betterness ™. “Comrade Andropov, this year’s Lada is 3% better than last year’s Lada.” If people want to hear it, someone can presumably be paid to say it.

    Moving to your scenario 2: this is the familiar argument for the lending counterfeiter. It is easily refuted by the practice of normalized accounting, that is, measuring the balance of each account at the Fed as a fraction of a total which sums to 1.

    Under normalized accounting, there is no difference between creating dollars and draining them proportionally from everyone’s account. Similarly, there is no difference between destroying dollars and proportionally filling everyone’s account.

    Our lending counterfeiter thus becomes a lending thief. He raids your bank account for money which he lends to his friends, promising to put it all back later.

    Which might not be *so* bad, except that he doesn’t put it all back later. He runs a permanent overdraft, to use your terminology, said the overdraft is not only permanent but permanently expanding. Thus, even though he is constantly restoring the money he stole, he is constantly stealing more.

    Our counterfeiter’s friends may, of course, be good and deserving people. But this is simply a regime of involuntary lending. The idea that this can somehow make a society more prosperous is the result of computing the accounting in a way that shows the output, but neglects the input.

  83. winterspeak writes:

    MENCIUS: You are correct — it is strange to think about a sovereign “overdraft” which is why I don’t like the phrase “Government debt”. The Government can always pay any debt it carries. “Deficit” is the most neutral way I can think of the simple G-T entity, and I took pains to stress that it would be carried in perpetuity.

    Your analogy of an infinite goldmine is correct, but even being the monopoly holder of such a wonderous contraption does not vitiate the need for accounting. How much gold have I extracted from this mine? How much have I returned to it? What is the amount of gold I have extant at this moment? You know that by counting G, and T, with the difference (G-T) being gold still out in circulation. The Government must take more gold out of the mine than it puts into the mine in order for there to be any gold above ground at all. Whether you calculate this as G-T (a positive amount) or T-G (a negative amount) doesn’t matter — you either count a surplus or a deficit. But *something* must be in deficit for something else to be in surplus.

    A goldbug views gold as an asset to someone, and a liability to no one. In a proper accounting system though, every asset must be balanced with a liability. Depending on what you choose to label things, this can create some negatives in your spreadsheet cells. Deficit does not mean loan.

    — Let’s take Scenario 2 first

    I ask you to desist from the term “counterfeiter”, “thief’ etc. etc. Let us discuss before we judge.

    I hope you agree that Government action, further slugs of G-T, are not the sole cause of dilution in an economy, *any* type of debt financing is as it grows balance sheets. In fact, anything that increases balance sheets causes dilution. So your target is not Government deficits, but any kind of debt at all. Maybe the Christians and Muslims were right?

    So, we can dismiss *increasing* Govt deficit financing as the villain in this piece and shift our cross hairs to debt, or if you prefer, net expansion of private sector balance sheets.

    — Scenario 1

    I struggled to make sense of what economic growth would mean in a world of truly fixed money supply (so, no debt financing at all, every balance sheet remains fixed in size, etc.) I find it challenging to keep distinct real goods and services from the money used to count and trade them.

    I don’t think the standard GDP number is useful, as it’s nominal and we really want to talk about whether a real economy can grow and have balance sheets — which are always to be in nominal terms — stay the same.

    But while all balance sheets will sum to the same (which is what I said), GDP need not. If I choose to spend, that transaction gets recorded in GDP. If I choose not to spend, then it will not. Balance sheets can stay exactly the same, but everyone’s income can fall to zero. Velocity matters to GDP calculations, and cash flow and income statements for the non-Sovereign sector as a whole. It does *not* matter to the consolidate balance sheet of the non-Sovereign sector. We have banished debt deflation, but animal spirits are still very much with us.

    Still, no credit means no credit collapse. The fortunes of Investor, Bystander, and Entrepreneur may change, but one’s gain will always be another’s loss. The system as a whole is stable, but it may also be stagnant.

    I’m finding it hard to imagine an economy with no loans at all that can still grow. Ironically, maturity matching, which you have convinced me on, has become a moot point! D’oh!

  84. BSG writes:

    WS – while I hope Mencius weighs in more thoroughly, I’ll address a few issues you raise:

    If every unit of currency represents, for example, a fixed amount of goods at a given moment in time, lending that unit does not cause dilution of anything.

    Now, things do get a bit dicey given that as a practical matter money is used to represent future delivery and you can have delivery failures. While that is probably impossible to ameliorate completely, why add the massive problems guaranteed to arise when some people (banks, government) are allowed to create new money out of thin air representing nothing and automatically diluting the holdings of those who actually produce tangible goods and services?

    As for imagining an “economy with no loans at all that can still grow”, try this: people become more productive, find previously undiscovered resources, have more children who grow up to produce things, productive immigrants arrive, etc. and so forth. Loans can help, even a lot, but are not an essential condition of growth.

    Even though a barter economy is far less efficient, it can still grow. It is likely that the increased efficiency and convenience provided by money is what allows monetary systems to survive even when they are so frequently abused.

    It’s one thing to say that an economy can grow even when its monetary system is abused and quite another to turn around and suggest that said abuse is a necessary condition for that growth as so many defenders of the current system seem to do.

  85. winterspeak writes:

    BSG: “If every unit of currency represents, for example, a fixed amount of goods at a given moment in time, lending that unit does not cause dilution of anything.”

    Nope. Every unit of currency is simply a number in a spreadsheet.

    During lending, the lenders balance sheet and the recipients’ balance sheet gets larger. If that loan is repaid, both balance sheets stay at this new, inflated size. If the investment fails and the loan is not repaid, both balance sheets shrink back to the original size (or smaller). If the borrower takes the money, then his balance sheet remains as big (and his liabilities get restructured) while the lenders balance sheet shrinks. It’s the same as if the lender simply gifted money to the borrower. Certainly in the gift scenario, there is no dilution.

    Are loans essential for growth? Do balance sheets have to grow as an economy grows? Can a corporate entity grow, and have its balance sheet stay the same size. What does “grow” even mean if your assets, and liabilities, are remaining constant in size?

    It really is important to think of this from a balance sheet perspective, as “loans” gets our intuitions wrong. Can a real economy grow and have a constant number of assets and liabilities on its consolidated balance sheet?

    You have balance sheets in a barter economy too — they would just be different from what we think of as a balance sheet. I don’t think that not having money gets you away from accounting.

  86. mencius writes:


    I will enthusiastically agree to desist from labeling a sovereign as a “thief” or “counterfeiter.”

    Actually, though it is probably not clear from what I said, I did not mean that a sovereign can be a thief or a counterfeiter. This is indeed quite impossible. The definition of a sovereign is that a sovereign is above the law. Its will is its own law. (Eg, in the US, the Supreme Court holds final sovereign authority: the law is whatever the SC says it is.)

    Therefore, a sovereign cannot steal from its own citizens, because any such transaction already has its own word: taxation. Similarly, it cannot counterfeit, because this transaction too has its own word: minting. The illegal profit that would accrue to the private counterfeiter, the difference between the exchange value of the minted currency and the cost of making it, is the legal seignorage of the sovereign. I suspect there are few here whose belief in sovereignty is as strong as mine.

    Rather, I intended to make three points with the comparison. Let me try to state them more clearly.

    One: counterfeiting is equivalent (both economically and legally) to theft. This is clear intuitively, and even clearer under normalized accounting. Both a thief and a counterfeiter are economic parasites.

    Two: a lending counterfeiter – one who does not spend his money on goods and services, but lends or otherwise invests it – is just as parasitic as one who blows his laser-printed roll on Rolexes and a Rolls-Royce. Ie: his presence in an economy is harmful overall – though it may be welcomed by Rolex in the latter case, and Wall Street in the first.

    Three: the impact on all other economic actors of an authorized money-printer (ie, a sovereign engaging in legal seignorage) money-printer is no different from the economic impact of an unauthorized money-printer (ie, a private citizen engaging in illegal counterfeiting). This is true regardless of whether the new money is exchanged for loans, or for direct goods and services.

    If you disagree with one of these points, I would find it helpful to know which. Otherwise, I will continue to maintain the analogy.

    Nope. Every unit of currency is simply a number in a spreadsheet.

    I think you have been slightly overcome by the admitted elegance of double-entry accounting. If your currency is gold, for example, it is not simply a number in a spreadsheet. It is a physical object.

    Similarly, a house is most certainly an asset. But it is not a number in a spreadsheet. It is a house. Let’s say, for clarity, that there is no mortgage on the house. We can say that the house exists on the balance sheet of a corporation, which is owned by the homeowner. So it is balanced by equity. But who owns the equity? The homeowner. At the end of the chain, we are just looking at property, no matter how many intermediaries you add.

    (Why, then, must corporations have balance sheets that balance? Because a corporation is not a real person. Directly or indirectly, it must be owned by some person or persons. So any excess of assets over debts is balanced by equity held by the stockholders. But (a) this is equity, not debt; and (b) since no one can holds shares in a person, a person can have more assets than liabilities. And indeed many of us do.)

    Moreover, if we accept normalized accounting (a principle you have agreed to, I think), in which a monetary balance is represented as a fraction of the dollar supply outstanding, all balances summing to 1 – it is impossible (not physically impossible as with a gold standard, but mathematically impossible) for all normalized balances to increase at once.

    This reveals another fact which is intuitively obvious, which is that a monetary system is not a mechanism of production. Currency is not capital. A Krugerrand, even a pile of Krugerrands, is not a factory. Therefore, our accounting system should reflect this reality, and be zero-sum as well.

    Note that all the economic practices you endorse are possible, for a sovereign who can adjust everyone’s monetary balance at will, under a normalized accounting regime. However, normalized accounting displays a significantly different reality. In order for G to exceed t, the sovereign must confiscate normalized balances. The apparent free lunch disappears.

    Thus, rather than a philosopher’s stone, the inflationist has discovered the next best thing, at least from the sovereign’s perspective – an invisible tax. For me, this is quite sufficient to explain the historical popularity of these policies, and the various theories behind them.

    During lending, the lenders balance sheet and the recipients’ balance sheet gets larger.

    Depending on the definition of “gets larger,” this could be interpreted in a number of ways. But using the simplest definition – the sum of both columns increases – I disagree.

    The recipient’s balance sheet expands, because he adds cash on the asset side and a debt on the liability. The lender’s does not, because the lender is simply exchanging one asset (cash) for another (the recipient’s bond). Presumably, at least when the loan is instantiated, the market prices of these assets are equivalent. And, of course, the lender assumes no liability.

    So the above statement is true only for a counterfeiting or minting lender, who creates the currency he lends. But even his business can be separated into the minting side, which does expand his balance sheet, and the lending side, which does not.

  87. mencius writes:

    I’m finding it hard to imagine an economy with no loans at all that can still grow.

    An economy with a closed money supply most certainly contains loans. It’s just that each borrower who gains the use of money for the term of the loan, there must be a lender who loses it. The system is, again, watertight.

    But I suggest you reread my answer to scenario 1. The word “grow,” which is of course an analogy, conceals quite a few assumptions.

    Here is another way to think of it. Again, consider “the economy” as the consolidated balance sheet of all businesses. This balance sheet can be seen as a business in itself, Allcorp. Then “growth” consists of an increase in the sales of Allcorp.

    An interesting question is: what is the market capitalization of Allcorp? In other words: what is the net market price of all the debt and equity that Allcorp can issue?

    Clearly the market capitalization corresponds to the total returns (sales minus costs) that Allcorp can receive over time. There is an obvious bound on this value: the total amount of currency that Allcorp’s customers have. If it extracts all of it from them, leaving them with none, it will have done its job to perfection. If can perform this feat instantaneously, its value will be equal to the value of said currency. But this is truly unimaginable, so it must be less.

    Suppose, for example, that all of Allcorp’s shareholders and debtholders are aliens. They have an insatiable desire for pictures of dead presidents. When they receive these lithographs, they pack them off to Alpha Centauri and save them forever. None of the returns of Allcorp are “recycled.”

    But the value of securities cannot depend on who holds them. And Allcorp, in particular, cannot possibly increase its market capitalization by printing Allcorp shares and issuing them to its customers. If aliens hold 100% of Allcorp, they cannot increase their returns by doubling the shares outstanding and giving each share to a randomly selected Earthling. The same is true if they start by holding 1%. At best, all the returns will be recycled and the value of the aliens’ holding will not change. In reality, it will probably decrease, because the loss from dilution will not be made up fully by the customers’ increased purchasing power.

    (It is interesting to observe, therefore, that the market capitalization of all industries today exceeds the quantity of currency in circulation by almost two orders of magnitude. I feel this is a good indication of how dependent our financial system is on the printing press. Ie, it indicates how far securities prices would fall if the Fed suddenly and permanently recused itself from the sovereign privilege of minting new currency. True hyperdeflation.)

    But the most important observation is that with a fixed (or normalized) money supply, Allcorp’s sales will be stable (but not static). So should its profit margins. Therefore, its capitalization should be stable (but not static). And therefore, so should its balance sheet.

  88. mencius writes:

    I should also mention the use of the words “nominal” and “real.” I always translate “nominal” as “actual” – ie, “nominal GDP” is the actual number of dollars received by Allcorp’s sales department in the year.

    The term “real,” in the sense you are using it, is a piece of jargon whose origin is in 20th-century (or late 19th-century) inflationist economics. Historically, the individual most responsible for our present use of the word is Irving Fisher, whose argument for a “commodity dollar” presaged our modern use of the CPI to “adjust” actual numbers. You’ll note that no such adjustments are, or ever have been, used in private-sector accounting.

    Use of normalized accounting completely obviates this technique, because instead of correcting subjectively and imperfectly for monetary dilution, we are correcting objectively and precisely. And obviously, if the money supply is fixed, normalized accounting and “nominal” accounting produce identical results.

  89. mencius writes:

    And finally, I need to note the extremely dangerous and subversive suggestion that everyone in the economy should be able to act as a self-minting lender, not just the Fed.

    In other words, everyone can borrow an infinite quantity of dollars from Pluto, at an infinite term, in exchange for noting on their balance sheets that they owe a debt to Pluto.

    What happens? Everyone does exactly this. They borrow an infinite amount of money with a thousand-year maturity, and spend it. They need have no concern for the amount of debt they are incurring, because it matures only long after their deaths. And in any case, the currency will become worthless, thus making the debt easy to repay. Even if Pluto limited his loans to 10 years, the result would be the same: instant hyperinflation. Despite the fact that all balance sheets are balanced.

  90. BSG writes:

    I’ll just add to mencius’ opus magnum that it is plainly evident that humans are capable of producing things in increasing amounts (at least for a time.) That seems like a plain, albeit highly simplified, definition of economic growth. If you are looking at the world through a filter that includes accounting conventions that obscure that fact (with or without loans) why not come up with a better accounting system?

    Just because some people way back when imposed on us a system that served their purposes (for economic parasites, aka rent seekers, to extract the most wealth from most others with least resistance), doesn’t mean it is set in stone. While it does appear quite difficult to change, it is also quite worthwhile.

  91. john c. halasz writes:


    You really should improve your manner of addressing arguments, if you actually want to address any real issues rather than just adventitiously skew matters to accord with your own peculiar obsessions, as if to “prove”, to yourself, if no one else, your own “cleverness”. (And Humpty-Dumpty games of perpetual nominal redefinitions do your case no credit).

    “This claim is unequivocally false. Production of goods and services does not require production of currency.”

