Value for value

Would it have been better if Timothy Geithner had had the power to guarantee all bank debt early on? As James Surowiecki reminds us, that was part of the Swedish solution. Justin Fox plausibly suggests that we might have avoided a lot of pain with a fast, full guarantee.

But that’s not the point. The question isn’t whether we could have avoided this crisis, if only we had cut a big check. We could have, and that was not lost to any of us debating these issues more than a year ago. (See e.g. me or Mark Thoma.) Had we done so, the near-to-medium term fiscal costs might have been less than they probably will be now. So, with 20/20 hindight, would it have been a good idea?

How you answer that question depends upon how you view the crisis. Is it an aberration, a shock to a basically sound financial system, or is it a painful symptom of an even more dangerous condition? Under what circumstances would our political system be likely to impose reforms that would prevent large scale misallocations of capital and shifting of losses to taxpayers in the future?

If you think that our financial system just needs some tweaks, some consolidation of regulators’ organizational charts and sterner supervision, then you should prefer that we had just cut a check, passed Sarbanes/Oxley Book II, and moved on. But that is not what I, or most proponents of nationalization temporary receivership for insolvent banks, believe.

If you believe, as I do, that we need a root-and-branch reorganization of the financial system, which must necessarily involve the dismemberment and intrusive restraint of deeply entrenched institutions, does that mean pain is the only way forward, “the worse the better” in the old revolutionary cliché? It need not mean that. But it does mean that palliative measures, like giving the banks money, would have to be attached to curative measures, like enacting capital requirements and imposing regulatory burdens that would force financial behemoths to break themselves up or become boring narrow banks. For almost two years, policymakers at the Fed and the Treasury, including Secretary Geithner, have offered bail-out after bail-out and asked for nothing serious in return.

Do I regret that Henry Paulson was not empowered to issue a blanket guarantee of bank assets early on, as the Swedes did? No, I don’t regret that at all. Why not? Because I think that “Hank the Tank” was a crappy negotiator, not only for taxpayers in a fiscal sense, but also for the economy and the polity more deeply. He would have offered the financial system sugar without requiring it to make the medicine go down. He may believe, quite sincerely, that a cure would be worse than the disease. He may believe that, but he is wrong. If we “get past this crisis” by restarting a consumer-credit-based, indiscriminate-investor-financed, current-account-deficit-making, income-inequality-expanding economy, we will have increased, not diminished, the likelihood of a major collapse.

You may believe that we have learned our lesson, that if we can just get some stability and comfort for a while we are prepared to do what must be done. That’s a respectable position. But I don’t share it, and neither do the majority of Americans who are unwilling to allow their representatives to sign off on any more expensive aspirin. We want value for value, an ironclad commitment of root and branch reform in exchange for the unimaginable sums of money we are being asked to hand over.

Surowiecki has in the past suggested that those of us who favor nationalization would criticize any alternative simply because it is not precisely the policy we advocate. But it is not we who have refused to compromise. We’ve seen variations on the same basic proposal over and over again. Geithner’s PPIP really does resemble Paulson’s TARP, besides the part about actually asking taxpayers for the money. Each latest plan from our incestuous cadre of economic Mandarins demands only symbolic concessions from the dysfunctional organizations we are asked to support. The “moderate” political class goes on and on about how Geithner and Bernanke have to go all Enron, funding the banks via off-balance-sheet guarantees and special purpose vehicles, because “populist, childish” Congress won’t put up the money. Setting aside how audaciously corrosive that sentiment is to Constitutional democracy, it is simply wrong. Congress would, because the public would, support large, explicit transfers, if they were attached to reforms sufficiently radical to prevent a recurrence, and suitably punitive towards the people who managed the system that brought us here. Value for value.

I am a true believer in American-style capitalism. So I would like to see people who earned profits lending to banks in good times bear the high costs of failing to monitor the organizations they funded. Investor fear is what is supposed to prevent the indiscriminate misuse of capital. To the degree that creditors have leaned upon “implicit” government guarantees, I think it would both be just and set a useful precedent if they were reminded that investors have to take responsibility for where they place the precious capital they steward.

That said, like Paul Krugman, I would be willing to hold my nose and tolerate a Swedish-style guarantee of bank creditors. I’d acquiesce to that even without formal nationalization. Nationalization is no one’s idée fixée. It is a means to an end, and the desired end is a world in which too big to fail is too big to exist for any financial institution that originates or holds credit risk in any form. Secretary Geithner could send a bill to Congress today that would put all banks with a balance sheet of over $50B into run-off mode, while clearing away legal obstacles so that larger organizations could arrange their own breakups over time. I’d fax my Congressman and support a $2T on-budget buyout of bank creditors as part of that bill, as long as it had teeth. (“Teeth” would include making sure that off-balance-sheet and derivative exposures were included in the size cap, etc.)

It’s not that us pitchfork-totin’ populists are unwilling to pay the bill. It’s that we want to know that in exchange for writing a very, very large check, the people that we are paying will actually deliver the goods. Given the behavior of bankers before the crisis and of shifty policymakers during, we have every reason to watch warily and to insist upon every precaution while we hand over suitcase after suitcase of freshly printed Federal Reserve notes.

Update History:
  • 28-Apr-2009, 1:20 a.m. EDT: Thanks to the excellent Nemo of self-evident fame for pointing out that I’d forgotten the tricky distinction between “to” and “too”. Fixed, I hope. From now on, I’m jus’ gonna write “2B2F”.
  • 28-Apr-2009, 2:40 a.m. EDT: imply include

69 Responses to “Value for value”

  1. reason writes:

    Great post again. I particularly like that the current account mess is brought into play. The problem is that a solution to that must come at the expense of foreigners who are desperately trying to keep the status quo going. I don’t see that banks can individually do much about that.

  2. Anon writes:

    “… I’d fax my Congressman and support a $2T on-budget buyout of bank creditors as part of that bill, as long as it had teeth. …”

    Wrong. No cash for trash, period.

  3. beezer writes:

    “I am a true believer in American-style capitalism. So I would like to see people who earned profits lending to banks in good times bear the high costs of failing to monitor the organizations they funded. Investor fear is what is supposed to prevent the indiscriminate misuse of capital. To the degree that creditors have leaned upon “implicit” government guarantees, I think it would both be just and set a useful precedent if they were reminded that investors have to take responsibility for where they place the precious capital they steward.”

    Bingo! These folks were supposed to be the “cops on the beat” responsible for wise stewardship. They blew it. Some of the changes needed are those that restore investor control of corporations they fund. Complacency should be a victim of all this, but government bailouts prevent it becoming a victim.

    The Citi crew was re-elected, for heaven’s sake. The only reasons for doing so, in my opinion, is that investors figure the current crew hold the best hope of continued bailouts!

  4. mittelwerk writes:

    the banks have been lending to the banks.

    this is not sweden — a comparitively highly homogenous society.

    if you haven’t noticed, this is not really a “constituional” “democracy” anymore, either — but a corporatist oligarchy. fancy words to describe the logical outcome of a century of increasing moral hazard. we NEED liquidity now in mulitples of output.

    the dismantling of banking functions you speak of means the dismantling of a social class, the subordiinate classes that serve them, and a big hit to the population at large, which is implicated in it all via excessive debt.

    this doesn’t happen because someone made a good argument about “receivership”: it happens because there are mobs in the street that want it to happen.

  5. winterspeak writes:

    Great post Steve.

    And I agree — Congress will not give more money to the Finance industry not because they are childish, but because they want to see the industry fundamentally reformed. Or even slightly reformed.

    One lesson I hope everyone, particularly free-marketers, take from the crises is how market discipline simply does not work on the liability side of banks’ balance sheets. Should the Fed just guarantee it all? I say yes. Why in God’s name is FDIC capped at $250K? Why do we have money market accounts at all?

    Of course, if the liabilities are Govt guaranteed, then you want the assets to be extremely regulated. It’s a shame that people bring up securitization as a helpful financial innovation, because I think it should be outlawed. Let bankers do something useful before their 3pm golf game, like assess credit risk based on actual local knowledge. Gasp!

  6. ws — speaking of banks and soft information, did you see this rortybomb post, and the very nice study behind it?

    turns out that bankers can make use of specific knowledge in evaluating credit. FICOs ain’t everything. who knew?

  7. winterspeak writes:


    I had not seen that, and an elegant study too. Thanks for sharing.

    There is lots of room for soft knowledge to make better credit decisions. I remember when I graduated college (a good school) and worked at DE Shaw, but could not get a credit card because I had no history. An uncle at Citi (of all places) helped me out, but discretion and a brain would have had a lender make the right decision.

    I remember years later having trouble getting a cheap car loan because I was a student (last year of MBA) and so was my wife (last year of med school). Our income as lousy, but we were great risks. The computer said no.

    Gary Becker (U Chicago, Nobel, etc. etc.) had a good paper on discrimination in Boston, looking at home loans and redlining. He found that, although black applicants were rejected more frequently than white applicants, both loan portfolios performed equally. This is the same logic as the ppt on Rortybomb.

    My final point on credit “efficiency”, particularly in housing, is to ask how it’s a net benefit at all? More efficient markets lower interest rates, great, but prices go up to compensate. It’s a transfer for buyers to sellers, and a net wash across the economy as a whole.

  8. ws — i have very mixed feelings about economic notions like “efficiency” and “optimality” in general, but i will make a small point.

    credit allocation affects more than who wins and loses, but also affects what gets done. so, in housing markets, very EZ credit did result in a transfer from buyers to sellers. but it also resulted in a lot of people purchasing homes who otherwise would not have purchased homes (and some lesser number of people not purchasing homes who otherwise would have, for a variety of reasons). as i know you know, “efficient” in the way that economists use the term is not so much about who wins and loses, but about aggregate welfare however defined.

    (of course no reasonable definition of aggregate welfare can be indifferent to who wins and loses, given radically different “marginal utilities” in the lingo, but like most convention economics, we’ll put that aside for the monent.)

    so, a case can be made for efficient credit allocation on the grounds that it results in changes of behavior that increase aggregate welfare. the case is much stronger with respect to business lending than consumer lending. (the case for consumer lending beyond non-rolling transactional credit and the financing of necessity long term durables is weak in general.) consider two businesses looking for loans. if funded, one business will create immense value, while the other will produce little, although both entrepreneurs are sincere. neither business will get off the ground without the loan. ideally, bankers would price the loans accurately, giving an objective view of the project prospects to the budding capitalists. by offering the poor business plan loans at only a high rate of interest, bankers at the margin discourage the squandering of real capital on bad projects, while by offering favorable loans to the good prospect, at the margin, they encourage good use of real capital. there is a range of loan prices that both the banker and the good business would accept, and where in the range of prices the loan is eventually made doesn’t matter from an efficiency perspective. that’s just a split of the proceeds. what matters, what makes lending efficient or not, is that the pricing of the loans is consistent with encouraging good projects to happen and bad projects to get nixed.

    you can extend this kind of analysis to houses and mortgages, but the informational task facing the banker is much, much trickier. the market price of housing has nothing to do with whether a mortage is “productive” in an economic sense, although bankers may be tempted to believe that it offers them security in their collateral. but market prices only offer security when markets are rational and circumstances are stable. a rational banker would have to evaluate the home/homeowner pair as a single utility producing entity. she would have to ask 1) to what degree is the homeowner capable of meeting the terms of the loan despite the fact that the good produced by the investment (housing consumption) is nonsalable; and 2) to what degree is the utility gained by purchasing the home certainly greater than the utility that could be produced by some alternative use of the borrower’s funds. just as a business lender will price loans differently based on the probability of business success, even among secured loans that in all probability be repaid, a mortgage lender should price home loans based on these criteria, assuring that homes are easily purchased when they’ll be worthwhile to the consumer, not in terms of comparable market price, but in terms of consumption foregone in purchasing the home. but of course bankers have no way of independently evaluating the utilities of alternative consumption baskets, so they defer to the potential borrower, offering a very limited informational contribution to the question of “should this person buy this home at this price”. capitalists are supposed to be gatekeepers, crosschecking the enthusiasm of entrepreneurs and thereby contributing to efficient real behavior. but with respect to homes, other than (sometimes) evaluating capacity to pay, they defer to enthusiastic homebuyers who don’t reliably check the burden of market prices against alternative consumption baskets. so they enable inefficient use of capital.

    i think lenders should be very skeptical of financing consumption goods, for this reason. they cannot evaluate the projects, and may be enabling myopia and foolishness. myopic borrower sometimes service their loans like reliable slaves once they realize their error, but people paying for a mistake are much less likely to pay than people getting continual value for their interest payments. for both the borrower and lender’s sake, questionable consumption benefots should be self-financed. of course if this became the law of the land, mortgage finance would in general be restricted to say $200K max (the cost of a utilitarian home for a fam of four), and anything over would be paid with cash. i think that would be a better world, frankly, but the real-estate community would disagree.