    No, it requires the production of profits, which is the actual “hoarding” taking place, i.e. what can’t be reduced to prior accounts. And the point is those profits need to be recirculated through reinvestment and/or increased consumption, in order to absorb increased real output and to maintain the “value” of the extant network of real capital stocks, which have no other “value” than that which is realized through revenues from the sale of output. (This is part of why the predominance of oligopolies is such a problem for “free markets”, because, though their rents are partly “justified” by high technical levels of productivity and the need to manage long-term fixed capital stocks, they tend to stove-pipe and seal-in profits, skewing income distributions and domestic “terms of trade” and forestalling further productive and innovative investment that don’t accord with their market-dominance, not to mention their ability to convert their market dominance into capture of government policy and regulation. “New Keynesians” note that “imperfect competition” results in “sticky prices”, but that does not go anywhere far enough).

    Sez the man who believes that counterfeiting isn’t counterfeiting, if the government does it.”

    Sure, ask any lawyer or Secret Service agent. Though, again, adventitiously redefining terms out of their stable natural language meanings won’t do. Even qualified lawyers aren’t really capable of doing that.

    “John, are you curious at all about the history of the ideas that you yourself hold? You might enjoy looking into the biographies of Richard T. Ely, Irving Fisher and John Maynard Keynes.”

    What exactly would those 3 economists have to do with one another, a German historicist Progressive reformer, a neo-classical disciple of Wicksell, and a critic of neo-classical optimum equilibrium assumptions? And why would you assume a) that I am not aware of the history of economic ideas, and b) that any of those three would be prime sources of my thinking or understanding. But more to the point, you’re just gratuitously engaging in ad hominem argument, without making any “substantive” point. I myself don’t mind if someone entertains religious beliefs, so long as s/he is otherwise rational and responsive, and I wouldn’t evaluate such beliefs or their import outside their historical context, wherein they might be quite commonplace and not especially “maniac”. (“Religion is what a man does with his privacy”, said Whitehead. And I find the matter about as interesting as someone’s boasts about his/her sex life, which is generally in inverse proportion to its passion or meaning). And if that buggerer Keynes was an “aristocrat” more dissolute than Nero, someone should have informed the hierarchs of Cambridge U. But then a person’s motives do not necessarily vitiate the quality or validity of his/her ideas, eh? What point could you be making, other than to manipulatively defend your own sacred idols, by slandering all other economists? On the other hand, my placing the Austrian school in its historical/ideological context serves solely to bring out the oddly archaic quality of its thinking, at once highly concretistic and fixedly static. Further, it oddly abstracts from some key features, which make capitalism, er, well, capitalistic, which abstraction is not at odds with its concretism, but its counterpart. And the version of “liberty” it extols is fairly transparently a defense of hierarchical privilege, which it oddly imagines a completely “free” market would uphold and reproduce,- again abstracting from some obvious features and consequences,- if it were only “free” enough. None of that is a matter of any ad hominem or genetic fallacy. Rather it’s a matter of educing some peculiar lacunae of that school of economic thinking. And pointing to their reactionary origins.

    And if you want to appeal to the history of economic ideas, might I recommend Karl Polanyi’s “The Great Transformation”? He was a senior editor of the Viennese equivalent of the WSJ, until he was fired for suspect political sympathies, after street fighting in Austria had resulted in the ascendancy of the hard right, several years before Anshluss. So he was thoroughly versed in the ideas of Austrian laissez-faire “liberalism” and part of the point of the book, though it’s not often noted, is an extended polemic against the effect of those ideas, especially in giving the lie to its typical plea that it is the antipode of German fascism and had nothing to do with its rise.

    It’s amusing to me how you utterly fail to see how unrealistic and unworkable your ideas are, though you’re obviously caught up in some gratuitous moralizing. But economics concerns functional matters, and has only a glancing relation to morality or ethics. Then again you’re apparently lacking in any “serious”, plausible account/analysis of human agency, “freedom”, imagining that it resides in some factitious “quantity”, volition, which is the property of atomic individuals, and any “violation” of which is an obvious outrage to morality. I like to ask libertarians, though I don’t know if you’d so label yourself: how is freedom possible in a causally determined world? It’s not like no answer can be given; several plausible accounts might be offered, provided suitable conjunctions of causality of suitable sorts, at least partially indeterministic in “nature”, which would approach our feelings or intuitions that human agency is a real phenomenon. But the question forces one to realize that it is a limited, finite “quantity”, and never entirely unconditioned and self-determined. Further, any such account would involve language/symbolic thinking, which means that agency is always social and bound up in relations and interactions with others. Hence not only is society not a matter of additive relations between atomic individuals, (and hence neither are markets, which can only occur within an institutional frame, which supports and “authorizes” them), but such atomic individuals themselves do not, because can not, exist. Hence no arrangement of social relations is or can be entirely voluntary, nor could that be a self-sufficient criterion for evaluating the former. Further, there is no system of contracts that could resolve such relations without large and consequential externalities, (which, e contrario, would amount to assuming the omniscience of lawyers, which even their grandiose self-conceit wouldn’t allow for), nor could any system of social exchanges, of which markets are a dominant case, remain static and recur to a prior equilibrium based on prior intentions, since any such system is one of cross-secting mutual constraints, which “constitute” both the system and its agents in the first place, such that the “system” is formed and re-produced through its functional capacity to handle the consequential transformation of its conditions, which has to do with agents being agents and, er, acting, in the first place. Which is to say, that such atomic individualism is of no “moral” worth, and fails to “ground” any sort of ethics. Its claim to speak on behalf of “liberty” or “freedom”, above all else, anything short of which is violent coercion, are not just incoherent, but actually involves a displacement of such violent coercion. But still more fundamentally, it appeals to a system of exchange that would vouch-safe it, that is, er, without consequential exchange, mirabile dictu, which is, indeed, an archaic “aristocratic” fantasy. Passingly strange as an account of a “free” market economy, still stranger as an account of
    its functional “virtues”, as producing real economic growth on behalf of the hoi polloi. I don’t want to get into some elaborate dispute over “methodological individualism”, though I obviously think it’s an inadequate approach to developing a realistic account of an economy, (since institutions and organizations are more relevant “agents”), but I’ll just note it does not “translate” into any account of “substantive” individualism, and the mistaking of the former for the latter, of functional criteria for “moral” right is the very height of folly, indeed, “divine comedy”.

    You apparently fail to grasp that money is simply a “universal equivalent”, rather than any actual “object” of exchange, a reduced symbolic medium, a semiotic instance, representing “value” without having any of its own, which only acquires “value” through circulation and exchange. And that is the case whether its “medium” is gold, sea-shells, “real bills”, $, or secured electronic transmissions. You further fail to grasp that any lending involves “credit”, (Latin for belief), the obverse side of which is debt, (which probably goes back etymologically to blood-money in tribal systems of exchange, Geld = guilt). I don’t lend my ox to let you plow your field and forgo plowing my own field, but rather I lend my surplus corn to allow you to buy an ox. Money, derived from the circulation of exchanges, is different from debt, but also allows such debt transactions.There is no way to reduce such debt to “prior” monetary exchange, anymore than one can convert corn to oxen, other than through a sequential production process. I’ll go further. The notion that lending is time-preference might seem intuitive: I forgo current consumption in order to allow you to engage in consumption, while increasing my resources for future consumption. Except that a) there is a more-or-less large population of potential lenders with differing time-preferences, which evens-out such an issue, and b) productive consumption increases the very surpluses, derived from prior exchange, which “pays” for such lending and more. Further, the extension of such time-preference into such productive consumption and into an account of the “round-aboutness” of production misses that production is a constantly ongoing process. Yes, there are vertically integrated stages of production, but they are constantly re-adjusting to one another, under ongoing, changing conditions, and there is nothing to say that such production processes are market mediated, “outsourced”, rather than contained in vertically integrated oligopolies or Japanese-style “keiretsu” arrangements. Which is to say, that there is nothing to say that the current rate-of-interest is determined by time-preference, rather than the current state of the productive economy, (and, er, the need to recycle such corporate profits and the lack of available opportunities for such). The “marginalist” account of production,- (and, yes, the Austrian school is a fully paid-up member, since it was only after they realized that their account couldn’t “work”, that they turned to eschewing the “scientism” of mere mathematics),- fails precisely on the score that production systems can not be reduced to market exchanges. There are re-production requirements that introduce constraints that are far different from market-exchange constraints.The sphere of production is cross-dependent or cross-implicated with the sphere of circulation and exchange. But neither is reducible to the other

  92. winterspeak writes:

    MENCIUS: I have more points to add, but it’s late and I’d like to spend more time digesting your posts.

    When I asked you to desist from “thief” and “counterfeiter” I was not talking about the Sovereign, I was talking about banks.

    You keep slipping into gold standard definitions of money again, which are just not true. “Currency is not capital.” Yes!

    “Therefore, our accounting system should reflect this reality, and be zero-sum as well.”

    I’m trying! But someone keeps bringing up these non-zero sum gold entities!

    In my money system, also known as fiat money, also known as current reality, money and accounting is zero sum. You have the public deficit that funds non-public savings (both netting to zero). And you have non-public debt financing non-public savings and investment (both netting to zero). The non-public assets and liabilities can both cancel each other out to zero.

    Note that in the former a deficit is *funding* savings while in the latter, debt is *financing* consumption and investment.

    Before going any further, can we agree that, at least initially, the Sovereign needs to run a deficit in order for the private sector to have some currency at all, and that this is true whether you’re talking about a monopoly infinite gold mine, or numbers on a spreadsheet? I think that if I cannot get you to agree to that, then I cannot get you to embrace a balance sheet perspective on money and we can go no further.

  93. winterspeak writes:

    Mencius: (You’re also too quick to dismiss Irving Fisher. Yes, he was crazy, but his craziness also caused him to lose all his money on the stock market crash, and *change his mind* about his economic ideas afterwards. Anyway changing their mind as a consequence of reality is so rare, that I think it’s worth comparing before and after. This account of debt deflation is excellent, and I think you would agree with much of it.)

  94. winterspeak writes:

    MENCIUS: A few more thoughts, but not too many. I still think that we should settle the key question first before moving forward: “can we agree that, at least initially, the Sovereign needs to run a deficit in order for the private sector to have some currency at all, and that this is true whether you’re talking about a monopoly infinite gold mine, or numbers on a spreadsheet?”


    A normalized account system would net to 1, not 0. A balance sheet system sets to 0. If you introduce a Sovereign who can carry a debit, and never needs to pay it back, then the non-Sovereign money would collapse to the credit that debit enables. Combining the non-Sovereign and Sovereign balance sheets gets you to zero.

    Balance sheets always balance, because equity is in the liabilities column. The quantity of equity vs other liabilities is “leverage”. Even individuals who own their house have balanced balance sheets — it’s just that their only liability is equity, which they own.

    “”During lending, the lenders balance sheet and the recipients’ balance sheet gets larger.”

    Depending on the definition of “gets larger,” this could be interpreted in a number of ways. But using the simplest definition – the sum of both columns increases – I disagree.”

    Hah! How about we use the normal definition? The asset column always equals the liability column. Just pick one sum. This was my point about not allowing any debt. If you allow debt, both sides of the balance sheet grow. If you only allow equity, then one balance sheet must shrink for another to grow. This gets us to your normalized accounting world.

    In a debt world, the lender creates a credit entry on their own balance sheet (receivable) and a debit entry (loan). The borrower creates a credit (cash) and a debit (payment due). Both lender and borrower have bigger balance sheets as a result of this transaction. If you eliminate debt, then the lender writes down cash and creates an “ownership” credit–balance sheet stays the same size. The borrower increases cash and adds a “payable” on the liability side–balance sheet is larger. The borrower needs to make someone else’s balance sheet smaller to sustain this larger size. If he succeeds, he keeps the larger size, and pays off the lender, who changes his payable asset back to cash. If he fails, then he reneges on the payable, and maybe he still has the cash or maybe he’s blown in. The lender writes down his receivable and equity — smaller balance sheet.

    “I’m finding it hard to imagine an economy with no loans at all that can still grow.

    An economy with a closed money supply most certainly contains loans. It’s just that each borrower who gains the use of money for the term of the loan, there must be a lender who loses it. The system is, again, watertight.”

    It also has loans that look exactly like equity investments.

    “And finally, I need to note the extremely dangerous and subversive suggestion that everyone in the economy should be able to act as a self-minting lender, not just the Fed.”

    It is dangerous and subversive! In practice, we have one “private” entity with the charter to do this, known as a “bank”. Banks mint money, and they are meant to only be able to mint money in proportion to increased economic growth (does not show up in CPI, does dilute currency).

    This was the point I was trying to make. In the real world, we have two classes of entities who can mint money (via making loans). One is the Sovereign, and one is banks. Your target as chief dilutionist has been the Fed. I’m trying to point out that Banks dilute as much. If you want to target dilution, you need to target debt financing (as it exists today in reality).

  95. BSG writes:

    WS: “Your target as chief dilutionist has been the Fed. I’m trying to point out that Banks dilute as much. If you want to target dilution, you need to target debt financing (as it exists today in reality).”

    As I recall, several here, including Mencius, JIMB, IA, myself and perhaps even Steve have repeatedly said that fractional reserve banking (your “as it exists today in reality”) is a major culprit and explained why. Have you been agreeing all along?

    >”Banks mint money, and they are meant to only be able to mint money in >proportion to increased economic growth (does not show up in CPI, does >dilute currency). ”

    Since the Fed was created, has there been a single year outside the Great Depression in which the CPI did not rise? How about asset prices, most of which are excluded from the CPI?

  96. mencius writes:


    But more to the point, you’re just gratuitously engaging in ad hominem argument, without making any “substantive” point.

    If you read my original response, it (a) was directly preceded by a substantive argument which you have not addressed, and (b) was posed directly in response to your phrase “Austrian aristocrats and wannabes, nostalgic for the lost glories of their dying empire and the hierarchical “liberty” it allowed.”

    Was this (a) not ad hominem, or (b) not gratuitous? What on earth can it possibly have to do with the logic of monetary dilution?

    For your information, the individuals I named (Ely, founder of the AEA; Fisher, inventor of the commodity dollar; Keynes) bear substantial personal responsibility for the abandonment of what was once called “orthodox finance,” ie, the adoption of the 20th-century policy of unrestrained fiscal and monetary expansion.

    BTW, you have studied well – your clouds of meaningless pseudo-rhetoric are indeed reminiscent of your master Keynes. I wrote:

    Production of goods and services does not require production of currency.

    You wrote:

    No, it requires the production of profits, which is the actual “hoarding” taking place, i.e. what can’t be reduced to prior accounts. And the point is those profits need to be recirculated through reinvestment and/or increased consumption, in order to absorb increased real output and to maintain the “value” of the extant network of real capital stocks, which have no other “value” than that which is realized through revenues from the sale of output.

    Arguing with this great cloud of pure doxology would be like arguing with a Marxist or a Derridan deconstructionist. Ie, like wrestling with a pig.

    But to get slightly muddy for a second, note that you have completely failed to answer my point – no one in your loop produces (ie, mints) money. You start by redefining the word “production,” then introduce the completely undefinable concept of “real value,” a fudge factor which can mean anything at all, then appear to end up with some kind of aggregate tautology of the “amount of rain that hits the ground” variety. And this is perhaps the most coherent paragraph in your comment!

    If readers want to see said pig-wrestling actually performed, on Keynes himself, by someone with much more patience than myself – try the work I have already recommended, Henry Hazlitt’s Failure of the New Economics.

    Note that Hazlitt was writing in the ’50s. By my count, the New Economics has failed about six times since then. Possibly seven. Whatever it is, it has quite a few more lives than a cat.

    But since I do not have Hazlitt’s patience, and since I find this exchange typical of our interaction, I doubt further discussion will enhance either of our understandings of economics.

  97. mencius writes:


    Before going any further, can we agree that, at least initially, the Sovereign needs to run a deficit in order for the private sector to have some currency at all, and that this is true whether you’re talking about a monopoly infinite gold mine, or numbers on a spreadsheet? I think that if I cannot get you to agree to that, then I cannot get you to embrace a balance sheet perspective on money and we can go no further.

    How on earth could I accept this? Before the 20th century and excluding the brief innovation of Kubla Khan, all currency was metallic and was produced by finite, non-monopoly gold and silver mines. Not government deficits.

    You: “During lending, the lenders balance sheet and the recipients’ balance sheet gets larger.”

    Me: “Depending on the definition of “gets larger,” this could be interpreted in a number of ways. But using the simplest definition – the sum of both columns increases – I disagree.”

    You: “Hah! How about we use the normal definition? The asset column always equals the liability column.”

    Yes, of course. By “both columns” I meant “each column.” This was not some new accounting innovation, just clumsy English.

    But go back and read my point again. The lender’s balance sheet does not increase, because he is exchanging one asset (cash) for another of equal price at the time of the transaction (the loan).

    You: “In a debt world, the lender creates a credit entry on their own balance sheet (receivable) and a debit entry (loan). The borrower creates a credit (cash) and a debit (payment due).”

    No. This is the heart of your error, I think. This is just wrong.

    Your lender is always a minting lender, and you are combining two transactions into one. One: the lender creates money. Two: the lender exchanges money (present cash) for a loan (a promise of future cash).

    Operation one creates money and expands the lender’s balance sheet. But it has nothing to do with lending. Operation two is the essence of lending – but it does not create money or expand the lender’s balance sheet.

    The essence of a loan is that present cash is exchanged for a promise of future cash. This transaction does not involve monetary creation.

    It is dangerous and subversive! In practice, we have one “private” entity with the charter to do this, known as a “bank”.

    No, not even. We have one entity with the charter to do this, the Fed. (Whether the Fed is legally “private” or not is to some extent a matter of opinion, but it is obviously official in reality.)

    The Fed can create dollars and lend them. In the US today, a commercial or investment bank cannot lend dollars it doesn’t have. It has to borrow them, ie, from depositors. Or sell shares. Or whatever. But it most certainly does not have the right to create dollars out of thin air.

    The banking system in effect, through the magic of guaranteed maturity transformation, can create money. Or rather, it can teleport it from the future. But this mechanism, she is not working well at the moment.

    Banks mint money, and they are meant to only be able to mint money in proportion to increased economic growth…

    Aha! This doctrine is a key part of 20th-century loose-money doxology. About a century ago, the term was “elastic currency” – the idea was that a fixed supply of money (or even a gold supply that expands through mining only) does not allow for, well, “economic growth.”

    (Since, as I explained earlier, “economic growth” is a statistical artifact of inflation itself, this is in a sense true. But not in a meaningful sense.)

    But the point is that said elastic currency was supposed to, in some mysterious automatic sense, expand “to meet the needs of trade.” Since there is no ordinary commercial transaction that involves, much less requires, the creation of currency, this makes about as much sense as most early 20th-century pseudointellectual propaganda.

    In my money system, also known as fiat money, also known as current reality, money and accounting is zero sum.

    All I mean by “zero-sum” is that every transfer of money is zero-sum in the game-theory sense – ie, the transaction does not increase or decrease the amount of money in the system. This total is most certainly positive.

    As for fiat money, again: fiat currency is not in any way qualitatively different from a gold standard, as my infinite-gold-mine standard generates. It is a difference of degree. You can go from an infinite gold mine to a finite but very large one, to Fort Knox, etc, etc.

    You can also see the continuum in terms of equity: from USG dollars, to IBM shares, to GLD shares, to actual pieces of gold. The question for both IBM and GLD is: what gives the shares value? What does their owner own? A piece of property. This is not a liability, except in the sense that it is the property’s liability to be owned by its owner.

    Perhaps this is the sense you mean? In your abstract world, does each Krugerrand have a balance sheet? Nonetheless, it is a physical object and cannot be created by Excel.

    [Fisher’s] account of debt deflation is excellent, and I think you would agree with much of it.

    I think I’ve seen an excerpt, and from what I recall I think you’re right. Although I’m not sure Fisher’s analysis was all that original. Mises had certainly provided quite a compelling explanation of the structure of a credit contraction, back in 1912, and I don’t think he was the first either.

    I am not quite sure as to what the post-1929 Fisher’s views were. The Fisher I know is the “commodity dollar” man, ie, the champion of price indexes. I don’t know that he repented of this. But who has?

  98. winterspeak writes:

    BSG: I don’t agree. Fractional reserve banking is a red herring. I wish they would do away with it once and for all. Reserve requirements have no impact at all on bank lending, and does not “multiply money” they way people think it does.

    Banks multiply money because, thanks to their charter with USG, they can create the money for loans ex-nihilo and expand both sides of their balance sheet at the same time. They’re allowed to carry overdrafts — that was the point I was trying to make in the example. Just as USG runs a deficit to fund private sector savings, the banking sector extends loans to finance private sector consumption and investment. I cannot loan you money I don’t have, but banks can. As we have seen, they don’t need deposits to do it — those really just, essentially, sit doing nothing. Goldman Sachs has no deposits, and it can make loans just fine.

    Capital requirements are a better constraint on bank lending, but they are easy to violate (as we have seen).

    “>”Banks mint money, and they are meant to only be able to mint money in >proportion to increased economic growth (does not show up in CPI, does >dilute currency). ”

    Since the Fed was created, has there been a single year outside the Great Depression in which the CPI did not rise? How about asset prices, most of which are excluded from the CPI?”

    I would not for a moment claim that the $ has not been systematically diluted since the Fed was created. I don’t like CPI any more than Mencius does, but between 1909 and 2009 fall in the dollars value has been dramatic and obvious, whatever CPI says or doesn’t say.

    I don’t know how much of this dilution has been driven by the Fed, and how much by banks. The run-up in house prices was primarily a private sector credit phenomenon, although USG’s hands are not clean either. Certainly it was massively dilutive to those who are short housing (renters).

  99. mencius writes:

    Perhaps it’s worth pondering the reason we have balance sheets at all. What is a balance sheet, anyway? Why do we have balance sheets, rather than nothing?

    Imagine you are building a virtual world, like a MUD or MMORPG. The first feature you need is an inventory – a list of the goods carried by a player. These are assets. There are no liabilities. Before debt, there is property. An inventory is not a balance sheet.

    A balance sheet is needed when players start making promises to each other. What is a promise worth? It depends what a player has to fulfill it with. Therefore, any purchaser of a promise will demand to see the balance sheet of the promiser, which describes (a) what assets he has, and (b) what promises he has made. The essential question is: can he be expected to fulfill all his promises?

    On the left, the balance sheet contains a copy of the player’s inventory. On the right, the balance sheet lists all promises to deliver that the player has made. The left side is not to be confused with the inventory or with the assets itself; the right side is a list of the promises, not the promises themselves.

    (Note how this can be extended from the usual, simplistic scalar definition of balance-sheet solvency (total realizable price of assets, sum of promises) to prohibit maturity transformation. We construct a time schedule of incoming and outgoing cash flows, requiring that outgoing cash come from incoming cash which is scheduled to come in before it goes out.)

    But a balance sheet is just a tool. It is not reality itself. In reality itself, people own assets and they make promises. They cannot make promises without owning assets, but they can own assets without making promises.

  100. mencius writes:


    Banks multiply money because, thanks to their charter with USG, they can create the money for loans ex-nihilo and expand both sides of their balance sheet at the same time.

    I am not a banking expert, but if this is so, why is there such a thing as interbank lending? Why don’t the banks just borrow infinite amounts of money from themselves – at zero interest? What is the interest rate, anyway, when you borrow from yourself? And the maturity? And why bother with deposits, at all, ever?

    In the reality I’m familiar with, Goldman may be able to borrow (or, more precisely, repo) dollars from the Fed. (Thanks to the latter’s new alphabet soup of garbage-pawning facilities.) But Goldman cannot either (a) increment its own bank account by borrowing a dollar from itself, or (b) print an actual dollar bill, any more than I can. When Goldman lends dollars, it has to get them from someone else, just like anyone else.

  101. winterspeak writes:

    MENCIUS: I don’t want to mess with the great Khan! And yes, there have been various forms of currency throughout the ages. That said, if the great Khan established a longer period of Sovereignty, he would establish a fiat monetary system. And when he did, he would have had to run a deficit to give the private sector their seed money.

    I’ll rephrase my question as “in a fiat money system, the Sovereign needs to run a deficit in order for the private sector to have some currency at all”.

    I think the question of non-fiat money is very interesting, actually, as all third world countries live on a mix of local currency, gold, foreign currencies, etc. Very few countries have the luxury of the Sovereign being a pure currency issuer, and subjects being pure currency users. There is room for a global switch to gold yet.

    — Interbank lending

    Excellent question. The Fed wants an interbank lending market because that is the mechanism by which they set the Federal funds rate. They issue Treasuries to drain excess reserves for the Fed, and thereby create this interbank lending market. It’s this bizarre gold standard remnant, and they should do away with it.

    The constraint in bank lending is capital requirements (how big can their balance sheets get on the equity sliver they have?). The purpose of SIVs was to circumvent there requirements. The purpose of CDS, CDOs, etc. etc. was to circumvent them further. The purpose of the Fed’s recent alphabet soup is to maintain those circumventions.

    We live in a world where the currency is fiat, but those in charge of it think we are on some flavor of the gold standard. That’s what gives us ideas like “reserve requirements”, “US needs to fund its deficit”, “the Govt should balance its budget” etc. etc. etc. It’s why they think you need to save banks to save the economy. If you’re going to insist on driving a tractor, sit in a tractor. If you’re going to sail a boat, sit in a boat. But sitting in a tractor and trying to find the center board is just a recipe for disaster. Sadly, that is the situation we are in.

  102. john c. halasz writes:

    Note: I hadn’t meant to post last night, and I have no idea how it happened. Rather I decided to go to bed and finish off later, simply copying to word. Apologies. However, here’s the remainder of my screed.

    And because there are proportional re-production requirements, as industries produce for each other “before” they produce for final consumption demand, production is not determined by short-run consumption demand aggregated from individual preferences, (which is why large firms attempt various devices to control their markets, to “capture” adequate demand). The standpoint of “consumer sovereignty”,- (aside from that being a very odd use of the word “sovereignty”, far removed from Bodin’s classic definition),- is largely fictitious, and does not generalize well to account for the goings-on of the whole economic system. But the more fundamental point is that it is the level of industrial profits that ultimately fund the level of supply of loanable funds, and not just the demand, i.e. need, for them, and the primary function of the credit system is to fund productive investment. In principle, consumer lending could be entirely done away with, though that would impair markets for consumer durables, such as houses. But there is no need for households to borrow to finance tertiary education, since that could just as well be funded for qualified applicants through public taxation, with those taxes paid out of the higher incomes of graduates. And, in fact, much consumer lending nowadays really amounts to vendor financing for industrial output. But, in that light, it is less a matter of loans deriving from forgone consumption than from forgone alternative production possibilities. Additionally, since productive investments are ever only an incremental addition to productive capital stocks, if they are not just a replacement for depreciated stocks, the rate-of-interest is not the sole factor determining investment decisions, as if it were a simple function of matching supply with demand. (Input/output accounts of production, such as Leontief’s and Sraffa’s, are a big improvement over purely market based accounts, but even they get their accounting somewhat wrong, in that they neglect that there is always a stock, fixed or variable, tied up in production, which, though a product of prior production, to be sure, nonetheless needs to added and subtracted from current flows for a proper accounting). Hence manipulations of interest rates, deviating from the supposed “natural” rate determined supposedly by the supply of household savings, can’t be used to account entirely for “malinvestment”, as the supposedly eliminable cause of business cycles,- (the real reasons lying in the dynamics of production itself and its resultant income distribution dynamics, as productivity increases outrun that income/demand that it generates, i.e. in the invalidity of “Say’s Law”)-, though it’s not as if bad investments and losses don’t occur and they would occur under any system of finance, since it is a matter of fundamental uncertainty with respect to future outcomes, which accompany any investment decision, and not the failure of some imaginary equilibrium condition to obtain. Of course, one could insist that real productive investment be entirely financed through selling equity, (as Islamic banking pretends to do, by disguising interest as profit-participation), but the likely result of that would be an interlinked ownership structure that would tend toward monopoly, since the source of that equity would ultimately be the profits already accruing to corporate equity. It’s little wonder that independent firms often prefer borrowing to finance their operations. Nor would an entirely equity- based finance system eliminate the cycles of corporate profits and ensure “equilibrium”. How then is the fundamental Austrian claim that time-preference determines interest rates and thus a time theory of production results derived. From the fundamental marginalist claim that “utility” is entirely “subjective” , psychological, since “interpersonal comparison of utilities” must be technically prohibited to allow for the calculation of utility preference functions to result in the determination of welfare “efficiencies”. (Yet utility is actually practical and functional and entirely bound up in our social relations, as we tend to discover our needs and desires precisely through those dreaded “interpersonal comparisons of utility” in sharing the conveyances of life among our practical activities and limited resources). Combined with the truism that supply must match demand, as the solution to all economic problems, the result is adherence to a static conception of equilibrium, (or “simultaneous determination” of the variables of those supposedly entirely independent supply and demand curves). Hence time must be reduced to an entirely subjective perception and is illegitimately abstracted out from ongoing real temporal processes, while ignoring the fundamental uncertainty that actually attaches to the future, and styled as an entirely subjective preference of future over present consumption, whereby savings is denied consumption, due its just reward for virtue, (ignoring, of course, any issue of the distribution of wealth and income, which might enable such “virtue”), rather than a matter of enhanced production and profit therefrom. When it was discovered that the systematic elaboration of time-preference accounts of interest and equilibrium savings into a time theory of production could not be worked out, (basically because it couldn’t quantify a “unit” of capital in terms of time), latter-day Austrian “saints” took to denouncing quantification and denying the very static equilibrium that their whole account, in fact relies on, all the while denouncing the actual economy for deviating from the very equilibrium that they at once asset and deny. This is logical flim-flam and sheer denegation. Levi-Strauss famously understood myths as synchronic systems that sought to “recover” and deny “lost”, i.e. passing, time. The Austrians notion of “time preference” seeks to do the same; hence its reactionary stance.

    What is so remarkable about Mencius’ account is its utter unreality and unworkability. And his refusal to recognize that fact. It’s as if an engineer would double the size and power of an engine and recommend that it run on the same amount of oil at twice the rate of circulation. Why on earth would capitalists invest in increased technical productivity and thus increased output and lower output prices if their gains would be subject to severe deflation of output prices? They so invest to increase their profit and favor lower output prices, if it gains them competitive market-share, but not so much that it diminishes rather than increases their profit. It’s like saying capitalism would be wonderful, if only there were no capitalists! It’s not any sort of serious realistic mode of economic analysis. It’s not even serious ideology, since an ideology, to be at all effective, must remain in contact with real facts, even if it must deny or conceal some set of facts to maintain the applicability of its distorted normative claims. It’s a sheer subjective fantasy world, endlessly circling about its magical fetish-object, “hard money”, in self-referential loop-de-loops which it inspires in his perfervid imagination to the point of ecstasy, without ever noticing that there is no actual or potential real object that could correspond to his fetish. Then again such cyclopean, mono-causal thinking resembles traditional mythic/metaphysical thinking, subsuming everything under its total ordering of “eternal being”. But then if reality itself fails to oblige in subsuming itself under such an ideal order, it must be because of a conspiracy of evil authorities, who are interfering with its perfect workings, rather than anything so commonplace as dysfunctions and errors. Which, or course, is paranoid thinking, unsurprising in a mind-set that exalts in the
    superiority of the individual ego, to which the world itself must submit. Hegel had a term to designate those who denounce the world for its corruption and withdraw from it into the “purity” of their perfect ideals: he called them “beautiful souls”, and the astringent irony of the terms derives from his conviction that no ideal, i.e. norm, is worth diddley-squat, unless it is anchored in a social reality and can account for itself by making sense of the reality in which it is applied.