  9. reason writes:


    I think you should rethink that last one. Here in Germany there aren’t any housing bubbles. Partly that is due to a slowly growing population and highly regulated rental markets, but partly it is also due to a tradition that says housing should be financed with 33% down, 33% financed from a savings contract (bau-spar) and 33% from a mortgage loan (slow repayment of principal). And you can’t walk away from a mortgage (you or a guarantor stay on the hook). You don’t need absolute limits, simply insisting on down payments and a guarantor would do the trick.

    P.S. I had a brainwave on an issue we discussed before – limited liability and leverage. In stead of saying that limited liability meant limited leverage – why don’t we upgrade shareholders to have the same rights as bond-holders?

  10. oh, the 200K thing wasn’t intended as an absolute limit, just as an example of where a philosophy of not financing consumption would lead (with basic housing being evaluable in clear terms as investment, relative to a lack thereof). and there wasn’t a policy proposal there, just some thinking out loud about how the world might be different under an ethic of “efficient lending” in the sense of using credit allocation to increase overall economic efficiency.

    i have mixed feelings about the german model as you describe it, but i am sure it would be much closer to “efficient” because borrowers are so much more on the hook that they’d evaluate the opportunity cost of consumption more skeptically, and because high down payments permit one to evaluate the opportunity cost of consumption in the near rather than far future, which is easier. but what is a “savings contract (bau-spar)”, as opposed to a down payment or mortgage loan?

    making shareholders look like bondholders either means a lot of fragility (if dividends or redeptions become like interest or principal payments, enforceable by bankruptcy), or interpreting in the other direction, making bondholders look like shareholders so no one gets to force bankruptcy, takes us towards all equity financing, which would be much less fragile, and more in a direction i’d like to go… (more on this soon, i hope.)

    but i feel like i’m missing the point, missing the essence of what you had in mind. i remember our conversation about eliminating limited liability, and the idea that excess leverage might be a trigger for doing so, but can you flesh out where you’re going? is it your proposal to make firms effectively all equity by flattening the capital structure, thereby making limited liability moot? all claimants get an equal share of the corpse if stuff happens, and that’s that?

  11. reason writes:


    Re Bauspar – that is a contract with two parts one a debt with a fixed interest and one a savings account which pays interest – that is it is much more flexible than a normal mortgage. But the catch is you have to start it in advance.

    Re Making share owners more like bond holders – I was actually thinking that if in the case of insolvency all were treated equally, then leverage would be fairly automatically limited. I haven’t thought it through fully though. Nothing changes in normal times – just in case of insolvency.

  12. dave writes:

    The problem with the regulated model is always what to regulate. As you’ve mentioned, the availability of credit is power. Those with credit use it to successfully bid on scarce resources, shaping what our society produces.

    How does the regulator know what should be produced? Here in America for instance it was decided that housing was something we wanted produced, so it was subsidized heavily (interest deduction, Fannie/Freddie, etc.) That seems idiotic now, but it was all the rage a few years ago. If your going to have regulators giving banks direction on how to allocate capital exactly what guidelines are you going to use. How are regulators suppose to know what a responsible investment is and isn’t?

    Going back to a little Hayek, isn’t the information needed to know which projects are best distributed out there in the market, beyond the reach of a central regulator.

    To me the entire finance debacle is a principle agent problem. Ownership of capital has become so diluted that owners aren’t excersizing oversight of their capital. They are trusting it to agents, who have their own interest, not the principles interest, in mind. Will adding another layer of government agents to monitor the first set of agents change the fact that owners are not managing their capital?

    The more layers of agents you add just distorts the fundamental problem: that people are not investing in real economic activity, but whatever they feel they can sell to the next guy for more. Simply put, its more profitable for owners to speculate on which accounts will be blessed with a government guarantee then figure out what companies have promising products. Telling principals that they don’t need to worry about that stuff because a government agent is looking out for them isn’t a long run solution, assuming of course those government agents are flesh and blood human beings with all the same limitations as private agents.

  13. winterspeak writes:

    srw – i hear what you are saying about “efficiency” etc. In the mortgage market though, if more people were buying homes it also means that an equal number of more people were *selling* homes. The transfer from buyers to sellers, or rather, from the banks financing the buyers to the sellers, is complete. Net net, a total wash.

    Starting new businesses should be entirely around credit risk. Why not have interest rates at zero and then have lenders think very very carefully about credit risk? Is there a purpose to a FRM vig on new enterprise?

    Just thinking out loud here. The Obama administration is working day and night to put Humpty back together again, and I look at it and simply can’t see the point.

  14. mittelwerk writes:

    that’s cool, dave, i always like a little rentier thought with my afternoon gruel.

  15. winterspeak writes:

    Dave: An obvious step 1 would be to stop securitizing. If that would lead to rates “too high” we can just set FMR to zero. Let all the risk be credit risk, and let all the work be assessing that credit risk by people who will keep assets on their own books.

    Govt backs all liabilities.


  16. dave writes:

    What do you mean by FMR. Google was no help:

  17. mittelwerk writes:

    isn’t a zero-percent interest rate merely communism with a twist?

  18. dave — i’m with you. that’s why i very much prefer structural to supervisory approaches to regulation. ensure financial intermediaries are structured in such a way that funders have incentives to monitor. government supervision can never be adequate, skews credit allocation, and creates implicit warranties and guarantees.

    ws — mostly agreed, though it’s worth pointing out that patterns of who buys houses for how much are real economy facts with real economy consequences that might be more or less “optimal”. also, bank credit allocation led to much more obvious real economic costs, in that scarce physical capital was pretty clearly squandered in building new homes and subdivisions and other residential investment (improvements). it didn’t, and doesn’t just lead to transfers of the existing stock between buyers and sellers.

    mittlewerk — i think ws is talking about a zero-percent (real) risk-free rate on gob’t obligations. borrowers would still pay interest as compensation for risk in the form of a credit spread. that its quite the opposite of “communism with a twist”. on the contrary, governments offering positive real returns on risk-free short-term lending is socialism for savers, because it implies a transfer of real goods and services to savers who bear no risk on behalf of any productive enterprise and make no other contribution, to the producers of those goods and services. you might argue that lending to the government is a contribution to the broad economy, as after all the government does invest in crucial public goods. but as ws will point out, the government doesn’t need to borrow to fund anything, it has the power to conjure funds out of thin air, and it can “sterilize” the inflationary effect, if it chooses to, by straightforward taxation. government borrowing is a kind of subterfuge, a means of obscuring the inflation and/or taxation implications of its spending. politicians may find it convenient to pay lenders a real return for their contribution to that subterfuge, but there’s little reason that the rest of us should be enthusiastic.

  19. dave writes:

    Thanks for the clarification Steve. I started typing something, but a second and third glance over of your last paragraph was a bit of an enlightenment.

    At the end of the day the government has to pay people a fee for borrowing, because they must be willing participants. Taxes and inflation are things unilateraly hoisted on the public backed by the threat of force. No ROI is required because there is no transaction, save handing over the loot at gunpoint. Borrowing on the other hand requires a willing buyer. Since nearly all of us would prefer to spend now versus later you’re going to need a positive yield to attract willing participants.

    Borrowing is as a substitute for inflation or taxation. It spreads out inflation and taxation over time. It is thus a useful tool since high levels of inflation or taxation in a short period can have huge consequences.

    The government could increase current aggregate spending through current inflation instead of borrowing, but this does damage to the value of money, which distorts the entire system of economic calculation. Screweing with money has a real effect on economic allocation and productivity. See 70s stagflation for example of the consequences.

    Raising taxes has its own drawbacks as well. If people didn’t change their behaivor based on taxes then increasing taxation to match increased spending is a simple transfer of current spending power from the private market to the public market. Ignore ones preference on that issue for a second. In reality tax rates effect the actions that people in the private market are willing to take. Tax them too much and they become disinclined to engage in economic activity, thus reducing the tax base. Rather then get into a debate of where we are on that curve, lets simply acknowledge there is at some point a level of taxation past which one can’t go.

    While borrowing allows one to spread the pain out over time, it also obscures the level of the pain from the general public and investors. People don’t quite grasp how much current spending will cost them over time. The public can manage its own checkbook to a reasonable degree, but the government less so.

    A big part of this is that there is no market disciplin on the government. When people look to buy treasuries they aren’t just evaluating worthiness of what the government is doing with the money it is borrowing, as they would with non-risk free private borrower. The first thought in thier mind is that they are senoir creditor of the countries resources. The government, unlike any other body, has the ability to confiscate future productivity by force (taxation) in order to pay its debts. While there are other advantages to government debt (diversification into countrywide public investment), ultimately the monopoly of force is the biggest advantage the borrower has. This is extra special true for military superpowers holding the worlds reserve currency.

    The government could turn to inflation rather then taxation, but once burned creditors want ever better terms on their purchases, until they stop extending credit to the government at all. At this point if the government can’t live within its means it must either default or hyperinflate on its debt and then live with a balanced budget. The damage of such measures is why inflation is loathed as a method for debt service.

    Because of the governments borrowing advantages it does not have to justify expenditures in the same way private business does. Thus the market is not able to give the government an adequet signal as to whether or not its spending is productive or not.

  20. dave — two quick comments. i don’t think we disagree very much, although i am more skeptical than you are about the legitimacy of governments using borrowing as an alternative to taxation and inflation. i think that it’s important to point out that from a cash flow perspective, everything you could do with (domestic) borrowing, you can almsot simulate via taxation. a government could tax lump sums from the very same people who would otherwise purchase a bond, and reduce their future tax levels relative to the preexisting baseline to repay the loan. there are obvious problems with this thought experiment — govt doesn’t have perfect information on individual preferences to tax exactly the same people who would have purchased its bonds, plus the perception that arrangement are voluntary rather than coerced may have important effects despite identical cash flows. but to a rough approximation, anything governments do with borrowing they could do more transparently without borrowing.

    the second point i’d add is that government only needs to offer positive real yields to investors when it borrows in very large quantity or when there is uncertainty surounding govts ability to maintain price stability (ie expected volatility in the price level/inflation). government securities are the financial asset best capable of providing long-term insurance of wealth preservation, if the govt and economy are such that price stability is doable. zero-real-interest Treasuries are the best available option for people who are genuinely savers — that is people who are not maximizing returns, but are trying to transfer the wealth they have today into the future with a high degree of certainty. Transiting wealth into the future, without seeking any appreciation is hard. The next safest tack — storage of some commodity, is costly (storage and insurance), risky (think theft and decay), and speculative (who knows what commodities will be valuable when you retire?). A fiscally cautious state can offer effectively costless storage, as long as its economy does not collapse. For zero-cost, cautious investors can eliminate all idiosyncratic risk (including market risk) and gain assurance that their wealth will be carried forward unless the state weakens or collapses (which risk savers in the aggregate have absolutely no means of avoiding).

    unfortunately, the abuse of government debt as a financing scheme reduces its suitability as low-return insurance. A financial crisis, in a sense, comes about when no entity can credibly offer a risk-free asset. In good times, large bank obligations as well as USG obligations were considered that, not now some kind of volatility in the price level is nearly certain, and bank obligations are toast. my view is that the availability of the insurance provided by credibly risk-free assets to small savers (who shouldn’t be expected to take risk-investment decisions) is far more beneficial than whatever fiscal smoothing and consequence-hiding is offered by unrestricted borrowing of assets. rather than hawking its obligations far and wide and in quantity, i would rather that government ration the obligations it sells, quite analogously to how it rations FDIC insurance. A person should be able to hold no more than $250K in government assets. government debt should be low, so that members of the public can be confident that the purchasing power of these limited-size nest eggs is secure (for exactly the same reason you want your insurance company to have a conservative balance sheet). to the degree we treat govt assets as a license to avoid market discipline, we are likely to provoke market discipline and lose the govts ability to insure the wealth of the middle class. govt should really think of every Treasury security it offers as an insurance contract, not a loan.

  21. reason writes:


    you have discovered the difference between good regulation and bad regulation. Good regulation worries about what the regulator needs to acchieve (avoid systemic collapse) not what the regulated need to acchieve (return with controlled risk). Economists should stop pushing the regulated/not regulated band wagon and concentrate more on the good/bad regulation story.

  22. reason writes:

    P.S. Trying to use a financial fix for housing available is just dumb. It ignores the price of land – that any subsidy just pushes up eventually the price of land. You need to invest in infrastructure to increase the supply of good neighbourhoods (and research to know how best to do that). Neighbourhoods are mostly about externalities and so a clear case of a public good. Subsidising housing just subsidises developers and corrupts local government.

  23. reason writes:

    As I said on another site, if we want to pay more attention to dead economists then it in many ways sense to reconsider Henry George more than von Mises or Hayek or even Keynes. Von Mises, Hayek and Keynes all moved the mainstream (no matter what their often fanatical accolytes might claim), Henry George got buried. I think the fanatical accolytes of Henry George are a pain, but I do think we need his insights in a new synthesis.

  24. winterspeak writes:

    DAVE: Steve is right, by FMR I meant the Federal Funds Rate (that it targets and tries to achieve by open market operations). I should have been clearer.