    Finally, dear Mencius, I’m sure you’re aware of the elements of symbolic logic. The truth-function of “if p, then q”, where p is f and q is t, is “true”. Hence appeals to deduction from such hinky premises won’t do. “Reason” already called you out on that, and he’s a pretty sharp guy, a statistician by profession, who lives in Germany, though he’s an Aussie. Still, he wasn’t quite right in appealing to sheer empiricism, which is inadequate to account for the generation of adequate, realistic theoretical concepts. Something like C.S. Peirce’s account of “abduction” is required, but once such concepts are formed, their test of adequacy is, indeed, their capacity to fruitfully organize and explain the full range of empirical data. Rather than, er, replacing such data and analysis with sheer counter-factual deductions. Or accounting systems using Roman numerals. Rather than bothering everybody with your adventitious and whimsical style of argument, which comes of as more childish in its self-insistency than obscure, might I suggest you sign up of an account at “Second Life”. There you could play with your toy models of Austrian banking to your heart’s delight. (Oh, and remember that I told you that such Austrian thinking is rooted in a supposed ultra-skepticism that is really just a smoke screen for its unreflected reactionary prejudices? I checked out your web site, which fully confirmed that diagnosis. But then you probably lack the quality of intelligence,- he would say “will”,- to read Wittgenstein and grasp his trenchant critique of such supposed “skepticism”). But if you’re really starved for amusement, you could check out the comments on Steve Keen’s post on “The Roving Cavaliers of Credit”, where at the end a recent graduate of the Von Mises on-line Bible college, far less nimble and clever than you, shows up and throws a hissy-fit, convinced of his a priori righteousness, when an effort is made to explain to him in the shortest and simplest possible terms. He apparently thought “information surveillance” was some sort of reference to Big Brother, rather than market participants, and evinced no awareness that markets involve exchange of information as much as of goods, which it was one of the classic strengths of the Austrian school, especially von Hayek, to point out and develop.

  103. john c. halasz writes:


    Sorry. The latter-day remnants of the Austrian school are a cargo cult, with the likes of Murray Rothbard witch-doctor-in-chief. This is distinct from the classic thinkers of the school, especially Boehm-Bawerk and von Hayek, who did make genuine, if failed, attempts at economic understanding. The school effectively ended as a going concern when Sraffa and Kaldor, (who, aside from being an Hungarian, came from LSE, von Hayek’s stomping ground, and was originally a disciple), decisively refuted its account of business cycles.

    It’s not me who’s lacking in any serious conceptual understanding of economic thinking. Your obsession with “currency dilution” is trivial, though all-determining, preventing any actual causal analysis. I eye-balled a chart once, obviously a reconstruction, since NIPA didn’t exist for most of the covered history, of per capita British output. If one drew a straight line through the variations of the 1800-1900 period, it looks like maybe a 1% per annum rise in productivity. Then starting around 1900, the slope of that line sharply increases to maybe 2.5% productivity growth per annum. That should offer a strong hint why the alleged golden age of “orthodox finance” saw and lost its hay-day, since, er, economies evolve historically and not utterly without “reason”. Though you’re actually just referring to an historically circumscribed period of the gold standard under the hegemony of the British Empire, the passing of which nostalgists the world over mourn with plagent tears, as the passing of the very ideal of civilization itself. Oh, and you seem unaware that during that glorious 19th century, financial/economic crises in the U.S. were common, recurrent, and the economy spent nearly half the time at stagnation “equilibrium”, as opposed to 20% of the time since WW2, before now. So much for the utter failure of the “New Economics”, which Hazlitt so sagely declared in the 1950’s.

    I’m not aware of having “redefined” production. The word seems plain enough. It’s you who are constantly redefining terms, with an extreme nominalism worthy of Mauthner, (to make an obscure reference to an obscurantist). Nor do I recall using “real value” as a fundamental reference, though it is a question that all sorts of participants are constantly asking themselves and trying to ascertain. You obliterate the question, in accordance with your supreme sovereign whim, while constantly hankering after it through undiluted currency. But, in fact, all real productive investment contains a speculative element and “maturity transformation” is an aspect of all such investment, given the uncertainty that ineliminably belongs to future prospects, whether there is FRB or not. As for the “origins” of money, who cares? That’s an archaic metaphysical way of framing the matter: what counts is its current function as, er, currency. One theory is that the traditional ratio between gold and silver was established by the cycles of the sun and the moon. What’s fallacious is the notion that such currency would operate “freely” in the absence of its issuance by a sovereign. But the question I posed in response concerned the source of profits, since, in my account, it’s the recycling of corporate profits that effects the level both of loanable funds and aggregate demand, a connection that you would appear utterly clueless about. And that issue would occur, FRB or not. In fact, you have no actual analysis of the current situation, in terms of the connections between huge trade/CA deficits, housing bubbles, upward mal-distribution of wealth/income, excessive credit expansion, and low levels of real cap-ex. (But, of course, to know of any such matters, you’d have to consult NIPA figures, mere data, which according to you is at the root of all “fraud”). No, it’s a much better, sounder intellectual procedure to pursue your archaic obsessions, with a solipsistic arrogance that simply obliterates, with a variety of sophistical techniques, all other points-of-view.

    And you seems blissfully ignorant of the fact that conventional commercial banking has been in decline for decades, since the 1970’s, and the current credit-bubble debacle, (which, yes, I’ve been watching for years), is largely “rooted” in credit expansion through the “shadow banking system”, relying on “capital markets”. I’d be the last one to defend the Fed or the cult of central bankers and conventional monetary policy, but it’s not government intervention and regulation that’s at fault, but the very lack of it, since the rise of the right-wing corporate hegemony under the banner of “neo-liberalism” 3 decades ago, in which the promotion of “market” economics under the slogans of “liberty”, opportunity, and competitive vigor served as a smoke screen for oligopolistic rent-seeking, and especially Wall St./MNC sponsored globalization under the banner of “free trade”, which served to weaken governmental policy-making and regulatory capacity. But no, it’s all due to that evil gummint that seeks to oppress us free spirits, and the local yokels, who should bow to our superior insight. The whole world would be fine and just hunky-dory, if Mencius were king, and could sweep away all extant arrangements and impose his dictat, restoring the “true” currency, which is also lacking in anything so gratuitous as “real value”. But not even King Alfred could so command the tides.

    Wrestling with a pig, indeed. Coming from you, that’s probably a compliment. More like trying to argue with a snake, one not nearly half as intelligent, knowledgeable or cogent as his pedantic reptilian brain tells him.

  104. john c. halasz writes:

    I don’t know if this would help anyone here, but the following is an except from the above-mentioned Steve Keen blog-post:

    “Our starting point for analysing the economy should therefore be a “pure credit” economy, in which there are privately issued bank notes, but no government sector and no fiat money. Yet this has to be an economy in which intrinsically useless items are accepted as payment for intrinsically useful ones—you can’t eat a bank note, but you can eat a pig.

    So how can that be done without corrupting the entire system. Someone has to have the right to produce the bank notes; how can this system be the basis of exchange, without the person who has that right abusing it?

    Graziani (and others in the “Circuitist” tradition) reasoned that this would only be possible if the producer of bank notes—or the keeper of the electronic records of money—could not simply print them whenever he/she wanted a commodity, and go and buy that commodity with them. But at the same time, people involved in ordinary commerce had to accept the transfer of these intrinsically useless things in return for commodities.

    “Therefore for a system of credit money to work, three conditions had to be fulfilled:

    In order for money to exist, three basic conditions must be met:

    a) money has to be a token currency (otherwise it would give rise to barter and not to monetary exchanges);

    b) money has to be accepted as a means of final settlement of the transaction (otherwise it would be credit and not money);

    c) money must not grant privileges of seignorage to any agent making a payment.” [11]

    In Graziani’s words, “The only way to satisfy those three conditions is …:

    “to have payments made by means of promises of a third agent, the typical third agent being nowadays. When an agent makes a payment by means of a cheque, he satisfies his partner by the promise of the bank to pay the amount due.

    Once the payment is made, no debt and credit relationships are left between the two agents. But one of them is now a creditor of the bank, while the second is a debtor of the same bank. This insures that, in spite of making final payments by means of paper money, agents are not granted any kind of privilege.

    For this to be true, any monetary payment must therefore be a triangular transaction, involving at least three agents, the payer, the payee, and the bank.” ( p. 3).”

    That at least serves to bring out the public utility function of basic banking, as it emerged/evolved from historical mists. And I also think hints at why the banking “power” comes to be aligned and organized/regulated through the sovereign power, as the issuer of the currency, and the “ultimate” guarantor/arbiter of all transactions.

  105. winterspeak writes:

    Mencius: Banks absolutely do lend by borrowing money from themselves. In your zero overdraft world (scenario one) you need to have a deposit in order to offer it out as a loan. For the period of that loan, you, the lender, do not have that money since it is loaned out. A “loan” is a transfer of money, with the promise to return it in the future. Straightforward!

    In scenario 2, which is closer to the real world, banks have the ability to borrow from themselves and, as institutions, go into overdraft. How deeply they can go into overdraft is governed by capital requirements, which have proven disastrously ineffectual in the recent past. Here, instead of a deposit enabling a loan, the loan itself creates the deposit. The bank makes the loan with money it does not have, and credits the borrowers account with it (creating the deposit as a liability on its own books). It balances this liability with an asset, a receivable. The borrower balances his new asset (cash deposit) with a payable liability. Since the bank balance sheet has gone up, it is now more levered as its equity sliver has not changed.

    Two things to note: 1) the bank has diluted the currency. This is why you, targeting the Fed as the chief diluter, is wrong. In the real world, most dilution is this scenario 2 kind of loan. Banks carry overdrafts so non-bank, non-govt entities can be in surplus.

    2) Banning maturity transformation helps, a maturity matched balance sheet is much more stable than one which is not, but it does not fundamentally alter the dilution. If the bank had a time series balance sheet, and had the liability time matched with the asset, it would still have borrowed from itself and diluted the currency.

  106. mencius writes:

    I can’t resist getting a little muddy:

    What is so remarkable about Mencius’ account is its utter unreality and unworkability. And his refusal to recognize that fact. It’s as if an engineer would double the size and power of an engine and recommend that it run on the same amount of oil at twice the rate of circulation. Why on earth would capitalists invest in increased technical productivity and thus increased output and lower output prices if their gains would be subject to severe deflation of output prices?

    John, let me introduce you to a little something called the “semiconductor industry.”

    You’ll note that the price of a transistor has been deflating exponentially for the last 50 years. Nonetheless, the semiconductor industry is renowned for its extremely large capital investments (billion-dollar fabs), whose technical sophistication is constantly increasing.

    If one drew a straight line through the variations of the 1800-1900 period, it looks like maybe a 1% per annum rise in productivity. Then starting around 1900, the slope of that line sharply increases to maybe 2.5% productivity growth per annum.

    The idea that these numbers are meaningful is nothing short of hilarious. A 1% per annum rise in “productivity,” across the century of the Industrial Revolution! John, have you ever read a history book?

    And what could you possibly even mean by comparing the “productivity” of 1900 to that of 1800? Think about the goods that are being produced at each point. I cannot imagine that even the most deranged Bible-Code numerologist would have the stones to imagine that this comparison could be reduced to a numerical quantity.

    You’re not even comparing apples to oranges. You’re claiming that watermelons are 472% tastier than potatoes. Clearly, we are dealing with someone who has watched “Pi” way, way too many times.

    Or not. To be fair to you: your thinking is not original. This flavor of quasimystical numerology is absolutely par for the course in what passes for “economics” in the universities today. No wonder the country is circling the drain. Frankly, I’m surprised it’s lasted 75 years with madmen of this stripe at the wheel. It just shows what a great country America used to be.

    Dear John: spend a little while in Detroit and tell me how wonderfully your rubber dollars are working out. Alternatively, go to Google Books and find me a description of urban decay from 1908. Guess what? You won’t find one. Guess why? Because there wasn’t any.

    Here’s a good exercise: compare Jacob Riis’s How The Other Half Lives (1890) to Sudhir Venkatesh’s Gang Leader For A Day (2008). Then tell us about all the great things that a century of government check-kiting has done for the poor. Don’t forget this cute WPA poster.

  107. mencius writes:


    Here, instead of a deposit enabling a loan, the loan itself creates the deposit. The bank makes the loan with money it does not have, and credits the borrowers account with it (creating the deposit as a liability on its own books). It balances this liability with an asset, a receivable.

    I now understand the transaction you’re talking about. But I still think you’re wrong. The way in which you’re wrong is quite interesting, however.

    Again, we are looking at two transactions which are bundled into one. These are perfectly separable, and neither creates dollars ex nihilo. Nor does either require the bank to use dollars it doesn’t have.

    In the first subtransaction, bank B pays X dollars to citizen C, in exchange for C’s promise of X+y dollars at time T. This is the loan. Note that it does not expand B’s balance sheet. Nor does it create dollars.

    In the second subtransaction, citizen C deposits X dollars in bank B. Ie, he loans X dollars to B with a continuously-rolling zero term. This produces exactly the cumulative effect on the bank’s balance sheet that you describe. But it, too, does not involve the creation of dollars “ex nihilo.”

    Note that these subtransactions are perfectly separate. They are bundled only for the (dubious) convenience of B and C. Nothing prevents C from reversing the second transaction instantly, and storing his cash as cash, depositing it in bank K, etc, etc.

    And of course many loans – mortgages, for example – do not show the second transaction at all. If I buy a house with a mortgage from bank B, the cash goes to the seller. I do not get an account, even a temporary account, at bank B which contains the money I have borrowed.

    Note also that this double transaction is pointless – as you note – except for maturity transformation. MT is creating the illusion that banks can create money, by teleporting it from the future. If the illusion is backed by loan guarantees – formal or informal – the result is as you describe. If it is not so backed, we see what happened to the “shadow banking system.”

    But this is not a natural consequence of balance-sheet arithmetic. It is a Rube Goldberg machine which is fundamentally dependent on “the camouflaged hand.” The hand was well enough camouflaged that the shadow bankers forgot that it existed, and thought that MT was a safe operation in a free-market financial system.

    I agree that reserve requirements are an extremely gamable, and basically failed, way of controlling this mechanism. Note also that according to inflationist doxology, it is not supposed to need control. The money supply is supposed to expand automatically to meet the “needs of trade,” as the inflationists of a century used to argue.

    (The inflationists of today seldom argue – they can just assume. Or even just bloviate. Power hath its privileges.)

  108. mencius writes:

    Graziani (and others in the “Circuitist” tradition) reasoned that this would only be possible if the producer of bank notes—or the keeper of the electronic records of money—could not simply print them whenever he/she wanted a commodity, and go and buy that commodity with them. But at the same time, people involved in ordinary commerce had to accept the transfer of these intrinsically useless things in return for commodities.

    This is great! You’re actually quoting from people whom even Keynesians regard as monetary cranks. Tell me, are you a fan of C.H. Douglas (Mr. Social Credit, and Ezra Pound’s favorite) as well? How about Silvio Gesell? If you don’t know them, perhaps you should look them up. Sounds like just your alley.