    SRW: I hear what you’re saying, but I’ve become less concerned with the real asset misallocation of the past. Not because I think it was a good thing, or I underestimate the long term opportunity cost of misallocation, but because when one takes a walk outside in the US it becomes clear that this is a nominal crises, not a real crises. The real misallocation is a sunk cost, there’s no sense in trying to reverse it or cry about it now. The stupidest proposal I’ve heard for ending this crises is to destroy houses (and that’s in the face of some stiff competition).

    Take a walk outside in Pakistan, and the problems there are very real, but here they are nominal, it’s a matter of inconvenient numbers in a spreadsheet. The Govt could simply add $1M to every American’s bank account, or they could simply buy every house for twice the zillow value (Moldbug’s favorite remedy). Housing crises solved, although we may have problems with inflation.

    Instead, the Obama administration gave some americans a check for $400, and is funding private savings through unemployment. Unemployment is a *real* cost to the economy, both in terms of opportunity, and the degradation of real skills. There is a number between $400 and $1000000 that would solve this crises *without* enriching the already grossly overpaid Goldman Sachs.

    I don’t have the same attitude towards savers that you do, Steve, which is fine, but I will point out that positive interest rates on Govt debt are probably the mildest form of redistribution the Govt engages in! Taxes and spending move a far greater amount of money around than the current 0.25% a t-note gets you (down from a recent high of — gasp — 3% or whatever). If you want to talk about transfer, interest rates are the pimple on the elephant.

    I also don’t think that we think of Govt borrowing the same way. I don’t think the Govt needs to borrow at all — it can issue no notes or bills and simply run an overdraft at the Fed. JKH will jump all over me for this, and he may be right, but it remains a fact that the US Govt does not need to borrow dollars from anyone. It should just stop pretending. Yes, lots of other things would change, but I think how US Govt spending is *actually* “financed” would become clearer, and then maybe we could deal with nominal crises by changing numbers in a spreadsheet instead of putting 30M out of work.

    As for your call for low Govt debt, you really *do* have it in for savers don’t you! ; )

  25. JKH writes:


    “will jump all over me for this”

    Not at all, WS!

    Instead I will rise in support of a fundamentally important economic truth in its most pristine form.

    I agree USG doesn’t need to borrow.

    Government expenditure has three immediate effects:

    a) Increased money supply

    b) Increased bank reserves

    c) Government overdraft

    The overdraft is equivalent to a negative deposit, or credit in drag, channelled via the Fed to the USG, as a result of money and reserve creation.

    No borrowing is required.

    USG borrowing is a choice.

    Where to borrow is a further choice.

    A choice to borrow from the Fed changes nothing of substance; it only changes an internal USG/Fed spread sheet entry, which is inconsequential to the economics of the consolidated USG/Fed sector.

    A choice to borrow from the market results in an exchange of USG debt for privately held money and matching bank reserves. Although this is still only a spread sheet entry, it is one that changes the economic balance between the USG/Fed and private sectors, because of interest on debt. (Although the payment of interest on bank reserves these days is now an important consideration.)

    Your “tautology of absent need” is true regardless of the choice of whether to borrow, or where to borrow.

  26. mittelwerk writes:

    am i missing something here?

    US external debt (currently about $14 trillion) does a lot more than pay the bills; it’s probably our dominant foreign-policy instrument. and also the best way to “tax” every trading country for use of the dollar and implied use of the US military.

  27. ws — i’m also unconcerned with real misallocations in the past, but am arguing that credit allocations generally and interest rates on types of collateral specifically can and do lead to real misallocations in general, and therefore we should care, even if we do not care about arbitrary transfers that creates private winners and losers. obviously we can’t undo what has already been wasted, but we can learn why we’ve squandered in the past, and not reproduce those conditions. (we also have to be careful with “sunk costs” arguments, because in a repeated-game context, it sometimes makes sense to expensively punish the sinkers of sunk costs, while the usual sunck costs story says you should just write them out of your analysis. this applies to the debate on how we treat the incumbents associated with troubled institutions, though, not in the present context of how much we should care about credit efficiency.)

    i don’t agree that the crisis in the US is only nominal. i do agree with you regarding that the nominal part, which is the most painful, it is within our power to fix the problem by writing checks, and the main issue becomes how large should the checks be and to whom. so far, we’ve adopted policies of writing large checks to already wealthy and well-connected cohorts, including but by no means primarily “middle class savers” who in fact are an unusually wealthy and connected group. that is a very, very poor choice. i’d be really happy if we did something along the lines you suggest, like writing large flat checks or de-taxing/subsidizing payrolls. the question we the “nominal crisis” is not can we resolve it, but what kind of country do we want to have after we have resolved it, in terms of both what would be efficient and what would be legitimate/fair. i think we are on the road to resolving the nominal crisis, but poorly.

    poorly, because i think this is a real crisis, in the same way as walking the steets of pakistan or wandering in the romanian countryside (i’ve done the latter, not the former) evidence a real crisis. we do not have a real crisis yet in the US if you define that as a crisis of present scarcity. but, in my view, past and continuing trends in human and capital misallocation point us inexorably in that direction, or in a direction of international conflict to ensure continued provision of goods and services we no longer produce. from my perspective, resolving the nominal crisis while leaving in place the trends that underlie a predictable near-future real crisis is a very bad outcome.

    re savings — i only “have it in” for savers if you mean savings in the Winterspeakian sense, ie hording government obligations. i am a huge fan of savings in the traditional sense, the kind that is tautologically investment. people hold government securities for insurance purposes more than for any other reason, but in the aggregate there is a typical moral hazard problem in that the more easily people can insure, the less incentive they have to take actions that would keep insurable risk low, like participating in the capital allocation decision with sufficient care that they are willing to accept downside risk. like any other insurance company, a government that offers insurance at a low premium without regard to moral hazard will find itself bust. bust for a sovereign doesn’t imply default, but does imply welching on a price stability commitment if default is to be avoided. that’s where i’m pretty sure that we are going.

    mittelwerk — it may be true that we have used sovereign debt as a foreign policy instrument and as a means to tax rest-of-world for security services that we arguably provide on their behalf. but that we have used debt as an instrument doesn’t mean it is a good instrument, either globally or even narrowly for the United States. i’d argue that the very poor economic choices we have made under the assumption of cheap external credit will prove to be more costly, even narrowly, for the United States, than the high consumption it enables, under almost any reasonable accounting. as far as rest-of-world is concerned, they’d probably be better off with more transparent and legitimate foreign policy constraints and collective security arrangements than being subject to the strange unpredictable tides of US military and financial decisionmaking.

  28. ws — from the above, i hope it’s clear that the reason i object to positive expected real-return on government securities is because it creates incentives for the government to overinsure and the private-sector to underprovide capital-allocation and risk-bearing. (when the private sector fails to bear the risk of economic projects, project quality deteriorates. the overinsured are disproportionately paid off, then, during bad outcomes largely of their creation, while those who invested privately get swallowed by systemic crisis without recourse to insurance. unlimited access to government debt creates a game where the nash equilibrium is everyone-is-insured, aggregate real economy tanks.)

    the fiscal costs and distributional ramifications of real-return-to-government-debt are concerns, but less important than the effect of “moral hazard” on the quality of private real investment.

  29. Blissex writes:

    «Here in America for instance it was decided that housing was something we wanted produced, so it was subsidized heavily (interest deduction, Fannie/Freddie, etc.)»

    It was not housing — it was conservative voters, as home ownership correlates very strongly with conservative voting. The Republicans have been very explicit about their aims, and blue dog democrats have also had an interest in promoting conservative-leaning voting.

    As Norquist said:

    From Norquist

    «The growth of the investor class—those 70 per cent of voters who own stock and are more opposed to taxes and regulations on business as a result — is strengthening the conservative movement. More gun owners, fewer labor union members, more homeschoolers, more property owners and a dwindling number of FDR-era Democrats all strengthen the conservative movement versus the Democrats.»

  30. blissex — i wonder how well norquist’s logic holds after the “investor class” loses its shirts in a crash clear incubated during a Republican administration…

    (btw, i say that by no means as any kind of capital-d Democrat. for the moment the dems seem lots less bad than the repubs, except on the financial issues that most exercise me, on which the two parties are mostly indistinguishable. my politics don’t jibe well with either party. jus’ sayin’, norquist’s logic included a hidden assumption — “and we manage the economy at least well enough that the investor class thinks we’re better than the other guys” — which the repubs extravagantly forgot to sustain.)

  31. winterspeak writes:

    JKH: Thank you for keeping me straight.

    SRW: The cheques can be metered out until aggregate demand is restored, and can go to everyone. Raising a tide to life all boats is not complicated, and it can be done in dribs and drabs. Mosler likes payroll tax holidays, but there are other mechanisms for this also. The nominal crises *is* easy to fix, and writing cheques to bankers does not fix it. Aggregate demand is being supported on the back of unemployment, which is a terrible way to do it. But that may not be the worst actions of the past 18 months.

    We are on same page re the long term bad consequences of malign incentives.

    I have not pieced together the connection between savings and investment beyond the identity, S=I. I know that money in the bank is not required to fund loans, the loans create the money in the bank. To that end, I guess that if the loan works out, the money remains (ie. you get savings) while if the loans do not work out, the money vanishes, and what was thought of was “investment” was merely a consumptive transfer from party A to party B. If this is true, then savings are merely the record of real investment, that is, real investment leaves a trail of savings in its wake.

    Is this so different from “winterspeakian” savings, income minus spending? I’m not sure. The loan pays the salary of the entrepreneur, and to pay that loan back he needs to reliably generate cash to service that debt. Paying down debt *is* winterspeakian savings, and lo, we find that your “good” savings and my “hoarding” are two sides of the same coin. I am not at all certain that the line you’re drawing is real.

    Most dangerously, if you believe in the above, you may have found a mechanism by which a growing economy needs to run chronically *higher* deficits. The more successful investment you have, the more savings you must account for, and thus, the larger Federal deficits you need to fund that as a matter of identity. I’m certainly not here yet, but if deposits do not create loans, then savings do not enable investment.

    I also hear what you are saying re: moral hazard. But there are two ways to skin that cat. There may be places that just aren’t appropriate for market discipline, and bank liabilities may be one of them. Certainly we know that without FDIC, simple demand deposits are inherently unstable. Why stop there?

    If you’re arguing for permanent ZIRP, up and down the yield curve, I think that’s less crazy than I did once upon a time.

  32. ws — I think we are basically on the same page re the nominal crisis and how it could be fixed (and how it cannot really be fixed). And I’m glad, and not surprised, that you are worried about longer term bad incentives.

    A couple of things: When thinking of savings and investments, I don’t think very much about banks or money. Suppose I want to invest in Winterspeak, Inc. I have claims on a mutual fund. I exchange them for clams on Winterspeak, Inc. The fact that, for the moment, I must instantaneously go into and out of “cash” is an implementation detail of the transaction. Winterspeak, Inc. takes the cash and buys a roboWinterspeak, so your brilliance is not lost to the world even while you sleep. The seller of the robot purchases a share the mutual fund. On net, I gave the robot-maker a claim on a mutual fund in exchange for a robot, which I then traded with you for a claim on Winterspeak, Inc. What I hope to realize from my claim on Winterspeak, Inc. is not a dollar return, but a real return, even if, as an implementation detail, my payoff will be instantaneously in dollars, before I save it in some other worthy investment.

    The quantity of dollars, or bank deposits, in these scenarios can be constant, as long as it is larger than any individual transaction requires. The quantity of government obligations held by the public need not grow, if its purpose is only to serve as a medium of ephemeral exchange. There might be transactional frictions that imply some convenience in letting the quantity of one grow with the scale of transactions, some rigidities in “velocity”, but these are entirely second order. There is no investment need for anyone to ever hold claims on the government in significant quantities, no reason why the private sector should hold cash or Treasuries, or insured deposits, except for transactional convenience. The reasons why the public does hold claims on government beyond transactional convenience are precisely as insurance, to avoid investment risk, or because the government/banking-system is offering real returns that make hoarding government claims an idiosyncratic-risk-free substitute for real investing.

    Paying down debt is not “Winterspeakian savings”, in the sense of hoarding government obligations. If, oddly but still very commonly, debt is denominated in dollars, then Winterspeak, Inc. paying its debt involves instantaneously converting goods and services to dollars, delivering them to me with an appropriate return as compensation for my robot advanced, following which I purchase a claim on some nongovernment investment vehicle. Individually, you have paid down a dollar debt, which feels to you analogous to putting dollar savings in a bank, because you are not so deluded as to think there is very much special about the number zero. But in fact, the position of the aggregate private sector with respect to the public sector has not changed at all when you paid me your debt, unless I hoarded the money as cash instead of investing it.

    Again, there need be no connection between savings, investment, and the aggregate indebtedness of the public sector to the private sector. The public’s desire to hold government obligations is a portfolio choice that the public sector can accommodate or not, at whatever price it chooses, so long as sufficient “money” is made available to mediate ephemeral transactions. If the government imposed a steep tax on holding money or government claims, transactions could still be mediated by dollars, but portfolio balances would shrink to near zero. (Imagine a tax of 0.5% per day — far less than the credit card fees that are worth convenience of transaction to the private sector.)