    The entire early 20th-century movement toward “endogenous money,” which I’m delighted to see revive in the hands of 21st-century quacks, is rooted in pure hatred of savings and the rich, and the desire to confiscate the former from the latter.

    Lenin, a member of the same school, once said that the only use for gold in the workers’ paradise would be to plate the handles of the faucets in the workers’ bathrooms. Indeed the 20th century made great steps in the direction of eradicating money and saving. Fortunately, it did not quite succeed. But who knows – maybe this time it will. I’ll save a potato for your gulag, comrade.

  109. john c. halasz writes:

    Wow, Mencius! You really out-do yourself. You prove yourself an even bigger jerk, in your self-conceited hysterical paranoia, and adventitious, ad hominem piss-poor style of argument, which only indicates the severe limitations of your intellect in grasping any conceptual point, let alone any concatenated set of such concepts, i.e. any genuine theory. What would you ever do, if you were deprived of your fetish-object, “hard money”, and had to deal with the world on realistic terms, by acknowledging its fundamental non-existence/impossibility? Would your whole entirely virtual world collapse, and, with it, your grandiose self-conception?

    By what arrogance or divine right do you assume that I’m unfamiliar with “Moore’s law”, or that I am unaware of the effects of innovations in new processes and products. That the neo-classical price-index concept is flawed because there is nothing to say that a set of nominal prices will be the “same” in composition and “value” 30 years hence, as 30 years ago is a point I’ve made numerous times. (Though I utterly fail to see why that would support your unhealthy obsession with “currency dilution”, as if it were the root of all evil in this world, and as if the historical defects of intergenerational “justice” would be remedied by its reversal, and as if it would assure you of gaining your rightful place on the throne, or at least close enough as your worship of hierarchy would allow). But you’re really proposing to proceed utterly without data and uterly without quantification? A brilliant approach! Why didn’t anyone think of that before? Let alone without any mode of explanation of how we get from here to there, since obviously history is a narrative of sheer decline from the golden age, literally, and any illusion of “progress” involve theft from the present by the ever so decadent future!

    Now while I do tend to take statistical data with a grain of salt, as a rough approximation, let me offer an ad hoc interpretation of that chart I eye-balled. The changed rate of productivity had to do with a) the shift from a still primarily agrarian economy,- (in the case of the U.K. perhaps through imports),- to a more fully industrial one, and b) the “second industrial revolution” involving oil, electricity, chemicals, etc. Apparently you’ve never heard of or considered the problems of joint production, jointed techniques of production, differing capital ratios between sectors, or differing rates of change in productivity between sectors, nor, mirabile dictu, that technical conversions of capital stocks must occur at economic rates, which will be determined by a) their profitability to capitalists, and b) the sequential states and developmental dynamics of the overall economy.

    But probably your deeper defect is that you simply lack the capacity to grasp the principle of real effective aggregate demand, which you think, in your superstitious mind, must be a witch’s brew, concocted by that devil-worshipper Keynes, since you remain wedded to the notion that markets must be perfectly self-regulating due to nominal price-adjustments of supply-and-demand. But actually, there’s a fair argument that Michael Kalecki scooped Keynes, except that he wrote in Polish and a bit in French, so no one noticed, and I’m offering a more Kaleckian version. To be brief, there are lots of reasons why ready price adjustments don’t occur, not least is the monetary illusions that wage adjustments would be under, which Keynes’ emphasized, since efforts to lower wages to keep up with falling prices would further lower the price level, while still further reducing demand levels, etc. But still more fundamentally is the factor I focused on, which is that long-lived investments in capital stock would be gratuitously destroyed by such price declines and lack of aggregate demand. Though the Austrian school, with their Viennese indulgence in more-than-Spenglerian pessimism, since Vienna was the “laboratory of world destruction”, as Karl Kraus put it, were more than pleased to recommend such catastrophic liquidationism. And this based on their time-preference theory of interest, since economic “value”, or prices, if you repudiate the term, though “time-preference” was actually supposed to be an account and measurement of such, would supposedly be determined sheerly through subjective perceptions and their aggregation, without any objective factors or constraints. You sneered at “deconstruction”, though I’d guess your too dim-witted to have any understanding of what the term might mean, but “dismantling conceptual criticism” might be a fair paraphrase, and, if you had bothered to actually read what I posted, in however an infelicitous style, I offered precisely such a criticism of the Austrian “time theory of production”.

    You prattle on about “maturity transformation”, as if you’d discovered the Holy Grail, but the world is condemned through failure to heed your brilliant discovery. But, aside from the fact that you fail to adequately distinguish between the issues of illiquidity and insolvency, though, to be sure, the hinkiness of financial accounting can readily obfuscate that issue, it’s true that investments can always fail and result in net losses, and the further out into the future you project, the more likely outcomes will deviate from current expectations. However, any long-lived real productive capital investment is financed out of current production,- you can call that “savings” if you like, though it’s likely basically the “savings” from the profits of current owners of capital stocks, or you could call it “production switching”. But it means that “maturity transformation” is always an issue, regardless of the exact arrangements involved in a financial system, which, whatever its pretentions, is always just a superstructure inter-mediating real production and the revenue/income flows therefrom, whatever tendency it might have to out-run those flows. The core issue is the recycling of profits from real productive investment, which will effect both the supply of loanable funds and the level of real effective demand, and, if an adequate set of opportunities for further productive investments and/or an adequate redistribution of incomes to reflect productivity gains does not occur, then profits will tend to bleed into the inflation of financial “asset” prices, until a bust results. But none of that has to do with the “evils” of currency dilution, as if wealth were really a store of “hard money” to be distributed through noblesse oblige, (a notion that befits the illusions of a landed aristocracy), rather than deriving from claims on the ownership/output of real illiquid capital stocks. And you utterly failed to answer my question: why on earth would capitalists invest in increases in productivity and output, if it would only serve to dissipate their profits into deflated output prices in a “hard money” regime? (And as a further aside, rapid increases in productivity and quality in the IT sector,- and much recorded productivity growth during the tech bubble was self-referentially confined to the tech sector,- do not result in general productivity growth unless they bleed out into applications in other sectors, whereas, in a stagnant economy, they might just result in “Baumol’s cost desease”, whereby the gains accrue to other sectors through increases in their relative prices, a version of the worry that Ricardo classically expressed in his theory of rents).

    So nothing really has to do with any “hard money” standard. In fact, you’re seriously confused about the difference between base money and credit, and you think the two should somehow be the same, else there must be fraud involved and aborginal wealth is being confiscated. In fact, AFAICT, Winterspeak has been correct and clear in his colloquy, and I don’t see why he’s interested in attending to your picayune insistencies, as you just seem to be making concretistic errors. (If C takes out a mortgage to buy a house, then seller S, unless buying another house, deposits his money in a bank, which creates further reserves for l
    ending: that’s the “fractional” part of FRB). Now it’s not as if FRB is unproblematic. For one thing, it involves tight inter-linkages between banks, which can spread financial contagion. But the same sorts of inter-linkages occur between businesses in the real economy, except that they tend to be intermediated by banks, hence the problems in the real economy tend to first show up there. And for another, there is embedded leverage involved in the FRB system, except that such leverage would be involved in any system of credit, insofar as it involved credit and not just the transfer of hoards of “hard money”, converting increased investment into decreased consumption. Which is why the dangers of FRB require backstopping and tight regulation by governments. Austrian “banking” proposals would involve both such a low level of loans as the production cycle would be slow to develop and perhaps never get started, (which is what is meant by the endogeneity of credit), and be so capital inefficient as to be entirely unprofitable. If you’re recommending a public system of credit, I’d be willing to consider it. But if you want a market-based profit driven system of credit, then such proposals wouldn’t do. But they also would not bring about the illusion of static equilibrium accounts, whereby all market prices would adjust instantaneously and fluidly to clear all markets in a perfectly self-regulating process. Indeed, a perfectly unregulated, “free market” banking system would inevitably end up extending credit anyway, but would do so in ways that a) resulted in frequent bank bust and bank runs and b) would result in “market discipline” restricting credit to minimal levels, stifling the real economy, which is to say, not an optimal “free market”, but a permanent stagnation “equilibrium”, which seems pleasing to the Austrian mind-set, since the rich would remain rich and the poor would remain poor, except that they fail to grasp that, though the poor would suffer most in such a condition, the relative wealth of the rich would actually be lowered more. But Austrians, of whatever degree of “sophistication”, are always a broken record, repeating monomaniacally that its all do to an evil conspiracy of the government interfering in the perfect workings of the market, which bears a metaphysical identity with “freedom” itself, via the Fed and the debasement of the currency due to its back-stopping FRB. Blah, blah, blah. Analysis never proceeds any further or more deeply, because it’s the perfect righteousness of an invariant ideological answer for “everything” that’s at stake, not any genuine effort at understanding. Your adventitious, impressionistic, and evidence-free style of argument strikes me as an indication that you simply lack the capacity for clear consecutive conceptual thinking by which to connect any dots.

    As for the Lower East Side back then being free of street gangs and full of wholesome happy families, based on undoubted contemporary testimony, and all subsequent social decay being due to people having stopped snorting gold dust and thus started snorting crack, is something only one addicted to Austrian fairy dust could “believe”. And Graziani is a contemporary economist, not a crank. His first name is Augusto, if you want to google him, just so you don’t mix him up with some soccer player. And Post-Keynesian accounts of endogenous credit are not the same as nineteenth century accounts of endogenous money, though their certainly different from Monetarist or New Keynesian doctrines, which rely on macro-economic models that strangely lack any specification of money and credit. But it’s not a matter for “monetary cranks”. In fact, you’re the one who’s a monetary crank, though the mystifications surrounding money as a symbolic “universal equivalent” have thrown up a wide variety of such cranks. The way that you just randomly through up names of such,- (really, Ezra Pound!), is just a further indication of the quality of your style of and capacity for argument.

    I also don’t mind that I’m thoroughly unoriginal. If I weren’t, then that would indicate that I hadn’t learned anything, since, while truth is not identical to consensus, truths of any sufficient generality will tend toward consensus, at least locally and historically. But then you’re utterly unoriginal yourself, endlessly mouthing stale discredited dogma. And the idea that money is a “stationary bubble” has long been standard in academic economics, though it comes literally from the physical of fluid thermodynamics, with a well-defined mathematics. It’s more turbulent fluid dynamics that raises the mathematical difficulties. I also don’t mind being compared to a pig. They are quite intelligent animals actually, and not at all uncleanly, except that their human owners keep them so. And I certainly don’t mind rooting about in the mud for some earthly understanding. Pigs also have an excellent sense of smell, by which they can discriminate between a fine truffle and a steaming pile of sh*t.

  110. winterspeak writes:

    Mencius: I think we’re making progress.

    1. You now understand the transaction I am talking about. And it’s clear it has nothing to do with FRB, no?

    2. You also agree that it’s this transaction that drives dilution, not Government deficit spending per se (although that is one instance of it, and the ties between banks and Govt are extremely close).

    All that remains is whether this transaction is an MT transaction or not.

    I argue that it is not. Your “it’s MT” claim rests on “In the second subtransaction, citizen C deposits X dollars in bank B. Ie, he loans X dollars to B with a continuously-rolling zero term.”

    It doesn’t have to. Suppose he adds a maturity-match clause to his deposit, essentially making the deposit a CD. The loan matures in 5 years, and the deposit matures in five years. We can take MT out of it and balance sheets still expand.

    Same thing happens with a mortgage. The bank creates the loan and credits the seller, thus creating a deposit. The bank has the receivable as an asset, and the deposit as the liability. The borrower has the house as an asset, and the payable as a liability. The seller has the deposit as an asset, but has lost the house. No change on his balance sheet.

    This 3 party transaction then has the bank creating a deposit by making a loan (as before) and there is an asset swamp amongst the seller and the buyer.

  111. i’m delighted, and very flattered, to host very smart, eloquent, and high-spirited commenters.

    and in the end you can speak as you please, i don’t “moderate” anything.

    but i do have preferences, and i would prefer that some of the high-spiritedness be expressed a bit more civilly. if one must be insulting, tone can be more effective than direct name-calling, and I find it more satisfying from an aesthetic standpoint.

  112. JKH writes:

    An earlier neglect of an important point:

    Mencius said (2/25/09):

    “Of course, we can (at least logically) draw an arbitrary line between any two sets of economic entities, consolidate the balance sheet of each set, and observe the flow between them. For example, we can consolidate the balance sheets of all the households whose head has a last name beginning with A-M, the same for N-Z, and observe the monetary flow between them, just as we can observe the monetary flow between the US and China. If we think of Winterspeak’s statement in these terms, he defines “saving” as lending by private citizens to USG. This equals borrowing by USG from private citizens, and (G – t) must indeed exceed 0. Thus, for the private sector to “save,” the public must borrow. But (as I’m sure WS would agree) this is an accounting identity, not an observation.”

    This is true. The liability of X must be an asset of (not X).

    Be careful and precise when creating theories from the equivalence of logical complements.

  113. winterspeak writes:

    JKH: I’m sorry, I did not understand your post.

    Were you agreeing with Mencius, or not?

    I take my statement to be both an accounting identity and also true. If we had a sovereign monopolist gold miner, the sovereign would have to take some gold out of his mine, and exchange it with non-sovereign sector for something. Let’s call this G. There sovereign also has the option to demand some of his gold back, let’s call that T. For the non-sovereign sector to have any gold to trade back and forth, the sovereign would have to spend more gold than he took back, let’s call that G-T>0, and yes, it is a deficit.

    But in no way is the sovereign “borrowing” money from the non-sovereign sector — quite the opposite, he is funding it. In a fiat system, it simply makes no sense to say that the monopoly currency issuer is “borrowing” from currency users.

  114. JKH writes:


    The point on which I was agreeing with Mencius was precisely the point that I made:

    The liability of X is an asset of (not X).

    This is true whether X is the US government, Warren Buffet, China, Brad Pitt, or UBS.

    It’s a somewhat broader point than setting X equal to the government and specifying what follows in that particular case.

  115. winterspeak writes:

    “The liability of X is an asset of (not X).”

    Did I ever say (or imply) anything different?

    I understood Mencius as claiming that there are assets out there that are liabilities to no one. This is certainly the standard concept of money in a gold standard world.

    So you and I are square (at least on this point)?

  116. JKH writes:


    No, you certainly didn’t imply anything different.

    I just liked the first four lines quoted from Mencius, which is what I’ve paraphrased, but could have left out the last four.

    I think we’re square. There’s no disagreement here – just an admission of the possibility that there’s a more general perspective on the relationship between a liability and a corresponding asset.

  117. mencius writes:


    Yes, we are looking at the same transaction. I feel we are quite close.

    My main point about this transaction is that it’s actually a sequence of two simpler subtransactions, neither of which violates the conservation of money. If money is not created or destroyed in either subtransaction, it cannot be created or destroyed in the sequence.

    And note the extremely suspicious nature of the sequence. A loans money to B – and B loans it right back to A! Does this smell like good accounting to you? Is there actually a legitimate economic purpose here?

    MT is indeed crucial to the existence of the structure. Think about it: if the term of A’s loan to B is the same as the term of B’s loan to A, the two cancel perfectly and there is no point at all in the transaction.

    But if A can back present-money obligations with future money, it can effectively engage in Mosler-style monetary creation. As long as it has the FDIC to supply the informal guarantees that make B willing to loan the money back to A, making A’s deposit obligation to B (which is not money, but a private promise of money) pegged to the dollar.

    Through “borrowing from themselves” with this transaction, banks can increase their capitalization, without actually having to raise capital from actual lenders. Thanks to the FDIC’s free option, they are in a sense borrowing from the Fed – but the transaction is invisible and informal. This, again, is the precise opposite of good accounting.