    I think the line I am drawing is very real. I think that money, as an asset rather than as a unit of account, hardly needs to exist at all, and technological change renders it ever less important. Except that the public loves assets with no downside risk and a positive real return, and so demands bank/Treasury obligations. That’s demand we should not much accommodate.

    Re moral hazard, I agree we’ve given up on small depositors disciplining banks with deposit insurance, and recently we’ve given up on large depositors disciplining banks with TBTF.

    But that’s not the moral hazard I’m talking about. It is individuals and firms in the nonfinancial sector who require discipline, who suffer from moral hazard. They (we?) are failing to tend to their real investment responsibilities, because the government offers investment insurance cheaper than free. The government/banking-system agglomeration is the soon-to-be insolvent insurance company, selling this insurance in a manner that changes the behavior of its customers in a manner that harms the insurance company’s investment portfolio, so that when its customers’ claims for real good and services come due, the government may not be able to deliver.

    Again, investors in government securities are the insured, and citizens (and China, etc.) are the customers who are led to misbehave by overgenerous insurance. The risk that is insured is loss of individual purchasing power, but by not tending to private investment, the insured are increasing the likelihood that the loss occurs in a manner that exceeds the government’s capacity to pay claims. That is the moral hazard about which we should be concerned. (I think we should be concerned about the moral hazard of banks too, but the government/banking-system’s discouragement of discriminating risk investment is the mother-of-all moral hazards.)

    I’m not really arguing for permanent ZIRP, because that implies low borrowing rates as well as low lending rates. I’m arguing for rationing of the right to lend to the government, rationing that in quantities sufficiently small that the government needn’t pay a real return on the funds it borrows. (I’m also arguing that the government not pretend it requires private lending to “finance” its expenditures. I think we’re on the same page about that too.)

  33. reason writes:


    it is perfectly reasonable for “insured” savings to give a positive return (particular if you see government borrowing as paying for infrastructure investment – as it for instance must by law in Germany) so long as returns elsewhere are greater. The real question, is why the return on investing in businesses in the US has got so low.

  34. reason — to be clear, i don’t think that government investment offers no or negative returns. i think the returns to government spending can be very high, actually.

    but conditional on government keeping its promise of price stability, lenders to the government don’t bear any downside risk on their “investments”. they are not, in fact, supplying risk capital, it is like they are buying an option that will certainly be in the money. usually the writer of an option is paid for the service, even when its eventual moneyness is in doubt. not so for “investors” in government debt that offer a real positive return. the risk bearing investors in government projects are the equityholders — taxpayers, as JKH has pointed out — not the lenders. to the degree there are financial returns on risky government investment, those should be use to defray tax costs, not to pay zero-risk lenders.

    all that said, it’s worth noting that historically, short-term US treasury debt has not offered much of a real return, so in a sense the price part of this argument is moot. as you point out, as long as alternative investments offer sufficiently high return at moderate risk, there needn’t be much enthuasiasm for lending to government. [the question of longer-term government debt is trickier, in that lenders do bear interest rate risk but not credit risk, which from their perspective requires compensation. the question is why a government would pay this compensation, when it can tax instead of borrowing and return the taxed funds (in lower future taxes or in public goods and services) as it sees fit.]

    there is no mystery to me why US aggregate investment returns have been abysmal. 1) quantitatively, we have had an investment boom during this decade, as foreign capital poured into the United States. even under the best of circumstances, diminishing returns would have kicked in; 2) qualitatively, the investment was skewed towards noninformation “safe” investment, which does not creatively discriminate to find hidden high return, productivity-maximizing projects, but instead flows into Treasuries and whatever Wall Street can mass-produce as AAA or low-risk, primarily housing and financial sector leverage. so we had an investment glut targeted at crappy projects (and destabilizing projects, because financial sector leverage tempts gamblers). the results should be a surprise to no one.

  35. dave writes:

    Don’t have time to write all I want, but wanted to hit on one thing.

    I really only care about past misallocation directely in that I don’t think people should think we are poping back to pre-crisis levels of GDP and consumption right away. It is not just a nominal crisis, the US did some real damage to its capital base (not maintaining infrastructure, selling some of it of to China, shifting it to now defunct industries such as finance, etc.) and as such previous levels of “output” were only maintainable so long as we imported capital from the rest of the world. Until we go through a period of savings and capital building, we can’t reach our old level of output. I say this mainly because I encounter all sorts of arguements about the “output gap” as if the shortfall between GDP last year and this year is a problem that can simply be solved by pulling monetary and fiscal levers. Not so much, physical and human capital needed needs to be rebuilt after a period of squandering.

    Now to tie that in with the main point. We need to restore our capital base. So let’s say american’s start to do the right thing and save money so we can rebuild. Our previous method for channeling savings into capital accumulation is broken. We need a new one. Without it, we can’t turn savings into economic growth.

    Personally, I don’t think we can come up with a new method for growth until we address past indiscretions. First, because past actors have become a huge part of the process of rebuilding our financial system, and their entrentched interests and history distorts the process. Figuring out what rules the new system should have is hard enough without a billion dollar banking lobby in washington. Second, because continued existance of the current regime crowds out new actors. Why start a bank to compete with one with a government guarantee? Lastly, one of the biggest losses we face in this crisis is trust. Trust in the financial system is as very precious asset. No financial system we come up with, no matter how sound it is in theory, will work without trust. If the perpetrators of previous fraud are not punished I don’t see how you’ll ever build trust again, and I don’t think that is the publics fault.

  36. dave — great comment. i don’t know how hard it would be to develop a productive capital base sufficient to support the lifestyle to which we have grown accustomed, if foreign financing / dollar support disappeared. it may become an empirical question.

    we have plenty of “capital” from the recent period: it’s not that we quantitatively didn’t invest enough, but that we qualitatively invested poorly, e.g. by overproducing houses. what we’ll miss before we miss imported capital is imported consumption capacity that masqueraded as capital. GDP will mechanically fall from the depression in residential construction, but those of us outside the construction/real-estate industries will feel is that it’ll get harder to make or borrow a buck, or at least harder to make or borrow a buck that buys very much.

    as you emphasize, we need to restore a capital base that, despite being flush quantitatively, is qualitatively all wrong. a high savings rate isn’t enough, if for example we use TALF to persuade savers to keep investing in houses. we may (or may not, that’s up to others) require a high domestic savings rate, but what we really need is a means of using investment capital wisely, of channeling savings to productive uses.

    i’m with you wholeheartedly on the last paragraph. one underappreciated aspect of the recent boom is that americans’ low savings rate was rational. ex-ante, returns on savings and the cost of borrowing were very low. ex post, actual returns on invested capital were negative. it was smarter to buy a plasma in 2006 than to buy an index fund.

    until americans perceive we have a financial system that can offer sustainable returns by funding productive problems, substantially all of our savings will be defensive hoarding of government obligations, which will eventually crack under the load. americans considering saving by investing in risk capital now have no idea who to trust, since as you point out, the intermediaries vying for their business are the same intermediaries who only just squandered unimaginable wealth, and who have not been punished, reformed, or reorganized for doing so. you’d be a chump to trust your funds to “high-reputation” wealth managers now, you’ve got to either be very selective, or select your own risks by purchasing claims on specific firms or projects.

  37. winterspeak writes:

    SRW: This conversation has probably moved on, but I’d like to challenge your assertions that winterspeakian savings is something different from your savings, and that the way you’re thinking about real vs nominal assets is correct.

    Firstly, “winterspeakian” savings is the common sense definition most people have of savings: they enjoy a certain income, and make certain expenditures (for requirement or choice), and what’s left over is saved. Savings = income – consumption minus investment minus taxes. Winterspeakian savings is exactly the same as paying down debt in the sense that you are increasing your hoard (decreasing your negative hoard) of Government currency. I don’t think calling them obligations is helpful, because they don’t obligate the Govt to do a damn thing.

    At a sectoral level, the private sector *cannot* increase savings (paradox of thrift). It enters into debt deflation as balance sheets collapse to zero. The only way they avoid this fate is if the public sector levers up, and gives the private sector the money it wants to hoard.

    Second, I’m struggling with your 3 way trade example. It seems that all you’ve done is describe a world without credit, where money simply acts as an enabler of barter, easing coordination problems and such. This is fine, but it does not resemble the fiat world we live in in any way. It’s much closer to the gold standard world, and hoarding *is* extremely toxic in that environment. But It is completely benign in this one (assuming that the currency issuer knows which end is up — asking perhaps too much I know).

    I’m also not clear why you claim “the public loves assets with no downside risk and a positive real return, and so demands bank/Treasury obligations. That’s demand we should not much accommodate. ” Fiat currency has large and obvious downside risk and return can certainly be negative. The dollar, reserve currency of the world, has lost 90% of it’s value in a mere three generations. And dollars stuffed in mattresses in 1900 are pretty much worthless now, as are dollars kept in FDIC accounts, particularly if you want to buy a house, college education, healthcare, etc. etc.

    Let me ask you think question to clarify my understanding of how you think about real goods and services: are imports good or bad for a country?

  38. ws — re imports, i think it depends. imports are good in terms of present consumption, but if they deter capital development in the importing country and are not indefinitely sustainable, than they can be bad, ultimately both for the importer (who is left with a depleted capital base) and for the exporter (who forewent current consumption that is ultimately not paid for in equivalent goods and services). debt-financed imports certainly provide short-term benefits to the importing country, and in theory, a well-managed importer could foresee future capital needs and avoid long-term costs. that would be a clear win. but in practice, countries with the luxury of debt-financed imports are often behave myopically, with bad consequences. although there are exceptions (debt-financed imports of tradables-producing capital goods, or asset-financed imports by an aging nation), as a rule of thumb, imports are much better in practice when trade in current goods and services is balanced.

    i think we talk past one another a bit, but i stand by my claim that deleveraging needn’t imply “winterspeakian savings”. private-sector balance sheets can be arbitrarily leveraged to one another without involving the government at all. if i owe you $1000 and you owe me $1000, and i have $1000 cash too, i can pay you, and then you pay me, and we have both deleveraged, while the public sector deficit has not increased. winterspeakian savings (ie hoarding govt obligations) must occur only when the aggregate private sector deleverages with respect to (or simply lends to) the public sector. but not otherwise.

    the private sector cannot increase “savings” only if savings are defined as holdings of government obligations. if building a factory counts as savings, the private sector can certainly increase savings without any government accommodation at all. balance sheets, even dollar denominated balance sheets, needn’t collapse to zero. i can use dollars as simply a unit of account, even if none of the actual asset exist in the world. if there were literally no dollars, transactions would be difficult. but as long as there are enough dollars to cover any given transaction, the quantity of dollars places no constraint on any particular private sector entity’s balance sheet. you are right that the private sector as a whole cannot net deleverage without govt accommodation, but that’s because the world is divided into two sectors, and there’d be no one else for the private sector to transact with. but the quantity of overall debt in the economy does not depend upon the government. in the recent past, the private sector levered up much much more than it borrowed from government. the public sector is being forced to lever up to counter the effect of intra-private-sector deleveraging on nominal GDP.

    fiat currency almost never has non-negative real return if placed in a mattress, but government guaranteed assets have had reliably positive medium-term returns. if the gov’t can maintain a price-stability commitment of 2% inflation and offers securities paying at least 2%, we have an asset with no downside risk and positive real returns.

    you are right that, in fact, there is no such thing as a guaranteed investment, and government price stability commitments are always contingent. i think we will see a US fiat money breakdown, and it will take the form of violating the public’s “risk free” expectation with inflation higher than bond returns. but ex ante, that is not what the public expects, nor what an investor extrapolating from past experience would expect.

    (in the 1970s, there were 6 years when short-term gov’t debt failed to outperform inflation. also 1980. also in the mid 30s and WWII related in the 40s. but otherwise the 50s, 60s, 80s, and 90s all offered consistently positive real returns to T-bills, and long-term real returns to T-bills have been positive. real returns on longer-term gov’t debt — bonds — are year-to-year volatile due to interest rate fluctuations, but holders of LT gov’t debt over time have earned substantial real returns. holders of FDIC-insured bank CDs would have also earned positive real returns, although holders of insured checkable deposits would not have.)

    the public believes that govt securities (and bank CDs) preserve purchasing power at zero risk. (at least the public believes that of TIPS and govt securities whose nominal return is >= ~2%) that’s a bad promise for government to make, but it is an expectation that govt has intentionally cultivated.

    i’ll agree with you that the phrase “government obligations” is an oxymoron. (“private-sector tax anti-obligations” might be more accurate.)

  39. Hi, nice post. I have been pondering this topic,so thanks for sharing. I’ll definitely be subscribing to your blog.

  40. winterspeak writes:

    SRW: My question about imports was to see how you separate real vs nominal. Based on your answer, we don’t see this distinction in the same way. If you look at real goods and services, physical stuff and work, then imports are always a real benefits, and exports are always a real cost. You can make sophisticated arguments (as you do) about long term capital erosion etc., but fundamentally, if you ignore money, stuff coming in is good.