  118. JKH writes:


    On second thought, and for completeness and full disclosure, here is something I wrote recently at another site, in response to one commenter’s apparently stiff allegiance to the Mosler paradigm. As you have become a major on-shore spokesperson for “the paradigm”, I’d be interested in your thoughts on my assessment:

    “My impression of it is that it’s a very insightful and detailed look at the operational guts of the banking system with particular emphasis on how government budget transactions can affect banking system reserves.

    But in my opinion, Mosler goes too far in extrapolating this knowledge to a larger monetary theory.

    As just one example, he says that government budgetary expenditures essentially create the banking system reserves that are required to support tax payments or bond purchases. In that sense, expenditures fund revenues. The essence of the idea is that government expenditures have the potential to create an overdraft position in the government’s account at the central bank. An overdraft is essentially an extension of credit. If that overdraft is left uncovered, the government has funded its expenditures by printing money, just as does the central bank in its normal monetization operations. The government in this sense can be viewed as an extension of its own central bank, and taxation and debt issuance can be viewed as extensions of monetary sterilization.

    This is true as an explanation of how the system can work. But Mosler seems to extend this to the notion that the system does work this way, in the sense that expenditures tautologically precede taxes or debt financing.

    That’s wrong, in my opinion. The system can work this way. But there’s also nothing to preclude the operational fact that governments can collect taxes or issue bonds (thereby draining bank reserves) before making corresponding expenditures (thereby replenishing bank reserves).

    The result is a grand Mosler theory which he calls being “in paradigm”. Those who don’t understand it are branded as “out of paradigm”. There are other deeper and impressive ideas associated with this theory which admittedly I haven’t had the time to explore yet.

    Nevertheless, a great contribution is his deep understanding of the operational details of the banking system that is very instructive in linking it with economic theory more generally.



    “JKH, you lost me here:

    “But Mosler seems to extend this to the notion that the system does work this way, in the sense that expenditures tautologically precede taxes or debt financing. That’s wrong, in my opinion. The system can work this way. But there’s also nothing to preclude the operational fact that governments can collect taxes or issue bonds (thereby draining bank reserves) before making corresponding expenditures (thereby replenishing bank reserves).”

    How can the monopoly issuer of a token not issue it before it collects it? Can an airline collect tickets at the gate without selling them beforehand? Can a subway collect tokens at the turnstiles without issueing them? Sure, the fact that ther are savings sloshing around in the system means there’s not necessarily a one-for-one relationship between issuance and redemption, but the logic is inescapable. I’m curious how you think it is possible.”)


    The private sector pays taxes via debiting of their bank accounts and crediting of government account(s). The government ultimately gathers those funds at a central point, that being its account at the central bank. This crediting of the government account is mirrored in the debiting of bank reserve accounts by the same total amount. This is the process whereby taxation at the margin drains reserves from the system.

    All that is required for this to happen independently of government expenditure is the existence of a sufficient outstanding stock of M1 type balances.

    The government is a monopoly issuer of currency and reserve balances. It is not a monopoly issuer of M1 balances. The commercial banking system issues M1 balances, and as Mosler would agree, is not constrained by central bank reserve requirements in doing so. So the M1 balances can be made available for people to pay their taxes before the expenditures that offset those taxes need occur. The system can work this way.

    On a cumulative basis, it has worked this way over certain non-trivial periods, in the examples of Clinton and Canada that I have previously noted (i.e. budget surpluses).

    It would in theory be possible for a government to collect taxes, and still be in budget balance from the beginning of its existence, with zero outstanding debt. Moreover, central bank liabilities could be invested in private sector assets. The currency that forces its own acceptance through the power of taxation would still be outstanding as it is now.

    All I’ve said is that taxation can precede expenditure in arriving at surpluses, balances, or deficits. I’ve not said that currency is not outstanding or hasn’t been issued.


    “How can the monopoly issuer of a token not issue it before it collects it? Can an airline collect tickets at the gate without selling them beforehand? Can a subway collect tokens at the turnstiles without issuing them?”

    I don’t see this as necessarily proving Mosler’s base case. Mosler’s argument seems to be that expenditure precedes taxation at an operational level. Governments spend, which increases bank reserves. Reserves are then drained by debiting the banks whose customers are paying the taxes. It can happen this way, but it doesn’t have to.

    Assume the airline is analogous to the government. Suppose expenditure precedes “taxation” in this case. The airline spends to pay for the cost of a flight (fuel, labour costs, etc.). It pays by issuing its “monopoly currency” – in this case airline tickets. The system then has excess “reserves” in the form of airline tickets. Supplier/customers of the airline pay their “taxes” to the airline by presenting their tickets as payment for the flight. Reserves are drained.

    It can happen this way. But it doesn’t have to happen this way. The order in which the airline spends and the customers redeem is operationally irrelevant. The airline may pay for its fuel bill or its labour costs before or after the flight. It may run a temporary deficit before the redemption of the tickets or a temporary surplus after the redemption. It is not constrained in any way by the notion that expenditure must precede taxation.

    The fact that the airline has issued its currency before the currency is redeemed is also irrelevant. This only means that the airline is running a gross surplus with respect to the lead time between the issuance of its currency and the time of the flight or redemption of that currency. In fact, it argues against the Mosler paradigm as a marginal effect, because the result taken on its own is a taxation surplus. But what must be compared against this gross effect is the other side of the transaction, which is the airline expenditure. That can happen with any timing – before the issuance of the currency (strong Mosler), between the issuance of the currency and the redemption of the currency (weaker Mosler), or after the redemption of the currency (non-Mosler). The net effect will determine whether the time distribution is one of deficit followed by a netting surplus, surplus followed by a netting deficit, or “simultaneously” timed balance.”

  119. john c. halasz writes:

    Let’s supposed that there is a fixed supply of money-stock, whether gold or any other medium, corresponding somehow to a fixed or constant supply of resources produced and exchanged, such that any exchange of transfer of a fixed amount of resources would exactly correspond to the exchange of a fixed amount of money. This would be a world in which Say’s Law would be immutably true, in which a system of monetary circulation and exchange would be a transparent reflection of an industrial barter system. Savings would be defined as deferred consumption, whether such savings were hoarded through being withdrawn from circulation, or lent, and lending would involve ego forswearing current consumption for claims on future consumption, in exchange for alter currently spending the money, whether on current consumption or investment in production, in exchange for alter supplying ego’s claim on future consumption. Now let’s suppose that productive investment is made by alter, to build productive capacity for future output and consumption, that would increase technical rates of productivity and thus increase output and lower prices, i.e. would produce more unit output per the same unit input of resources. However, given fixed resources and fixed corresponding money-stock, this would involve production-switching, as productive goods were withdrawn from meeting current consumption demand to build future production capacity. Hence, there would be a lower supply of production to meet current consumption demand, but, note, the work force building the future production capacity would still be paid out of current output. Hence a gap between current production supply and current consumption demand opens up, which can be meet either through inflating prices or deflating wages, but, either way, involves lowered current aggregate demand.

    Hence there’s a dilemma here, maybe even something approaching a paradox: the re-entry in the future of increased productive capacity would occur in a lowered aggregate demand condition, which it’s own production would have occasioned. It’s true,- even ignoring that the new capital stocks competitively would drive out older stocks and the costs incurred thereby of recycling the remaining value of bankrupt businesses and re-employing unemployed laborers-, that lowered output costs would nonetheless imply higher real wages and thus some increase in overall consumption demand and productive output. But the gains to output would be limited by the constraint of such lowered demand and thus the incentive to productivity-improving investment would be diminished. The “solution” that bridges this dilemma is credit, which is not primarily a matter of “teleporting” future money into the present, thereby creating “artificial” demand, but rather of “teleporting” current productive capacity into increased productive output in the future. It’s true that not all extensions of credit will work out and be repaid, not least because older capital stocks will be competitively driven out and result in bankruptcies, and also because the sharing of risks and uncertainties, as well as gains, through credit arrangements will not map, perfectly somehow, onto future prospects and outcomes. That’s what interest rates and term curves are all about, and it’s a matter not of avoiding any losses, but whether the gains out-weigh the losses.

    It’s also true that the extension of credit can outrun the availability of real productive possibilities, and, given the unavoidability of production/business cycles, rooted in the limits of available, unexploited technical possibilities and the need to convert capital stocks at economic rates that allow for their realization before depreciation, credit crises will periodically occur. But that’s why any system of credit,- (and any system of lending that gets beyond the concretistic account and its dilemma, with which I began, will be a credit system with embedded leverage, in which risks and rewards are shared between the operating leverage of productive enterprises and the financial leverage employed by aggregators of savings/underwriters of credit),- needs to be tightly regulated and supervised, to prevent their real dangers from getting out of hand. A completely “free market” system of credit would tend all the more to run out-of-hand, and be all the more disaster prone, or else, would be so limited and constrained as to utterly fail to get production cycles started. (It’s also ironic, given the Austrian’s emphasis on the embeddedness and dispersion of information, at the root of markets, that they fail to notice that their banking proposals would drastically increase the cost and lower the efficiency of information-gathering, so that such banking would both be too unprofitable to survive in a “free” market, and too unreliable for ordinary savers and tradesmen). Again, our current debacle is not due per se to government “intervention” and regulation in the banking/finance system, but rather to the failure of such under the auspices of “free market” ideology. (And, though an analytic distinction can be made between policy and market failures, there is not always a clear line to be drawn between the two, especially if there is a failure to recognize that market failures are potentially and pervasively endemic, not just the result of intervening policy failures, and policy failures are frequently the result of their entanglement with market failures, through regulatory capture and rent-seeking by “private” interests).

    It’s also not true that credit extension amounts to an increase in the money supply, or, at least, that’s an inverted understanding. It is true that credit-implements can be more-or-less readily exchanged for currency, (as with, e.g., secondary markets for bonds), but credit actually always out-runs and exceeds the circulating money supply. Indeed, it would be hard to see how there could be readily available opportunities for ordinary savings, if it didn’t, as then savings would have to seek out specific opportunities for investment,- (and as for the costs involved in that, just think of the whole largely useless “financial adviser” industry today)-, or would amount to a hoarding of $, withdrawing currency from circulation and decreasing its needed supply. What’s important here is that savings be recirculated into productive investment without impairing adequate demand. But, just as well, credit supply can not be controlled by controlling currency supply, since currency supply tends to follow after credit supply, (as the failure of Monetarist policies testify). Which, again, is why credit “creation” and finance need to be subject to tight regulation, to prevent the build-up of excess leverage that out-runs actual and potential real output, the revenues and incomes produced by sustainable realizations in the real economy, through corporate profits and wage incomes, which excess credit results in financial “asset” inflation and compounded interest, beyond any reasonable future prospect that such financial claims could be fulfilled, which inevitably results in bust and general default. (Though one should not ignore the income distribution variable, which if skewed upward, results in an excess of “savings” over real investment at the top, lagging effective demand in the middle and increasing debt at the bottom, which also results in financial “asset” inflation and bust). However, even with a well-regulated financial system, periodic downturns will occur, adjusting to failed realizations and mistaken expectations, and the public debt will increase, both automatically and at discretion, to compensate for the liquidation of “private” debt, providing the “liquidity” that is missing to over-extended private credit. (The point I tried to make earlier is that the tendency in developed industrial economies to incur increasing public debt is due to the the need to compensate for production/business cycles being accumulated across series of such cycles, since government will always need to correct for market failures, without there being a nee
    d for government budgets to be balanced across a given cycle, rather than debt maintained below the level of real growth. There are other factors involved, such as demographics, though government-mediated transfers at low transaction costs are not necessarily public debt, unless growth fails to be sustained at rates projected in budgets).

    But none of the above would be apparent from a neo-classical marginalist perspective, which reduces production dynamics to market-exchanges. If there is a highly inefficient distribution of resources, then a set of “Pareto-improving” exchanges could readily occur that would move toward a “Pareto-optimal” distribution, increasing overall economic efficiency. But once such a set of markets is in place, such improvements become increasingly “scarce” and markets would tend toward constant re-production/stasis. (Which is what Robert Solow “discovered” by running then reigning GE models and finding they could scarcely account for recorded economic growth). And analysis of the sources of productive surpluses and their distribution, as in classical and Post-Keynesian economics, is far more to the point, not least because by not reducing economic “value” to an account of nominal price-formation, it does not conflate money and financial markets with goods markets, and thereby fail to recognize the endemic tendency for the two levels to diverge. But I also think it doesn’t “pay” to focus sheerly on the monetary and financial level of Post-Keynesian accounts and fail to understand the linkages to Post-Keynesian accounts of production in the real economy. It’s correct, as far as it goes, to say that sovereign governments at once provide the currency that pays the taxes it collects and provides a compensating stock of public debt to facilitate private savings. (In fact, Australia provides an interesting empirical case in point, as it,- though this is about to change obviously, by one account has no public debt, the 20% of GDP debt after the last recessionary crisis having been entirely paid down by Howard and Costello, that notable comedy team, but, by another account, it provides a stock of debt at 15% of GDP precisely to provide “risk-free” implements to the private sector. On the other hand, Australia’s external debt is +100% of GDP). The government system is entangled with the economic system, to be sure, if only through the entwined issues of the fiscal budget and taxation. But it’s bootless to deny that the government depends on resources produced elsewhere in the real productive economy and draws those resources off of it, whereby, if it “produces” the means of circulation, money, and commands resources thereby, through taxation, there is a significant difference between those two “moments”. That would be especially obvious in the case where the government consists largely just of the sovereign and his hangers-on, with taxes derived from expropriated agricultural surpluses, and “public” debt accrued for purposes of sovereign warfare, and raised from merchant-bankers, who derived their fortunes from trade-credit, as in the good old days at Versailles, which case is the actual origins of the modern state, from early modern absolutism, (“lo stato” originally referring to the prince’s estate, which property he nonetheless had ostensible feudal obligations to protect and provide for).At least, in such cases, the pretension that the sovereign provides public debt in the public interest are, er, severely limited. Hence the government and its finances are at least as dependent on the “commonwealth” and the circulation and output that occurs therein, as the latter is dependent on the sovereign provision currency and fiscal/financial regulation.

    When NIPA accounts were devised, the accountants decided that there was no way to account for government expenditures in terms of output, as to whether any of it counted as investment or production, (though obviously some of it does, whether directly by government agencies or through government contractors), and whether government workers achieved any specifiable level of productivity, so government expenditure G was counted as sheer consumption and revenue T as undifferentiated taxation, (i.e. regardless of its sources or incidences or whether they involved any implicit subsidies, etc.) And, indeed, since governments don’t run a “profit” and their budgeting processes don’t distinguish between specific expenditures and outputs, such a convention seems well-nigh inevitable. But it is just a convention and all modern governments, regardless of how “democratic” they are, are deeply involved in variously regulating, supporting, subsidizing, and otherwise sustaining the real productive economy, for reasons of their own fiscal and financial sustenance, which have some unspecified limit, beyond which they can not compensate for the failings of the real economy, (which limits, of course, will vary, depending on the position of the country and its currency in the international pecking-order). Hence an account of macro-economic aggregates and gross fiscal dynamics will not suffice. A decomposition of government budgets and the policies and activities they sustain is as essential as an analysis dependencies of the “private” economy on government supports and expenditures, whether as sustaining a general “public interest” or as private capturing of rents, just as the “private” economy is susceptible to inter-sectoral analysis of the composition of outputs and activities. In short, a simple “macro” level of analysis, assuming a clear-cut distinction between public and private won’t do, not because “micro foundations” are missing from the macro-economic analysis, but because a more “meso” level of institutional analysis is required to understand the actual workings and dynamics of modern “market” economies. That might then permit a more general discussion of what ends are in the “public interest”, and what should be allowed to it.