    I’m not sure why “not being sustainable long term” makes imports any less good? Maybe this is an extension to your capital erosion argument (which I am sympathetic to, btw.) In real terms, imports are good for as long as you can get them. Long term sustainability, ie. exporters won’t remain stupid forever, doesn’t change that dynamic.

    Not sure I know what you mean by debt financed imports. Is it households going into debt to consume goods that happen to be imported? Or a nation running a budget deficit AND trade deficit at the same time? My question was very much focused on the real economy remember, financing does not come into it.

    The private sector as a whole cannot delever, although private sector balance sheets can shrink as a whole can absolutely shrink to zero. Your example actually makes it clear why this must be so.

    In your initial state, the combined balance sheets add up to $3000, $2000 of which is credit, and there is $1000 of equity (the initial cash).

    In the final state, the credit gets paid and cancels out, so the combined balance sheet for the sector if just $1000, all of which is equity.

    My point is that that $1000 of initial equity *has* to come from somewhere, and the place it comes from is another sector, the non-private sector, AKA government. In a fiat world, Government currency primes the pump. Now, if the private sector has $1000 in equity, the Govt balance sheet must have negative $1000 in equity, and this is why I resist JKH’s characterization of the Govt having positive equity in its combined balance sheet.

    If that equity was not there, the private sector balance sheets would simply collapse to zero. Private sector balance sheets *must* net to zero.

    Also, note how the private sector *cannot*, as a whole, increase the amount of equity it has. It began with $1000, and it ended with $1000. For it to get that number about $1000, it needs another sector to lend to it.

    Building a factory does not count as savings. That’s investment. Buying a car does not count as savings either, it’s consumption, or maybe investment too depending on the car. Buying gold does not count as savings, it’s investment as well, but might be consumption if it’s jewelry. Paying taxes does not count as savings no matter how you spin it. You save money when you don’t spend it. This is not some crazy definition I came up with, it’s how it’s defined in then national accounts, and how it’s understood by everyone. Try running a poll:

    1. If you get some money and spend it on building a factory are you:

    [ ] saving the money?

    [ ] investing in a factory?

    2. If you get some money and don’t spend it, keeping it in the bank instead are you:

    [ ] saving the money?

    [ ] ?????

    I don’t mean to be snippy, and I’m also happy changing “savings” to “hoarding” in the above example, but I think it would be really helpful to define savings as what it is — not spending.

    I’m using a narrow definition here, and certainly not claiming that hoarding defers consumption, or does any of the positive things people associate with savings. But money that comes in, and does not go out, is money hoarded.

    Private sector balance sheets can lever to any degree, and provide investment for any number of factories, and credit for any quantity of consumption. But, at a sectoral level, they must always sum to zero. The private sector is totally self funding in this sense. But it can also collapse to zero. For the private sector to have any equity at all, that needs to come from another sector, which could be the Government, or it could be some fixed third party, like Gold. Under a gold standard, with credit, a credit collapse would limit the amount of money in the world to the amount of gold in the world.

  41. winterspeak writes:

    One last point — I think I see now why you say “paying down debt” may not be the same as winterspeakian savings. If you just cancel out credit obligations, then yes, that is not winterspeakian savings, but it does reduce oustanding debt.

    Maybe winterspeakian savings is more accurately described as private sector capitalization? The private sector, as a sector, cannot increase its absolute equity.

  42. ws — we live in such different frames of reference! you chide me for being overly “nominal” wrt imports, but i think you are being overly nominal wrt balance sheets.

    the private sector requires no government scrip to have equity, even if we measure balance sheets in dollars. equity and cash/govt-securities are two completely distinct concepts. if my firm owns two cars, one outright, one financed with a nonrecourse loan (so its debt can be said to be one car), the firm has a equity of one car. i can redenominate the balance sheet in dollars, claim each car is worth 20K, and i have 20K in equity.

    this isn’t at all unusual. i prepare a balance sheet for my dormant consulting firm each year as part of a tax return. i’m too lazy to look, but it has maybe 40K in equity, and maybe 300K in cash or govt securities. that the balance sheet is written in dollars is a just a choice of units. (it is a “choice” required by the IRS for US tax purposes, but if my firm had significant operations elsewhere, i might be required to duplicate the same economic balance sheet in pesos for some other tax authority.)

    the private sector’s net equity is the value of the real economy privately owned. it is not zero. the balance sheets of firms tautologically net to zero, but the “balance sheets” of households do not, in any meaningful way. household equity is not an obligation to anyone, and can’t be netted away.

    neither balance sheets nor income statements nor transactions depend upon any form of government scrip. the private sector can increase its absolute equity. in Arnold Kling’s fave metaphor, begin with a seed, plant a fruit tree, harvest the fruit, then plant two more. the private sector has self-capitalized, in a way that is visible if balance sheets are measured in dollars or any other tradable asset.

    you are right that most people view savings as saving money, and view claims on government scrip as money. but that view is not what economists mean by savings, and this is one of the few places where i think it’d be better if the general public learned its lesson from economists. actually they have, kind of. most people do view their 401Ks as savings, but if their “money” is invested in an equity fund, their “savings” consists not of dollars (although they might think of it that way), but of claims on factories and fleets and the dance of business organization. in economic theory, savings and investment are an identity, savings is investment. in real terms, we’ve created financial investments that support consumption, or that might provoke no real economic consequence at all, like credit-card receivables and government debt. but those are and ought to be aberrations, you have canonized as “savings” the aberrant practice of putting money in a government-affiliated bank. but the bulk of people’s savings, at least until very recently, was not held as claims on government or claims on banks. it was held as claims on capital or future labor.

    it is simply a bad idea to define “savings” as the odd subset of the category that takes the form of unallocated claims on government. it has a certain intuitive appeal, because naive savers think they hold “dollars”. you are right to plumb deeply the consequences of the world in which we live, where people demand “dollars” and government can supply it. that fact does have real effects, and creates real economic opportunities and traps with respect to how the government issues its freely conjurable scrip. but you need to be clear about why it would be useful to call holdings of that stuff “savings”, and segregate it from “investment”. i’ll not let you get away with it in our conversations, i’ll make you qualify holding scrip as “winterspeakian savings”.

    i will grant you a distinction that i sometimes use: it is one we’ve much discussed. “savers” are primarily motivated by hedging, by minimizing the risk of losing consumption power as they move into the future. “investors” are primarily motivated by seeking positive real returns. but even here, the distinction is slippery, a matter of degree rather than kind. savers do seek positive real return, if they can get it at sufficiently low risk, and investors do avoid risk as much as possible while seeking good returns. for both groups, the question of whether to hold govt scrip of any sort is only a portfolio allocation decision. cash, T-bills, bank CDs, etc. are just ownable assets. to those who have faith in a govt price-stability commitment, govt scrip is just a very low risk, modest return asset, so “savers” tend to have portfolios disproportionately in those assets, while “investors” gravitate elsewhere. but there’s a continuous terrain between, and fundamentally the distinction is arbitrary.

    re imports, i suspect in actual conversation, we’d agree more than we’d disagree. i’m happy to concede that real imports are better than no real imports, ceteris paribus. however, bundling real imports with a short position on promises (whether denominated in dollars or any other thing) break the ceteris paribus, and such an arrangement is not necessarily a good thing.

    unsustainable real imports might be a bad thing too, even with no promises asked, if they break ceteris paribus by altering our behavior in maintaining the capacity to produce. but if we can maintain (however dormantly) our full production capacity, then sure, we might as well enjoy costless temporary consumption (or investment), and be no worse off when the good times end and we have to get by on our own. unfortunately, we tend not to maintain our capacity to produce while others temporarily supply, we are not collectively good long-term optimizers in the way that simple theory holds we should be. if our long-term optimization capability is sufficiently good, then we clearly gain from temporary real imports with no strings attached. however, if we are too myopic, temporary free imports can paradoxically provoke long-term harms. it’s hard to say no to a subsidy, but poor decision-makers can be better off when they are constrained than when they are free, and that yukky paternalistic argument might apply more strongly to national economies than it does to individuals.

  43. winterspeak writes:

    re: imports “however, bundling real imports with a short position on promises (whether denominated in dollars or any other thing) break the ceteris paribus, and such an arrangement is not necessarily a good thing.”

    Promises in dollars are worth the paper they’re printed on. The US can produce any quantity of dollars it wants at any moment without needing anything from anyone. China’s been exchanging real stuff for an excel entry. Good luck to them!

    Economists, not that I think much of them, define savings the same way I do. “personal saving has been defined as personal disposable income minus personal consumption expenditure.[Keynes] In other words, income that is not consumed by immediately buying goods and services is saved.”

    “The private sector requires no government scrip to have equity, even if we measure balance sheets in dollars. equity and cash/govt-securities are two completely distinct concepts. if my firm owns two cars, one outright, one financed with a nonrecourse loan (so its debt can be said to be one car), the firm has a equity of one car. i can redenominate the balance sheet in dollars, claim each car is worth 20K, and i have 20K in equity.” A firm with two cars has two cars (real). From a nominal perspective, it’s balance sheet is equally large, but more levered because of the financing. When we move into the world of finance and Govt obligations, we’re talking entirely about nominal things.

    Someone has a house. That’s real. Them being upside down on the mortgage is a nominal problem, not a real problem. Destroying a house to solve a nominal problem is vandalism.

    With respect, Arnold’s fruit tree metaphor was awful. Arnold does not understand fiat money. But if you found light in his metaphor, more power to you!

    I’m not denying that the private sector does not have real stuff. I’m simply pointing out that nominal claims net to zero, with any remaining residual being exactly equal to Govt debt. This is a fact of accounting.

    The distinction I’m making between what counts as savings and investment is straightforward, and useful. Savings is no one’s income. Investment is someone’s income. Money allocated to savings decreases velocity, money allocated to investment contributes to velocity. An increase in savings, at a sector level, is a decrease in income. This matters.

    Have I made myself clear on why money directed to investment (someone’s income) is different in important ways to money directed to hoarding (no one’s income)? Do you also agree that this distinction is real, useful, and important? Remember, banks don’t need deposits to make loans, so people hoarding does *not* enable investment.

  44. winterspeak writes:

    Steve, I also wanted to apologize if I come across as being curt. If anything, it’s because I have too much to do and cannot spend the time I’d like to thinking about your ideas.

    If you’re in the bay area ever, let me know.

  45. ws — yeah. i so think that the distinction is real, useful, and important. i like the idea of “winterspeakian savings”, but don’t like to conflate with savings broadly, which can be attached to investment.

    i’ll agree that economists use “savings” they sometimes mean what you mean, and sometimes mean what i mean. when Keynesians make “paradox of thrift” arguments, they always mean what you mean. but when economists write S=I, and when they calculate the “savings rate”, they mean what i mean.

    it’s an accounting identity that private sector financial claims net to total public sector debt. it’s also an identity that all non-steve financial claims net to net steve debt. so what? the accounting identity only matters to the degree that the debt of the excluded claimant is uniquely desirable. which it is in the case of government (which is why i enjoy your view of the world, and find it useful to consider). it is that desirability that gives the accounting identity (and the entity that gets to control the numbers in that identity) its power.

    however, the accounting identity doesn’t imply that private balance sheets net to public sector debt. you draw too deep a distinction between real and the financial. financial statements are descriptions of real things, and financial claims have meanings in terms of real things even if they are denominated in dollars and by definition net to zero. my futures contracts on gold are dollar denominated financial claims, but (assuming no chicanery) what i am owed in real terms is entirely insensitive to the government’s net debt. the government could change the real value of the dollars i’ll have to surrender for the gold. but i could have funded my long in gold with a short in silver, in which case there would be four zero-netting USD claims that would leave me entirely indifferent to the fate of the dollar. these are dollar-demoniated, but they are real claims, and they make up much of the financial world.

    a simpler example is common stock. a share of common stock is a very nebulous, net-zero claim, but whatever it is a claim on is not nominal dollars.

    i agree with you that nominal games can affect a lot, both in nominal and real terms, which again is why i do like thinking about things your way. given the present circumstance, where demand for government scrip is unusually high, i think that your analysis is useful and many of your suggestions would be very practical (although there are longer term consequences that require some thought).

    with respect to debt and imports, you are right that the short position in promises we have taken along with our imports is written in scrip we can infinitely produce. but i think you underplay the degree to which people’s understanding of the meaning of promises has effect beyond the letter of the contracts upon which they are written. i think that the US can (and should) devalue those promises a great deal. but just because printing dollars is costless doesn’t mean that it is costless if promises are perceived to be broken despite contracts fulfilled. from one perspective, the US was savvy (or lucky) to get away with buying goods with costless scrip. but that’s a zero sum game that our unsavvy partners ought to be but may not be prepared to concede losing. and, from another perspective, even if we get off easily and extinguish the promises without rancor, we were not, in my opinion, wise long-term optimizers during the period of cheap imports. we let a subsidy beguile us into poor choices, and i think we will bear some costs for that.

  46. reason writes:


    I dispute your claim and that of the Austrians that there has been a massive overinvestment in housing (never quantified). Look at this chart:

    housing starts

    It is the price of LAND shooting up that is the bubble not the very weak (despite massive fiscal stimulus) real economy. It asset prices that tell the story. But why is all the money going to speculation rather than real investment?

  47. reason — that’s a pretty good peak in starts, even though it’s not unprecedented. plus there’s more to residential investment than starts — see residential investments as a fraction of GDP here. it’s the tallest peak in four decades, and it’s understated in the graph, because GDP has grown rather dramatically in real terms since other large peaks in the seventies. (i think a graph of real residential investment would look quite dramatic relative to past peaks, but i can’t seem to navigate the BEA site to the historical data, and my friend FRED — data from the st louis fed — is mysteriously down.) why the difference? lots of renovations, upgrades, conversions, etc that don’t count as starts.

    i don’t disagree that speculative rises in land and existing home prices are a lot of the housing bubble story. but there was definitely a sizable boom in construction, renovation, etc, and I do think there was substantial “malinvestment” within that boom.

  48. reason writes:

    My reading is that the peak (never so high) has already been offset by the subsequent crash. And given the relatively low level of industrial investment, why shouldn’t resources be allocated in that direction? The problem is not that people didn’t want to consume more housing, but that they bought it at too high a price. I stick by my view that speculation in asset prices is the main story here.

  49. reason writes:

    Thanks for your rapid reply by the way, and I think I should further add that it was clear to me that something was VERY wrong, before the “house building boom” even started.

  50. Hi, interesting post. I have been wondering about this issue,so thanks for sharing. I will definitely be coming back to your blog.

  51. dave writes:

    Misallocation to construction is only part of the story. The credit bubble created distortions in all number of sectors. For one all those home equity extractions fueled boom in consumer goods that was also unsustainable.

    Mish had a pretty good post awhile back outlining all the different sectors effected by the housing boom/credit bubble. Can’t find it quickly right now. When you really think about it nearly every sector was affected.

  52. winterspeak writes:

    SRW: The connection of savings to investment is *exactly* the nexus that I’m trying to explore.

    In the typical model, and I used to think this as well, savings would be capital allocated towards investment, so S->I. Now, “I” could be inventory, so you get high savings and cars piling up on Long Beach, but the causality driver was clear. First you get savings, and those savings fund investment.

    Your point, I think, is that many forms of savings (ie. bank deposits) don’t drive investment at all, or do so in ways where the saver is abdicating all of the due diligence an investor is supposed to do (to justify their compensation). Why does a “saver”, who just puts their money in an FDIC checking account, deserve anything other than a 0% return? We want more minds thinking hard about the future and making asset allocation decisions, not fewer.

    My contention is that the causality is reversed, S does not drive “I”, instead, I->S. If you have “I”, “I” that works out, then that creates S.

    The S in S=I is the same as the Keynesian S. S=I comes from National Income Accounting, which defines GDP = C + I + G + (X – M). Winterspeakian savings is GDP – C – I – T = G – T (X-M). GDP (income) – C – I – T is exactly the same as income minus expenditure, and distinguishes between investment and consumption (which is someone’s income) and hoarding (which is no one’s income).

    Am I missing something?

  53. ws — the two definitions come from the same identity, but are different. In the simple, closed-economy case is a very different statement than:

    S = I

    S = G-T= Y-C-I-T

    Both are definitions, equally valid in a sense. We are quibbling over which best captures what people mean by savings, both among the general public and among economists. We both have a case, descriptively, but I’ll cling tenaciously to my definition for normative reasons.

    We are getting to less semantic issues here:

    My contention is that the causality is reversed, S does not drive “I”, instead, I->S. If you have “I”, “I” that works out, then that creates S.

    My view would be


    I think you are quite right that bank-intermediated savings does not meaningfully drive investment at all. (That’s a bit exaggerated, but to the very limited degree that it does drive investment, it drives it badly. Bank deposits do correlate with bank lending to some degree, but there is no reason to think that bank deposits correlate with good investing, since insured depositors don’t care.) Putting aside the parenthetical, I’d agree that as far as bank-financing is concerned, the causaility is mostly I->S. Banks will pay somewhat higher interest rates and cause a bit more savings to avoid the (usually mild) costs and risks associated they face for lending with insufficient deposits. (There are costs because they may have to borrow on the wholesale market, and manage around very weak reserve requirements.) Bank investing is only very gently coupled to savings at all, but to the degree it is, I->S.

    But other forms of (Interfluidian-but-not-Winterspeakian) savings do involve the saver as an investor, so that the decision to save also impacts the likelihood that some investments will be made. It’s a two way street: potential investors won’t save if there are no good prospects (they won’t even Winterspeakian save, if they doubt the quality of the price-stability commitment). So I->S. But projects that otherwise might not have occurred will occur, because people with more wealth than they need to consume are enthusiastic (see, e.g. dot-com era). So S->I.

    So my view is that in the general case, it is S<->I, while in the traditional bank sector, it is I->S. In the “deep non-bank intermediation” sector, which I’ll define as nonbank, nonstate intermediaries that do not substantially involve the saver in the informational process of selecting investments. Triple-AAA securitizations are the poster child for this sector, but index funds belong as well. For this sector, I think you need to add a third player, and all causality is bidirectional:

    financial intermediary<->I<->S

    If I could format it, there would be a bidirectional arrow also between financial intermediary and S (and I’d have arranged everybody as a triangle).

    That is, I don’t think the ex ante investment prospects of housing drove the flow of savings into MBS. There was savings interested in nonbank/nonstate investment products, financial intermediaries actively worked to create products for this demand, that led to an increase in residential investment (because mortgages were good fodder for securitization), the success of which helped to draw more savings. (A similar story could be told about the development of index funds and the stick market, I think.) Neither an increased-investment-prospects-draws-savings, nor an increased-savings-drives-investment story is sufficient,. The role of financial intermediaries was not passive, the choices they made when “innovating” affected both the allocation and quantity of savings and investment.

    Are we getting any closer?

  54. reason — FRED is back, so do check out this graph of real residential investment. The investment boom was not subtle. I agree with you that it was in part due to a lack of good investment opportunities in other sectors, that it was partly a “tallest pygmy” story in a time when the US was awash with financial capital seeking returns. And, per my comment to Winterspeak, much of why it was the tallest pygmy was because financial intermediaries knew how to fund residential investment “safely”. But all this did lead to a very large redirection of real resources to building and upgrading residences.

    You are certainly right that price inflation played a substantial role. But per Winterspeak, on the upside price-inflation just results in transfers between individuals. It is anti-price-inflation that becomes a systemic concern and produces harms. But overinvestment in residences destroyed real resources during the boom, even if many people didn’t know it. Dilapidating tracts in the exurbs, especially out West, now tell the tale.

    I won’t argue that things were “right” before the R.I. boom, although to get before the R.I. boom you really have to be in the middle of the dot-com bubble. Things haven’t been quite right for a long while now. (Maybe “right” is too much to ask for.)

    dave is certainly right to point out that poor investment was made in lots of sectors during the credit bubble. The R.I. boom is a salient and important example, but it is hardly unique. Private equity, commercial real-estate, etc were similarly overinvested. (Although with PE, it was more a matter of price inflation, the excess financial investment resulted in a lot of transfers, sometimes slimy transfers, but PE investors tended to withdraw financial capital rather than employ it, so real costs are side-effects of the fall, e.g. overlevered productive businesses fail unnecessarily, leading to less production than under alternative scenarios.)

  55. winterspeak writes:

    SRW: I think we’re getting closer. I’m comfortable saying that bank deposits don’t drive investment *at all*. It’s an exaggeration that clarifies.

    That doesn’t mean that bank *lending* doesn’t drive investment, it surely does, so I -> S is strong in the bank sector, it just isn’t driven by deposits.

    But I want to make sure I have things straight in definitionally.

    Closed economy:

    GDP = C + I + G

    winterspeakian savings = Y – C – T = G – T = what income is not expended is saved.

    S=I derivation:

    Y = income, C = consumption.

    Y-C = S(avings)

    Y = C + I by identity (all expenditure, spending and investment, is someone’s income)


    S – C = C + I


    It’s the *same thing*. All private sector expenditure, whether consumption of investment, is someone else’s income. All public spending is money creation. All taxation is money uncreation. Winterspeakian savings takes money out of circulation because having it in an FDIC account is the same as uncreating it. The Government, because it can print, is sitting on an infinite gold mine. The rest of us don’t have an infinite gold mine, but we can make a little, finite one through hoarding. If we’re China, it can get quite big, albeit still finite.

    Winterspeakian savings is exactly the same as S=I savings. It’s income that doesn’t go out again. It’s purely a nominal measure.

    I’ve actually completely lost what Interfluidian savings is. An attempt to successfully invest? Why not simply call it investment then?

    Savings is what is not spent. Let Y = income and C equal consumption. Therefore, Y – C = S. Using a national income accounts identity for AP Economics, Y = C + I. If I subtract C from both sides, I have Y – C = I. Since Y – C is Savings, I’m left with S = I. The following video shows this again.

  56. winterspeak writes:

    Pls ignore cribbed S=I derivation at the bottom of the post.

  57. ws — I won’t strongly object to your definition that clarifies. I think that at the margin, deposits do increase banks’ propensity to lend, but I think that effect is small, much smaller than the perceived quality of investment opportunities or bank-internal incentives to lend. I am comfortable, to a first approximation, with your claim that in the banking sector I->S, or financial intermediary<->I->S.

    Re the derivation, if you write an economy as Y = C + I and define Y – C = S, that’s the same as defining S = I. It is not meaningfully a “result”. It is a definition.

    (That’s arguably true of a lot of economic “models”, where models are crafted to imply a desired results. But usually the chain of deduction is murkier, so readers might think that the result was subsequent to model definition. You must be less transparent if you want to be an economist!)

    In a world where there is a public sector, the only kind of world where there could be a difference between Winterspeakian and Interfluidian savings (because Winterspeakian savings is not defined in a Y = C + I ecomony), the definition S = I is different from the definition S = G – T = Y – C – I – T.

    Interfluidian savings is income earned and but not spent on goods and services characterized (by the spender, by BEA, by whomever) as consumption. When I am making the (closed economy) definitions, I write Y = C + I + GC + GI, so Interfluidian savings would be S = Y – C – GC = I + GI. But if we go with the convention of presuming all government expenditure is consumed, then S = Y – C – G = I.

    Note that in general, Y – C – G != Y – C – I – T , because G != I + T ==> G – T != I in general. Winterspeakian savings and Interfluidian savings are different unless the level of investment in the economy and the government deficit happen to be identical. In a Y = C + I + G economy, Interfluidian savings matches is consistent with an S = I definition, Winterspeakian savings is not.

    The definitions are really pretty clear. I am defining savings as investment, you are defining savings as government deficits. Both map to conventional intuitions of savings, even though the concepts are quite distinct. Your definition maps to money in the bank or held as government paper. My definition maps to funds invested in capital goods directly, but also in paper that indirectly supports business activity or future labor. Depending on your view of banks and government (does bank/govt borrowing support production of exchange value like private sector borrowing is claimed to?), Winterspeakian savings might or might not be a subtype of Interfluidian savings. It’s a matter of definition, since investment can always include bad investment…

    Since much of the “savings rate” goes into nonbank/nongovernment investment, and since most of people’s “savings” is usually not in government paper, I’m unwilling to cede the word savings and take the name investment. I think we’ll have to live with the human ambiguity that a word can mean different things in different contexts.

    I do think it is misleading, and leads to errors in reasoning, to insist that the one true definition of savings is S = G – T, though. If you say it is impossible for people to save if the government runs no deficit or a surplus, that is not right under ordinary definitions of savings that include 401Ks. It is perfectly possible to have a high savings rate as conventionally measured and ordinarily understood and no deficit whatsoever.

  58. reason writes:


    (and this relates to your discussion with WS as well) I think the housing boom only got serious in about 2005. It looks to me like a short and narrow spike (which means that the total amount of investment is not that great). I still read the charts as saying that any “excess” investment judging by the area above and below a trend line, has already been wiped out. If we want to see that something is seriously wrong, we have to look at debt – particularly household sector debt.

    The housing bubble (post-2005), financed fraudalently, was a last desperate attempt to keep the good times rolling, I see it is something endogenous in response to underlying problems, not the source of the problems.

    Now I agree with WS that the US is lucky – they own the reserve currency and can just print it to pay off their obligations. And they should print it, and spread the money widely rather than narrowly. Some will still go under, some will prosper and some will hang on – hopefully the least prudent will suffer the most. This however, still leaves the problem of the disastrous foreign balance (which is the source of the problem in my view). If the US is lucky enough and prints enough money then maybe it can be rid of the reserve currency status. It works like an oil or gas industry (the dutch disease) to distort the real economy.

  59. reason writes:

    P.S. I think land prices are not given the importance in economics that they should have. We need to bring back Henry George and modernise him.