  120. winterspeak writes:


    “In the first subtransaction, bank B pays X dollars to citizen C, in exchange for C’s promise of X+y dollars at time T. This is the loan. Note that it does not expand B’s balance sheet. Nor does it create dollars.”

    This does increase B’s balance sheet, because B does not draw down a cash balance to make the loan. It just creates a receivable for X dollars as an asset, and the deposit (ultimately) as a liability.

    In a fixed balance sheet system, B would have to have the money (in cash, or whatever) before it could loan it out.

    “In the second subtransaction, citizen C deposits X dollars in bank B. Ie, he loans X dollars to B with a continuously-rolling zero term. This produces exactly the cumulative effect on the bank’s balance sheet that you describe. But it, too, does not involve the creation of dollars “ex nihilo.””

    We spoke about how MT does not help here. C’s balance sheet is bigger too — he has his deposit as an asset, and a payable as a liability.

    The point is that B did not have the money on hand that it lent to C. B created the money by making the loan. In a fixed money world, deposits enable loans. In our fiat world, loans create deposits.

    JKH: I enjoy pushing the paradigm to see how far it goes. It’s novel to me, and I think its core mechanisms, in actual central bank operations, is a better place to start than most. That said, there are parts of it I think are insane. At my roots, I remain a Chicago boy (for now).

    You raise good issues about whether spending needs to precede taxation or not. I spent some time looking into how actual taxes make their way to the Reserve bank, and it’s ugly and complicated, requiring deep coordination between the Fed and the CB. I completely agree that the US could run on such a horrendous system.

    Commercial banks could issue M1 for the private sector to then pay in taxes, before the Govt deficit spends. This would have the private sector in net dissavings though, for the Govt to run surpluses. We’ve had a discussion about this before, and I agree that it’s certainly possible, it just seems more natural to me for the situation to be reversed. I have a few (I think) good reasons, but no great one, so I could certainly be wrong. It also does not seem sustainable, as ultimately Federal surpluses would drain private savings away entirely. Commercial banks cannot keep creating M1 the way a Sovereign can keep creating M0. (My terminology may be wrong here, but you understand my point).

    One final thought about Canada — and implications for the paradigm. The paradigm is focused entirely on nominal goods, not real goods (Mosler would agree), and the introduction of required commodities interacts with it in interesting ways. Mosler focuses on Saudi, as it sets the price for oil, but Canada is a major commodity exporting country as well, and while the Canadian dollar is it’s own fiat currency, I would not be surprised if, in practice, Canada runs a mixed economy with a good number of US$ being in circulation there also. In many ways, the US is unique for being as in paradigm as it is, and I think this actually makes it *harder* for people here to think about currency.

    At any rate, the paradigm identity is Y-I-C-T = G-T + (X-M). If you’re a major commodity exporter, and have large enough trade surplus, then you could run a budget surplus and still have the money you need to fund private savings. Mosler would say this is a bad deal because you are exchanging real goods for tokens, so the terms of trade work against the exporter, but this is a point that I’m ambivalent about.

  121. winterspeak writes:

    JKH: Something just struck me. Could bank created M1 be used for anything other than consumption and investment?

  122. JKH writes:

    WS – not sure I know where you’re going with that question. But M1 for given periods of time won’t necessarily be “used” at all – i.e. periods for which the velocity is zero. E.g. a business borrows from a bank just to hold M1 balances as an asset for precautionary liquidity reasons. On the other hand, “use” may depend on whether you’re viewing M1 as an asset or as a means of payment.

  123. JKH writes:


    Completely off-topic, but here is what looks to be an interesting paper on the credit crisis by John Taylor.


    I haven’t actually read it in full yet, but it caught my eye because I have an extraordinarily deep and visceral hatred for the “global savings glut” theory, which he destroys in several sentences.

    I think I know, and suspect I would agree, but what would the “paradigm” have to say about the GSG theory?

  124. winterspeak writes:

    JKH: Re: M1 — I was wondering if it could be *created* for the purposes of saving (not spending). Certainly, once it exists as a deposit it could change velocity, but I’m wondering if it could be created for zero velocity use. G-T can be created for zero velocity use.

    Will check out John Taylor.

  125. JKH writes:


    “Re: M1 — I was wondering if it could be *created* for the purposes of saving (not spending).”

    I can give you an example of where this happens, sort of.

    The Chinese Central Bank buys dollars from exporters and “prints” RMB M1 in doing so.

    To the degree that dollars earned by exporters associated with China’s current account surplus, and to the degree that the CA surplus is a component of Chinese savings, M1 has been created for the purpose of saving.

    This association isn’t perfect. The fact is that saving as defined according to national accounts is a flow and M1 is a stock. The M1 in this example is simply an intermediation channel for saving. Also, the mapping is not precise from export dollars to PBOC dollar purchases to PBOC RMB monetization to current account surplus to domestic saving.

  126. mencius writes:


    Me: “In the first subtransaction, bank B pays X dollars to citizen C, in exchange for C’s promise of X+y dollars at time T. This is the loan. Note that it does not expand B’s balance sheet. Nor does it create dollars.”

    This does increase B’s balance sheet, because B does not draw down a cash balance to make the loan. It just creates a receivable for X dollars as an asset, and the deposit (ultimately) as a liability.

    My whole point is still going right past you. But I now think this is my fault, rather than yours. Let me try a slightly different spin and you may say “aha.”

    My point is that your (Moslerian) financial system is a special case of mine. In your system, the above transaction is a fundamental, atomic operation – an axiom. In my system, the above transaction is a composite of two operations – a theorem, as it were. (Another metaphor: CISC versus RISC.)

    In both systems, exactly the same transaction is possible. But since my system has fewer axiomatic operations (ie, special cases), I find it a clearer and simpler way to model the process of banking. Does this compute?

    In your financial system, banks have special Moslerian powers with relationship to the currency. They can create money through the fiat mechanism, effectively borrowing it from theselves. Ordinary individuals and corporations do not have this power, n’est ce pas?

    In my financial system, the only special power banks have is the power of insured maturity transformation. (Which is quite a power!) Ie, a fiat issuer provides depositors with free, and risk-free, put options to insure that their deposits (which are not money, but a loan to the bank) always trade at par. This does not make the bank a fiat issuer – just a distributor. Only the Fed can mint dollars, so only the Fed can mint risk-free deposit options.

    Other than this, a bank (in my system) has the same power as any individual or corporation. It can sign contracts and make exchanges. It cannot create dollars, any more than on a gold standard it can create gold.

    Now, let’s look at the transaction again. In your version of the transaction, the bank matches the borrower’s receivable with the borrower’s deposit, appearing to create money ex nihilo.

    From my perspective, your picture of this event is like a blurry picture of a double star, taken with a lower-resolution telescope. Through my telescope, your Moslerian transaction resolves itself into two independent, orthogonal, non-Moslerian and money-conserving transactions. Yet we are looking at exactly the same astronomical object.

    The bank dips into its vault and lends the borrower money – actual cash – with a term of 30 years. The borrower puts this bundle of Benjamins in a suitcase, walks across the bank lobby from the loan office to the teller window, and deposits it with a rolling term of 0.

    Same exact result. The bank expands its balance sheet. So does the borrower. The bank has exactly the same amount of cash in the vault at the end of the day that it started with.

    Nonetheless, two simple things have happened, rather than one complex thing. And both these things obey the rules of my “fixed world” (which can, except for the minor problem of issuing risk-free guarantees in gold, work with any currency – paper, metallic or electronic, closed or open loop).

    Indeed, banks were performing your Moslerian transaction for literally centuries before anyone thought it would be a great idea to go off the gold standard. Because it depends on maturity transformation (it is pointless for the borrower to lend the money back to the bank at the same term he borrowed it for!), any banking system that practices it is diluting gold with paper, a dangerous practice. Nonetheless, this practice exactly is the hallmark of the 19th-century Anglo-American banking regime.

  127. winterspeak writes:

    Mencius: I see what you mean (I think) but I still don’t think we’re there.

    “The bank dips into its vault and lends the borrower money – actual cash – with a term of 30 years. The borrower puts this bundle of Benjamins in a suitcase, walks across the bank lobby from the loan office to the teller window, and deposits it with a rolling term of 0.”

    Nope. A bank can start with zero money in the vault and still do the transaction. The borrower can begin with zero money, and still do the transaction. Balance sheets really do get larger. And it can all happen with maturities being matched.

    As before, you have a seller, buyer, and bank (for a house). The seller transforms his house into a 30 year CD deposit — balance sheet stays the same. The buyer gets a loan (30 year, installment) as a liability, and the payable as a liability. The bank gets the deposit as a liability (30 year, matched) and the $100 receivable as the asset. Both the buyer and the bank have larger balance sheets. The deposit is matched to the loan. The payable and the liabilities are matched too.

  128. winterspeak writes:

    JKH: re: Taylor. It was a fine paper, not sure what it said that was new. I don’t think anyone buys the “savings glut” explanation any more, or has done for at least a year. I am more sympathetic than most with his claim that Lehman was not the cause of the financial collapse. Banks are pro-cyclical, if the economy was doing well the banks would do well, and I don’t believe giving banks lots of money will fix an economy that is deleveraging. I do believe that helping an economy delever out of fiscal expansion instead of AD preferable, and would do a lot to help banks.

    I also see your point with China. wrt to US$, the Chinese public sector is currency user, same as everyone else. It’s only a current issuer wrt to RMB. Net US$ inflows are CAS, some flow to the Govt, and some are probably kept in private accounts. Most economies are mixed currency economies, where a fair amount of activity happens in non-sovereign fiat currencies. I think it’s actually easier to understand the paradigm in these situations.

    It’s harder to pry apart Chinese banks with the CCB than it is US banks with the Fed/Treasury. I actually approve of the CCB telling banks to make loans (that will not be paid off) — at least it’s a transfer!

  129. winterspeak writes:

    Mencius: Good god–let me try again without the typos!

    “As before, you have a seller, buyer, and bank (say for a house). The seller transforms his house (asset) into a 30 year CD deposit (asset) — his balance sheet stays the same. The buyer gets a payable (loan, 30 year, installment) as a liability, and the house as an asset. The bank gets the seller’s deposit as a liability (30 year, matched) and the buyer’s loan as a receivable (asset). Both the buyer and the bank have larger balance sheets. The deposit is maturity matched to the loan. The payable and the receivable are matched too. Have I missed something?

  130. JKH writes:


    Brad Setser remains a steadfast believer in the global savings glut thesis.

    And see this from Krugman just the other day:



    Rather disappointing.

  131. JKH writes:

    One could visualize a system where banks keep exactly the same type of currency “on deposit” at the Fed that they do in their own vaults. Each day’s clearing would consist of deliveries of currency in the amounts of the bulk bilateral settlements between banks for their share of the net clearing. The result would be the same as if there were a 100 per cent (currency) reserve system, except in those cases where short banks needed to borrow currency from the Fed. Given that the Fed “prints” reserves in the same way it “prints” currency, there really is no effective distinction between this pretend system and the system as it is, at least in terms of the net result of the clearing of reserves between banks.

  132. john c. halasz writes:

    The problem with Mosler,- or the sort of monetary American Post-Keynesianism that he espouses,- is that it fails to take into account the US$ as a global reserve currency, which is where the issue of “fiat currency” becomes problematic, i.e. in the balancing of global trade. (Alas, as Stephen Roach put it, the US$ remains the tallest pygmy amongst currencies). In fact, the U.S.A. has been running a dollar-kiting scheme, which was first recognized as the “Treffen dilemma”, but probably should be traced back to the “beginning”, with the U.S. rejection of the “Bancor” proposal and the rejection of the 1948 Havanna Charter. It’s that lack of a currency/trade adjustment mechanism, leading to the absurdity of “competitive” devaluations in a fiat world, (and the corresponding lack of any genuine domestic policy control), that renders the Moslerian vision of maintaining aggregate demand through control of the issuance of the currency blind, since it appeals to the very mechanism by which the Wall St./MNC/Pentagon triad has undermined the very aim which it espouses, (since the gap between US$ and PPP terms has allowed the sorts of rent-extractions from the RoW that undermines any domestic “compact”, except insofar as the “natives” can be paid-off and bribed by a share of such rents, or, insofar as it allows “mercantilist” free-riding).

    So there is no “global savings glut”, though there is plenty of de facto forced savings around the world, but rather it’s tantamount to a global investment drought, operating through a declining wage-share of global output, and therefore a shortfall in aggregate demand, except insofar as it is propped up through increased debt-financed consumption, recycling excess profit-shares into financial “assets”. That doesn’t contradict the notion that the Fed deliberately reflated the U.S. economy by blowing a housing bubble through excessively low interest rates, but rather the former account is insufficient, insofar as it ignores the role of the CA deficit in shifting from tradeable manufactures to the non-tradeable housing sector, (which pattern is to be found through-out the world with CA deficits). And insofar as it ignores the deliberate favoritism of such a policy toward the financial sector, which renders it hardly “accidental”. But then again, that pattern of the decline of U.S. manufactures long pre-dated the current cycle. Krugman is only behind the curve.

    As to the Taylor paper, which I read in full, I’ll outsource my comment to Bruce Wilder, a sharp perennial commenter at Mark Thoma’s site:

    “” . . . letting Lehman fail . . .”

    This is a good example of how narrative is not analysis.

    Narrative requires post hoc, propter hoc sequences, with the hero (or antihero) confronting challenges, striving, and, by her own actions, winning or losing.

    Taylor is a classic libertarian, engaged in the libertarian’s pastime of blaming everything on government’s shortcomings and failings. Others want AIG’s coincident failure and rescue to not have an ill effect. That wouldn’t be much a narrative, would it? “Gov’t announces giant bailout of AIG; LIBOR spikes.” But, it makes as much sense.

    The narrative of how Lehman’s failure triggered a financial crisis is still, at base, a signal argument — that is, an argument that Lehman’s failure was a signal to the market, not that its failure — massive as it was — caused substantive harm to the system.

    As far as I know, Lehman’s failure had some, but limited effects on other institutions holding Lehman paper. I think one money market fund busted a buck. But, even that was more a demonstration of what could happen more broadly — it was not, in itself, a broad, substantive effect on many money market funds. In other words, it was a signal that other banks and institutions might be in big trouble, too.

    An argument that Lehman’s failure was a signal is not necessarily wrong, but it is a much weaker criticism of government policy than would be a Lehman failure that had substantive effects throughout the system. In particular, it is harder to see how simply “rescuing” Lehman would have prevented the signal.

    AIG’s “rescue” was a pretty strong signal, too, after all. Because of the “rescue”, it did not have substantive effects — banks holding AIG CDS got paid, and paid billions. But, it was signal that lots of CDS were both in the money, and going bad.

    If Lehman’s failure was a signal, it was a signal of general weakness across the industry — not a substantive event that created weakness, but a signal that revealed and confirmed weakness already suspected.

    Given that a “rescue” could only have supplied funds to meliorate the substantive effects of Lehman’s failure, to argue that letting Lehman fail was a big policy mistake, you have to marshal evidence that would suggest either that the substantive effect was large and critically important (cause that’s what a rescue could do — reduce the substantive losses) or that a Lehman “rescue” would have been a different “signal”. What was the information content of that “signal” and how would a rescue have changed that content.

    Taylor is projecting, concerning the information content of the signal, supplying his own sophomoric libertarianism to manufacture an alternative argument about the content.

    The question remains: what objective evidence do we have about the actual information content of that signal? And, what about that information content would have been changed by a “rescue”?

    The substantive consequence of not “rescuing” Lehman, and letting Lehman pass through an expedited bankruptcy and liquidation just don’t seem to me to be all that great — as large as Lehman’s bankruptcy was.”

  133. winterspeak writes:

    JKH: I missed that Krugman article. And yes, disappointing. But Krugman’s disappointed for a while now.

    I did not pick up that Brad was a global savings glut adherent too (in general, I like his site a great deal). It boggles my mind that, in the face of massive global deleveraging, where even at ZIRP incomes cannot service debt loads, anyone could claim that the private sector *leveraged too little* 2000-2007 with a straight face.

    As for your proposed system, another benefit is that it removes interbank lending risk. As you know, I support such a system. Big change from today though, no?

  134. JKH writes:


    Yes, Brad is a strong advocate of the global glut explanation. I wouldn’t be surprised at all if he does his next post on the Bernanke speech.

    My point above is that there would be no substantive difference between “clearing balances” on deposit at the Fed, and currency on deposit. Both are forms of money that would equally serve the purpose of interbank clearing and settlement. Both money forms are liabilities of the Fed; the fact that one is electronic and one is paper would be immaterial.

    Obviously, 100 per cent reserves would constitute a substantive difference. But the difference wouldn’t depend on the form of money – clearing balances or currency – it would depend only on the amount deposited.

  135. JKH writes:

    Re above, from Bernanke’s speech this morning:

    “The world is suffering through the worst financial crisis since the 1930s, a crisis that has precipitated a sharp downturn in the global economy. Its fundamental causes remain in dispute. In my view, however, it is impossible to understand this crisis without reference to the global imbalances in trade and capital flows that began in the latter half of the 1990s. In the simplest terms, these imbalances reflected a chronic lack of saving relative to investment in the United States and some other industrial countries, combined with an extraordinary increase in saving relative to investment in many emerging market nations. The increase in excess saving in the emerging world resulted in turn from factors such as rapid economic growth in high-saving East Asian economies accompanied, outside of China, by reduced investment rates; large buildups in foreign exchange reserves in a number of emerging markets; and substantial increases in revenues received by exporters of oil and other commodities. Like water seeking its level, saving flowed from where it was abundant to where it was deficient, with the result that the United States and some other advanced countries experienced large capital inflows for more than a decade, even as real long-term interest rates remained low.

    The global imbalances were the joint responsibility of the United States and our trading partners, and although the topic was a perennial one at international conferences, we collectively did not do enough to reduce those imbalances. However, the responsibility to use the resulting capital inflows effectively fell primarily on the receiving countries, particularly the United States. The details of the story are complex, but, broadly speaking, the risk-management systems of the private sector and government oversight of the financial sector in the United States and some other industrial countries failed to ensure that the inrush of capital was prudently invested, a failure that has led to a powerful reversal in investor sentiment and a seizing up of credit markets. In certain respects, our experience parallels that of some emerging-market countries in the 1990s, whose financial sectors and regulatory regimes likewise proved inadequate for efficiently investing large inflows of saving from abroad. When those failures became evident, investors lost confidence and crises ensued. A clear and highly consequential difference, however, is that the crises of the 1990s were regional, whereas the current crisis has become global.”

    This is actually a tweaking of Bernanke’s original explanation of the glut, where he contended that the glut was truly “global” in the sense of excess global saving driving down real rates of interest. He backs off that somewhat here, noting “joint responsibility” instead of implying a “global” problem that was implicitly everybody else’s making. This is a much improved explanation from his original. Still, a Mosler consistent paradigm would view the US current account deficit as financing capital inflows, which I have believe for a long time. It’s the US originated financial engineering that generated the financing of the housing bubble that generated mortgage equity withdrawal that generated the CA deficit that generated the US capital inflow, in my view. This is also consistent with Taylor’s fed funds explanation. Of course, the fact that China didn’t want to buy any stuff from the US didn’t help.


  136. winterspeak writes:

    JKH: The new Bernanke explanation is a little better. Mosler would outright reject the old explanation (too much saving driving down interest rates) by arguing that CBs can set interest rates wherever they like, up and down the yield curve, if they so choose. We’re seeing more aggressive monetary policy take us there now.

    Mosler would not like the new Bernanke explanation either. He would argue that the Chinese desire to hold US$ means that the Govt should run larger deficits to fund that desire. Insufficient deficits means that the $s were produced as credit via banks. Banks created deposits by making loans, deposits were traded for Chinese goods, moving from US banks to Chinese banks, BoC swapped chinese money for US$ and saved US$. If USG had run higher deficits, say by lowering taxes, then HHs could purchase chinese goods directly, and the system would not need this precarious bank capital.

    I think that low interest rates play a bigger role in the 2000-2007 housing bubble than deficits (or surpluses).

    btw. Great find with “Worthwhile Canadian Initiative”. I was looking for a good source of canadian finance and banking since our last discussion. While Fareed Zakaria briefly catapulted Canada into the limelight a while ago, ppl were bored by details and it has slipped below the broader consciousness again.

    Finally, and I think Nick Rowe gets to this, once monetary policy gets unorthodox enough, it starts looking like fiscal policy. USG decides to buy MBSs. Fine, but at what price? Too high, and you have fiscal transfer to holders. As I personally don’t think banks are the key to this mess, I’d prefer to help HHs, I’m frustrated and bored with the ever more elaborate convolutions the Obama administration is coming up with the give money to banks on the sly.

  137. JKH writes:

    john C. halasz,

    Bruce Wilder wrote:

    “As far as I know, Lehman’s failure had some, but limited effects on other institutions holding Lehman paper. I think one money market fund busted a buck. But, even that was more a demonstration of what could happen more broadly — it was not, in itself, a broad, substantive effect on many money market funds.”

    This was not the case. Bernanke in this morning’s speech:

    “Another issue that warrants attention is the potential fragility of the money market mutual fund sector. Last fall, as a result of losses on Lehman Brothers commercial paper, a prominent money market mutual fund “broke the buck”–that is, was unable to maintain a net asset value of $1 per share. Over subsequent days, fearful investors withdrew more than $250 billion from prime money market mutual funds. The magnitude of these withdrawals decreased only after the Treasury announced a guarantee program for money market mutual fund investors and the Federal Reserve established a new lending program to support liquidity in the asset-backed commercial paper market.”

    Systemic money market fund outflows following the Lehman event caused the Fed to fear the entire system was at the brink. This was the pivotal event. Bernanke referred to it again in Q&A; Buffet also noted it in this weekend’s interview.

  138. JKH writes:


    I just left this for Nick:

    ” Here are two people who agree with you that it’s a liquidity crisis more than a solvency crisis – John Hempton and Warren Buffett:


    And here is the post that you referred to above, where the same John Hempton delineated various definitions of solvency:



    Now I can cover myself and say that it’s a crisis in the definition of solvency.

    Finally, stay tuned for the possibility of critical adjustments to mark-to-market accounting in the roll out of the full Treasury plan over the next few weeks. MTM treatment also relates to the appropriate definition of solvency.”

  139. winterspeak writes:

    JKH: Hah! I think you are right. How else to explain the fact that we’re still arguing over this two years into the greatest financial crises since the Great Depression?

    I think you are quite right that we’ll see MTM being suspended. But will financials rise or fall on that change? Eventually all the transfer has to matter, yes?

    Finally, could you drop me a note at winterspeak AT winterspeak DOT com? Just wanted to thank you for the good discussions.

  140. JKH writes:


    I just posed a couple of questions to Mosler that are central to our discussions:





    (I’m very rushed today and tomorrow; drop you a line later).

  141. JKH writes:


    I don’t know precisely what the MTM adjustment will be if they do it.

    But the effect will be positive.

    Bank stocks will rally big time.

    The reason is that it effectively extends the time horizon for capital adjustment. This allows for the Hemptonesque solvency model that includes internal capital generation.

    It’s a smart thing to do.

    MTM is not a necessary condition for transparency.

    BTW, did you see my Worthwhile Canadian Initiative guest post on MTM?



    (I really didn’t want to do it. They asked me. It’s a headache of a subject.)

  142. winterspeak writes:

    JKH: Mosler agrees with you. Spreads are incredibly wide — banks are making only the most profitable loans, on excellent terms. They are paying nothing out in interest (low interest, bad for AD) and are set up for a nice pop. Overtime they can completely recapitalize out of cheap Gov money and bad terms of the non-banking sector.

    I ask you, why aren’t we all in Finance?

    Will check out the rest.

  143. JKH writes:


    The mark-to-market issue has been gathering quite a head of steam in just the last few days, with a congressional hearing on it today.

    Holman Jenkins, Wall Street Journal, yesterday:


    “Now comes Warren Buffett, a big investor in Wells Fargo, M&T Bank and several other banks, who, during his marathon appearance on CNBC Monday, clearly called for suspension of mark-to-market accounting for regulatory capital purposes … Mark-to-market accounting is fine for disclosure purposes, because investors are not required to take actions based on it. It’s not so fine for regulatory purposes. It doesn’t just inform but can dictate actions that make no sense in the circumstances. Banks can be forced to raise capital when capital is unavailable or unduly expensive; regulators can be forced to treat banks as insolvent though their assets continue to perform. CNBC, sadly, has been playing a loop of Mr. Buffett’s remarks that does a consummate job of leaving out his most important point. Nobody cares about the merits of mark-to-market in the abstract, but how it impacts our current banking crisis. And his exact words were that it is “gasoline on the fire in terms of financial institutions.” Depressing bank stocks today, he said, is precisely the question of whether banks will be “forced to sell stock at ridiculously low prices” to meet the capital adequacy rules. “If they don’t have to sell stock at distressed prices, I think a number of them will do very, very well.” He also proposed a fix, which CNBC duly omitted from its loop, namely to “not have the regulators say, ‘We’re going to force you to put a lot more capital in based on these mark-to-market figures.””

    So what Jenkins/Buffett are saying here effectively is that MTM should be disclosed, but not necessarily systemically forced through profit and capital accounts, the same as I’ve said elsewhere.

    Bernanke and Buffett have both said over the past week that MTM should not be suspended. But they’ve also said that meaningful adjustments should be considered in translating MTM information to regulatory capital requirements.

    I’ll predict that there will be some sort of adjustment coming out in the coming weeks, and that it will have a positive effect on the banking crisis, the interpretation of the Geithner policy response (details still forthcoming), and bank stock valuations.

  144. winterspeak writes:

    JKH: All good points.

    This discussion, as well as the one on Mosler and the Canadian blog had started me thinking about capital requirements being a key element of bank regulation. The argument to relax MTM, or capital requirements now, is essentially using capital requirements to make banking counter-cyclical (or at least, less pro-cyclical).

    It at least has the benefit of only have one effect — transferring money to banks — as opposed to the dual (and opposing) effects of traditional monetary policy (lower rates make credit cheaper — stimulative — and also reduce interest income — counter stimulative). But like the Greenspan/Bernanke put, I doubt actively managed capital requirements will work in both directions.

    btw. can you make head of tail of what Sumner means when he says “unconventional monetary policy”? Does he just mean the Govt buying something other than Treasuries to reduce interest rates further up the yield curve, and/or in other markets? The point being, when you can no longer bid on one instrument, start bidding on others to increase the price? Since I cannot understand (or find) what he means, I cannot reconcile his views with Mosler’s.

  145. JKH writes:


    MTM implementation increases measured market risk, and therefore capital requirements, and so is one aspect of capital pro-cyclicality. Bank capital attribution models are based to a great degree on corresponding value-at-risk models; the “great moderation” reduced real and financial volatility and so reduced Var and capital attribution calculations. Nassim Taleb was all over this.

    Capital is crucial to regulation. One of the ironies of the common “reserve multiplier” mistake is that it distracts from the true dynamic of bank balance sheet credit and deposit expansion, which is capital based. The purpose of capital is to absorb losses and insure against risk. A bank’s allocation of capital involves choices for balance sheet expansion based on risk. The “reserve multiplier” has nothing to do with risk.

    It’s interesting that the Mosler paradigm is not a model of risk per se. That’s unusual for a financial or economic model of any type these days.

    Not sure about Summers – unconventional usually means quantitative (monetary base) and or/qualitative (asset credit risk) easing. The Fed has done both already except they haven’t yet starting buying treasuries under quantitative easing. The purpose of buying treasuries in this environment as I understand it would be to help get mortgage rates down.

  146. winterspeak writes:

    JKH: You are exactly right. Capital requirements *actually* work the way people *think* reserve requirements do. Hempton argues that, since capital requirements are a buffer, it is right to let banks be undercapitalized right now and now make them raise additional equity. Let the buffer buffer. I’m suggesting that the Govt alter capital requirements more dynamically — explicitly increase them during credit expansion, and decrease them during credit contraction. So, instead of MTM and VAR driving capital requirements, it’s done by decree. Certainly actual capital requirements are being driven that way today!

    Glad you are also confused about Sumner. How on earth can some blog well enough to get the ear of Paul Krugman and still be completely unclear as to what the actual recommendation is!?

    Buying Treasuries is one way to get mortgage rates down. Expanding what Freddie and Fannie cover is another. Mortgage rates are very low if you have any flavor of conforming loans today. The amount of money that’s been channeled to home sellers over the past 10 years is amazing. First banks, and now Uncle Sam. Sucks to be short housing (aka “renter”)

  147. JKH writes:


    Thought you meant Larry Summers. But now I recall reading the Scott Sumner blog also. Didn’t really connect with it in any event.

  148. JKH writes:


    I’m completely baffled by Hempton’s latest, where he is having some battle with Krugman over non-recourse financing for TARP II.

    I left a few comments, which left him impatient, so I guess I have no absolutely no idea what recourse or non-recourse actually means.


    I don’t know what “banks are non-recourse” means, a phrase which Hempton uses. Any thoughts? Why am I so clueless on this?


    SRW – much appreciate being able to use this post for discussion. It won’t go on forever.

  149. winterspeak writes:

    JKH: I read the Hempton thread, and am not sure I understand it either. I think by “non recourse” he simply means that the private capital is in a first loss position, but the bulk of the losses may be borne by taxpayers (in this instance). So, with a regular bank, equity holders (private capital) gets wiped our first, but the tax payer picks up the rest (via FDIC). I have no idea where private debt holders stand as they seem to enjoy a truly exalted place in the cosmos. Hempton is saying that, if we are OK with having taxpayer money at stake, so long as private money is in a first loss position, then the only question is: do we have a large enough private stake in that first loss position?

    Just my guess though. I don’t like putting words in other people’s mouths.

    My personal view is that all of this talk about “taxpayer losses and gains” is meaningless. All money comes from the Government — it has no need to “make” money on a deal as it can simply create its own, if it so chooses. The image is like “if the Government drives a bargain like Warren Buffet we’ll get a tax break because the Govt will have made so much money”. Nonsense — Govt can make as much money as it wants, whenever it wants.

    The real key is transfer here and now. The Govt could have guaranteed all liabilities in the case of Lehman (so counter-parties were made whole) and let debt and equity holders go to zero. They could so the same thing with Citi and BofA. Or they could just declare a 5% haircut — not big enough to kill anything that wasn’t already essentially dead, but enough to say that bondholders need to, ahem, think about what they invest in. When we think about what risks to take out of the system, and what to put in, counterparty risk seems unhelpful at this point. The cheap leverage the Fed is offering to private equity is a total transfer to a group of insanely rich people. I see no reason why this should be non- recourse (or why the PE firms should not be at risk for more than 100% of their “investment”).

    JKH, I’m on the road tomorrow and will be traveling for quite some time, so this will have to end the thread. Was serious about the offer though: winterspeak AT winterspeak DOT com. I’ll buy you a beer if we turn out to be nearby.