  60. reason writes:


    And I’m with Winterspeak in that we need to see the financial economy and the real economy as related but parallel worlds, large parts of which function independently from one another.

  61. reason writes:


    regarding the national accounting identities, doesn’t it help if you point out that Keynes made a distinction betweeen planned and realised savings?

    And further regarding the accounting identities, I don’t have the time to do the analysis (I’ll have to wait till I retire or my kids grow up whichever comes first) but it occured to me that the distinction between consumption goods and investment goods is murky at the margin (think automobiles for instance). So defining savings as investment is rather murky as far as I’m concerned.

  62. dave writes:

    One thing I think is valuable to understand is that investment must meet future demand i.e. demand from savers. Borrowers will not be able to borrow more and more forever so at some point they will have to pay back their borrowing to savers who gave it to them.

    This is important because businesses generally respond to past demand trends. They see what people are consuming today, what rate it is increasing at, and extrapolate. Investors do the same thing. This has some problems. Ideally, we would like investors to be expressing their own future consumption preferences so that the capital to produce them will be available. For example if an investor if going to need more healthcare as they age it would be nice if they invested in stock of healthcare providers so they could expand capacity to meet those demands.

    Instead, borrowers are generally driving capital allocation. Savers/investors, rather then expressing their future consumption preferences, are largely providing the capital to produce the things borrowers want. Herein lies the misallocation. After years and years of ever rising debt levels (the cause of which lies at the heart of our banking system) our entire economy is organized around producing goods to be consumed by a pool of ever indebted borrowers. No thought is given as to how those borrowers pay it back. What are they going to produce that the savers want in return and how do they produce it?

  63. In practical terms, I think that we are mostly all on the same page. I also think that the US should and will end up printing to resolve its intractable balance sheets, both internal and external, and that as much as possible we should distribute those newly printed claims braod and flat.

    I certainly don’t mean to claim the the financial realm and the real economy track one another very well; they are clearly very loosely coupled. But I don’t want to canonize loose coupling in practice as normative theory. That is to say, I think we ought to try to develop a financial system in which the connections of claims to the real economy is clearer. For the moment, we live in an odd “nominal” world, which has let the financial sector play lots of dangerous games without check, and now serves as the basis for proposed remedies good and bad to the maldistributions, misinvestments, and broken promises untethered paper has given rise to.

    I want to define S=I precisely to call attention to the ridiculousness of letting S=”loans for present discretionary consumption to people without substantial collateral”. Saying S=I doesn’t imply that all investment is good investment. But if you say “S=hoarding government paper” and that government-backed banks for a variety reasons lend in ways that are shitty when viewed as investment, you eliminate any possibility that private capital markets can steward aggregate wealth. The Winterspeakomoldbuggian thought experiment that “savers” lend to government and government lends to bank customers, may be accurate, in fact. But if we let that become a normative rather than descriptive framework, we are giving up on private-sector capital allocation and turning the allocation decision into pure politics. I do want to concede, to Winterspeak and to reason, the accuracy of the description. But I want to argue that it is not a sound prescriptive view. It is worth defining savings as investment, because that definition requires us to ask whether the savings is good investment, and induces savers to take responsibility not just for “putting aside money” (which as WS will note is like storing pencil marks), but for stewarding real capital well.

    dave gets to an idea very near to my heart, which is that savers who do not wish to be investors, in the sense that they are primarily concerned woth hedging consumption risk rather than with maximizing returns, ought at least contribute information about their future consumption needs. i don’t blame savers for not doing so, because we’ve defined no convenient institutional means whereby ordinary savers might buy, for example, future housing vouchers or somesuch. defining such things well is actually hard.

    but saving as generalized claims on “consumption”, as mediated by government paper, essentially deputizes the state to draw inferences about future aggregate consumption needs and to ensure that real resources in sufficient quantity and the appropriate pattern actually come to be available when savers wish to consume. i think that’s a bad system, since as dave and WS point out, borrowers drive state investment, and the state has incentives to encourage present borrowing regardless of its future productivity, because present borrowing increases current economic activity.

    as much as possible, we need savers to be investors. if in practice they usually are not, we need to alter their incentives. in my view, we do need to allow small, individual savers the luxury of unallocated savings, because investing is a lot of work (even planning out specific consumption, if we had ways of purchasing consumption forward, is a lot of work and requires the saver to assume risk that plans or preferences will change). but people who wish to save a great deal need to take responsibility for ensuring that the the real goods and services they will want in the future are actually produced from the capital they set aside.

    if you live in a dung hill, it is a useful exercise to come up with an accurate map of the dunghill. but it is not a good idea to declare that map a theory of the world and create a society in equilibrium with its perpetual existence. it would be better to develop a theory of carpentry, by which one might build a nicer home.

    our untethered, information-leaking, incentive-skewed, political-and-market power-based financial system is what it is, and i don’t mean not to acknowledge that. but i won’t accept a description of the dunghill that is to serve as the conceptual framework by which i understand a world whose future is still malleable.

  64. reason writes:


    I’m not too sure that savers know what their needs will be in the future (particularly as the future will offer options they don’t have now AND vice-versa). Savers just trust the system to deliver what they will need in sufficient quantities, they don’t have a crystal ball. But yes, the system hasn’t worked too well, but I’m more inclined than you to think it is the international system that has failed, that a big part of the problem is that their are NO profitable productive investments in the US because the US imports twice as much as it exports.

  65. winterspeak writes:

    SRW: I think we’re in a dung hill too. So we’re on the same page there.

    You are also quite right to call out my identies as definitions, not proof.

    To me, the key insight of this identity structure is the realization that, at the economy wide level, all spending (whether on consumption or investment) is someone else’s income. This is where Y = I + C comes from, and if you accept this (which I think you must), then I=S falls out as a neccessay logical consequence, assuming you accept the commonsensical notion of savings being hoarding (ie. no one’s income) while investment is money spent (giving someone income) in the hopes of having it return, with gain, in the future.

    This is *not true* at the household level, where savings and investment can be quite different. Macro is tricky because we think in terms of households, but the economy as a whole is not like a household.

    It follows that one individual saving does not flow into an investment. What it does is enable investment to happen somewhere else in the system (I’ll show how this happens in a moment). That is why putting money in an FDIC account does not flow into investment in any way, it just essentially takes the money out of circulation. Bank deposits *do not* fund investment. 401(k) contributions, while thought of as “savings” are not, they are investment. Income that is not spent on consumption or investment is saved, whether it’s put under the mattress, or in an FDIC account.

    To see why this is the case, let’s look at what S=I means in real terms:

    A closed economy’s real income, in any given period, is its production of real goods in that period. An economy can consume as much as it produces (Austrians add a useful and correct note about maintaining stock, a key driver in ABCT that I’m sympathetic too, but outside the narrow scope of this discussion). Suppose that economy chooses to consume a fraction of its production, the remainder is either inventory or capital stock. Inventory and capital stock are Investment (by identity). Not consuming everything you produce is savings. Thus, in real terms, what a sector does not consume of what it produces is investment, which is the same thing as savings (when viewed through the nominal lens). That is what the identity is pointing out.

    Note that this analysis says *nothing* about the quality of that investment. Unexpected inventory, like the Chrysler’s piling up in Long Beach count as “investment” in this definition, although you and I would simply classify them as waste. An overpriced house that was bought for “investment” purposes suddenly reveals itself to be consumption, and the household Y suddenly finds it’s been doing a lot of C and not much I. As it’s actual, realized I is lower than it’s expected I, it’s actual, realized S is lower than it’s expected S. D’oh! So investment can be good (new factory) or bad (unexpected/unintended inventory), just as consumption can be good (new car) or bad (fancy investment property just turned out to be a depreciating white elephant you do not really want, nor can you really afford).

    The only way an economy can have true savings is if it consumes less than it produces. That real difference needs to show up *somewhere* in the economy as savings.

    Let’s add Government to the mix. By taxing, the Government reduces the ability for the private sector to consume everything it (the private sector) produces. If I paid no tax, I’d buy more stuff, but since I have to pay tax, I have less money, so Ibuy less stuff. The Government then steps in and buys the stuff that the private sector produces, but can no longer afford. In this way, taxation enables the Government to spend without triggering inflation. Remember, the Government does not need to tax in order to have money to spend. The Government taxes to create demand for its currency, redistribute, and sterilize the inflationary impact of it’s own spending.

    National income, Y, now equals I + C + G.

    Private sector savings = Y – C – G – T = G – T. You know this story, we’re back at Winterspeakian savings. The more the private sector *chooses* not to consume/invest in (through hoarding) the larger a deficit the Govt can run without triggering inflation. It is a mystery only to Macroeconomists how Japan can run a 120% deficit, have the highest savings rate in the world, and still be in deflation. I simply ask why the Japanese Govt bothers to tax at all. (Joke: What does Scott Sumner call it when the Japanese Govt reduces all taxes to zero? Answer: Monetary policy). Also note, “national savings” is an anachronism. “Taxpayer profit” is similarly nonsensical. Government is a currency issuer, it has no need of savings, or profit.

    As I’ve defined things, S=I to the penny, and S = G-T to the penny.

    “I want to define S=I precisely to call attention to the ridiculousness of letting S=”loans for present discretionary consumption to people without substantial collateral””

    Agreed. Banks were not investing, they were transfering money to current home owners (who sold their house). I don’t think anyone regards that activity as Investment today. What they *thought* they were doing turned out to be quite different from what they were *actually* doing. Maybe we need a new word for this — invetmenting? It was unintended consumption, for sure.

    “S=hoarding government paper”” But it is. In real terms: if a sector does not consume everything it produces, something must be left over: investment. In nominal terms, if a sector does not spend everything it earns, something must be left over: savings. Hoarding paper is how you know you’ve invested (as a sector), but it tells you *nothing* about the quality or desireability of that investment. Imagine a world where every household had a zero bank balance, and all income went either to investment or consumption. Every investment would lead to some consumption until there was no investment left. *Some* real output needs to be left unconsumed for it to remain and be count towards a future period (ie. be invested), and this is not possible unless there is some hoarding.

    Also note, that allocation decisions can be entirely driven by the private sector in all of the above. Hoarding Govt paper does not mean the Govt is directing investment, it simply marks the fact that some present output is not being consumed and therefore (for good or ill) is being invested. The households who do not consume all that they produce may be different from the households who direct the investment. I save up money to buy what the entrepreneur is cooking up.

    Conflating investment (someone’s income) and savings (no one’s income) does not make sense, and I hope that my translation of S=I, particularly I->S in *real* terms makes this clear. Even better, think of hoarding as stuffing money under the mattress. Banks still lend, deposits are still created, but savings (hoarding) makes the money dissappear. Hoarding does not drive or direct investment decisions, but it creates space for investment by leaving some period production unconsumed.

    PS. In practical terms, I think we are on the same page, but where you see some actions as bad, I view them as being benevolent. Unfortunately, the Obama administration chose to fund savings the hard way — through unemployment. The economy is producing less than it was, and less than it can. This is a real impoverishment, and totally unneccessary.

  66. In practical terms we agree a lot more than in theoretical terms… we’d make very similar policy choices, but never stop jousting with our pens…

    So, there’s lots I disagree with here. In a closed economy without a government, one might define…

    Y = C + I

    We’ll let it be a barter economy, since there is no government scrip. Maybe some commodity evolves as a convenient medium of exchange, maybe it’s a high-tech world where frictionless markets and cross-exchange rates between goods and services render money unnecessary.

    In that economy, would it be true that

    Every investment would lead to some consumption until there was no investment left.

    No, of course not. There would, as you say, be some “hoarding”. In the accounts, my exchange of say gold for a television might count as “consumption”, but in fact, purchase of a television is more like investment than consumption. Nothing goes “poof” when I buy the television. To prevent some consumption black hole, all that needs to happen is people have to hold goods without destroying them.

    Let’s suppose now that there is an indestructible monetary commodity. Let’s call it “gold”. Assuming property rights are exhaustive (someone owns everything), it would be impossible for everyone to have zero monetary balances. So, some people would hold gold. But is there any relationship between the stock of gold and the fraction of other goods that is held or used in production rather than being destroyed? Of course not. Some “investment” would take the form of gold held. Other investment would take the form of lumber purchased and transformed into looms. Consumption might take the form of burning wood for heat. (The distinction between “investment” and “consumption” is a judgment call about the future value of a present use. Even burning wood is investment, if the alternative is starving to death. Output can’t be neatly decomposed into categories of what is “saved” and what is “consumed”.)

    But let’s simplify. Let’s consider a two good economy, where the goods are gumballs and gold. All exchanges are mediated by gold. There is a magic gumball machine that produces 100 gumballs each period, which are randomly distributed. I think this is fairly close to the economy you are imagining.

    People may be fickle about which color gumballs they like, so there may be exchanges of old gumballs and gold. But we don’t count that as Y. Of the output, the new gumballs, some gumballs get eaten within the period, some do not.

    Y = C + I = C + S by definition

    Does the quantity of gold have anything to do with how many end up “saved” and how many “consumed”? If it does, it certainly isn’t obvious to me. Assuming an equilibrium, probably the gold is primarily in the hands of those with a low time preference of consumption, since holding gold substitutes for holding gumballs, but only gumballs can be consumed. Eventually those with low time preference hold all the gold, so very few transactions occur. We end up in a spenders, savers equilibrium where gold is irrelevant, and aggregate savings is determined by the ratio of spenders to savers. If gumballs held yield more than one future gumball, the proportion of output saved goes to one in this economy: “savers” hoards grow exponentially, output does as well, and the fraction of output falling to spenders and consumed goes to zero. In every period…

    Y = C + I = C + S

    …but despite the fixed money balance, Y and I (or S) go to infinity while C remains constant. (A Marxist crisis of accumulation in a nutshell…)

    Now suppose that there is no gold, but there is government scrip. If the quantity of government scrip is fixed, it is identical to the gold case. However, if G collects taxes in scrip, and only issues scrip in exchange for gumballs, consumers will have to consume less, and savers will deliver some of their gumballs to the state for scrip, which script they will also deliver to pay taxes. If the government budget is balances, there is no net scrip issuance, since all the new scrip surrendered for gumballs is taxed. What effect does this have? The goverment can itself store the taxed gumballs, in which case we haven’t fundamentally altered the economy. Government will be like a saver, and we’ll have the same crisis of accumulation, but faster as some of what would have been consumed will be saved. If the government can destroy assets, and does so, the existence of a currency-issing taxing G will lead to less aggregate savings. If the government itself neither saves nor consumes, but simply distributes the gumballs it collects, the effect on savings and consumption will depend upon its pattern of distribution.

    Now suppose the government issues more currency that in collects in taxes. It does so by purchasing gumballs with currency, but not taxing back all the scrip. Savers in the economy sell gumballs for the excess scrip, as long as they believe that the scrip will allow them to avoid selling a future gumball to meet a tax burden, the scrip is “as good as gold”. From the savers’ perspective, holding a gumball or holding excess scrip are perfect substitutes. In their own accounts, this transaction has not affected savings, although some of their investment portfolio takes the form of scrip.

    But in aggregate, whether savings/investment increases or diminishes depends upon what the government does with the gumballs. If the government stores the gumballs, nothing changes. But if the government destroys the gumballs, or redistributes them to those who would consume rather than save them, savings decreases! This, “Winterspeakian savings” — a government deficit — can result in increased consumption and less aggregate savings/investment in real terms. This is an intuitive, and conventional, analysis. If governments borrow and spend their borrowings in a manner on goods and services that are consumed or destroyed in a way that does not improve future periods, or if they redistribute to those who do, aggregate savings diminished. Note that this can be a good thing: It is quite possible to save and invest too much (and if we allow for variations in investment quality, too poorly) so that more present consumption by parties who otherwise would be unable to consume is optimal. But “Winterspeakian savings”, or government budget deficits, can lead to less aggregate savings than would otherwise occur in exactly the manner that people conventionally understand.

    However, government can issue scrip to people who would otherwise consume. This seems odd: if scrip is convertible to gumballs, and I’d rather consume a gumball than hold one, why would I hold scrip? But governments can set the price in gumballs of scrip arbitrarily, and can make the penalties for not coughing up arbitrarily large, so that even those who ordinarily would prefer to consume prefer to hold scrip against future taxation, for fear that if they do not have scrip they will be badly penalized. If governments persuade those who otherwise would have consumed gumballs to sell them for scrip (as well as those who would have saved anyway), and if government stores rather than destroying, then net issuance of government scrip can increase aggregate savings.

    The lesson, I’ll say again, is there is no necessary relationship between the quantity of a medium of exchange and the fraction of output that is consumed or invested, and there is no necessarily relationship between government currency or debt issuance and the fraction of output consumed. It depends. But in practice, government deficits more often increase the fraction of output consumed rather than “marking” in some sense private savings.

    The economy you want is one where investors only hold government paper (that is, they do not directly store capital goods), and where the government holds but does not destroy the goods surrendered in exchange for that paper. In such an economy, yes, government paper would serve as a marker of private savings. Only government would hold real investments, and S = I = G – T would hold. In such an economy, storing money under a mattress implies government savings or investment, sure. But that is far, far from reality.

    If individuals can hold capital goods directly, or if they can hold paper claims directly on one another’s properties, then there is no necessary relationship between (G – T) and government savings and private investment. If government does not save or invest the goods and services it purchases with taxes (that is, if it destroys or distributes to people who increase their consumption), then (G – T) is in no sense a stand-in for investment, and the relationship between (G – T) and aggregate real savings can be (and often is) negative.

    I think you are hung up on accounting. If something is saved, it must show up in someone’s accounts as savings, right? It certainly does. But here we come back to balance sheets, and the lack of relationship between units of denomination and assets held. If I am a saver of gumballs, I may write a personal balance sheet that includes an asset of 10 gumballs, valued at $10 ea, so an asset of $100. My equity is then $100 on the other side. This is not a net-zero balance sheet, since my equity is no one’s claim. I have “savings” worth $100. But I don’t actually hold $100 of government scrip, and the quantity of government scrip out there has no bearing on my ability to save. If there were no government and gold were the medium of exchange, I might write a balance sheet for my gumball holdings in terms of ounces of gold. I can save individually, and the economy can save in aggregate, and the expressed-in-gold-value of the economy’s wealth can grow without bounds, even though no new gold is issued. (The in-gold-value of the aggregate balance sheet might do all kinds of crazy things, as nothing fixes the gumball/gold exchange rate. But the total quantity of gold never limits the total expressed-in-gold value of private balance sheets.

    Back in the dungheap, if you believe the Fed Flow of Funds, US household net worth is roughly $51T, while USG liabilities (including the Fed) are roughly $13T. If we consolidate liabilities of the whole financial sector as USG obligations (screams of agony!), we add 17T, bringing us to $30T. Anyway you cut it, household net worth exceeds the stock of US guaranteed claims. The quantity of USG claims tells us something about the fraction of household assets (not net worth) that is state guaranteed, but nothing about the fraction of assets that will be held forward or used to produce in future periods. Running a greater deficit implies that households swap other assets for government claims, so it probably implies something about the volatility of household assets in USD terms (since govt claims are guaranteed). There is a strong relationship between govt claims and the liability side of household balance sheets, since as you often point out, household sector liabilities must be to the public sector, since liabilities to firms are indirectly liabilities to households. (FoF US household liabilities ar ~14T, which is close enough for govt work to the $13T formally guaranteed and just a bit of the financial sector guarantees taken “on balance sheet” (so not added back into the household sector somewhere).

    I find Winterspeakian savings to be a very useful concept, because it gives us intuition about how guaranteed the private sector’s assets are, demand for such savings tells us something about private sector risk aversion, and choices we make in meeting or failing to meet (or partially meeting via unemployment &benefits) that demand are important. I love your consolidated view of money, where government speand or lends it into existence and borrows or taxes it out of existence, and never requires “financing” of any sort in its own currency. But I really think it’s misguided to imagine that there is any necessary relationship between G-T and real savings or investment (good or bad). I think the closest you can get is that if you define G-T as investment (maybe very bad investment, but on the theory that lending counts as investment whatever its purpose, like credit card receivables), then you can say S = I > G-T, with the quality of I due to G-T questionable. There may be second order effects — govt may borrow from “spenders” and then make transfers to “savers” savers so that govt deficits and even quality real investment covary. A lot of scenarios are possible, but there’s not a deterministic relationship.

    Anyway, we agree that, mediated by govt claims or not, savings/investment in an accounting sense tells us nothing of quality. I think we agree that the line between savings/investment and consumption is blurry at best. But however defined, I think we still have pretty significant disagreements about whether government deficits and real savings/investment (however defined) have a necessary, positive relationship. I don’t think so, but I’ll bet your unpersuaded. It’s remarkable, that despite that, our actual policy intuitions are very close.

    [Update: Struck an incoherent bit in the above. The liablity side of the aggregate household balance sheet can’t be interpreted in the way I suggested, it makes no sense. The FoF household sector’s liabilites are also household sector assets. The liability side just “grosses up” the balance sheet, without an effect on net worth or any plausible relationship to govt liabilities. Brainfart!]

  67. winterspeak writes:

    Ay ay ay! Gold, gumballs, scrip — hard to keep straight (and lots of excellent points throughout). Let me focus on the first issue you raise, as I beleive it is the crux of the matter. Does Y = I + C? I can *make* it equal that by definition of course, but is it a useful construct?

    I think this is easiest to understand by first looking at the economy in real terms, and then nominal. I hope we agree that the real income of the economy is the 100 gumballs it produces per period. Some of the gumballs are consumed. Some of the gumballs are not, and are available for consumption in the next period (I would classify that as investment, which includes inventory). Given that the economy (like all economies) is wealthy enough to afford to consume all of it’s output, how, in nominal terms, do you account for the income which is not spent? We have real gumballs left over, some of our real income has not been spent. Do you see why that has to show up as savings in *whatever* currency you choose for the economy?

    It took me *forever* to get my head around this. Loans creating deposits was easy, federal deficits being private savings was hard, but savings being the record for real investment is really weird. This is why it’s taken me so long to tie it back to “winterspeakian” savings.

    As for the Government buying some of the economy’s output (and taxing to sterilize that spending), everything still works. If the Government runs a deficit, then this may or may not trigger inflation depending on whether that deficit is sufficient to fund the private sector’s demand to increase it’s savings that period. If the Government runs a balanced budget, the private sector *cannot* increase its’ holding of Govt scrip as it has no source for that new money. This should be very clear. It should also be clear that the real income of the economy is the amount of gumballs it produces each period. Issuing more scrip does not make the economy richer.

    If the Government consumes the gumballs it buys, then there is less real investment (fewer gumballs for next period), and therefore less savings. But note, this is *not* because of the deficit. The usual argument against deficits is that it “crowds out” private savings, which is nonsense because deficits *fund* private savings. The argument that the Government should use it’s assets productively is absolutely true. Keynes’ government workers, digging ditches and filling them up again show how easily you can get full employment at wildly different levels of productivity. And opportunity costs are real costs. Keeping the amount of scrip the same in an economy, and dramatically reducing the quantity of goods it can buy will trigger inflation as readily as increasing scrip, but keeping the quantity of goods the same. People may still hoard the scrip, but it will be able to buy less (obviously) if the Government starts consuming large amounts of real output.

    “I think you are hung up on accounting. If something is saved, it must show up in someone’s accounts as savings, right? It certainly does…. etc.” You’re confusing real with nominal, and household with sector.

    Your $100 gumball balance sheet is net zero. If you begin the balance sheet with 10 gumballs, you will have 10 gumballs as assets, and 10 gumballs as paid in equity. You can value the gumballs at whatever you like, and that should make it clear why your $100 “savings” aren’t denominated in actual scrip — it’s because your savings have not been realized. When you convert your gumballs into scrip, then you will have the scrip, and you will have your savings in scrip. Remember, S and I and C are *realized* *actual* and *in the period*, not *potential*, *intended* and *in some other period*. If you are a household and you want to increase your savings, then some other household needs to decrease their savings, ie. you trade your gumballs for their scrip. If the sector as a whole wants to increase it’s savings it’s stuck, it has no one who will take their gumballs in exchange for scrip, unless the Government steps in and does it.

    If you and your 10 gumballs are suddenly transported to Japan, your intended, potential balance sheet will suddenly switch for dollars to yen. Wow! Of course your actual real balance sheet is in gumballs, just as it ever was.

    “I think we still have pretty significant disagreements about whether government deficits and real savings/investment (however defined) have a necessary, positive relationship.” I think they have a necessary relationship, in that one enables the other, and they are equal to the penny, but I don’t think it has to be positive (ie. good). Investment can be good or bad, iPhones or Chryslers piling up in Long Beach. Savings is merely the record that investment happened. I’m guessing you don’t agree with my first position, but we are in agreement with my second, and the third just weirds you out totally. Weirds me out too.

    Anyway, good conversation as always.

  68. winterspeak writes:

    Ack. I promised myself I would not do this, but I think I see the point you’re making.

    You’re saying that the private sector can increase it’s savings rate by simply altering the balance between cash in circulation and cash in the bank. No need for higher Government deficits. When I say “increase savings” I should say “increase savings while maintaining aggregate demand” because as velocity falls to zero (100% savings) AD falls to zero as well. In the real world, Fisher’s debt deflation keeps prices from adjusting, as lower income is not met with lower prices but with dramatic decreases in money supply as credit is written down, and equity is consumed. I think that puts us in the same place wrt to goods, scrip, and prices. I still don’t think you believe that I = S, to the penny, because S accounts for I.

  69. I think that much of our disagreement reduces to semantics, by your introduction of the word “realized”. If by “realized savings” you mean savings that is converted to government money, then we are (nominally?) in agreement. The total quantity of “realized” savings is constrained by the quantity of government money.

    I think that most savings (and the most important savings) is savings in the form of potentially productive capital and stored goods. I certainly think of this as “actual” savings, and when the goods are purchased with the intention of capital/storage, I think that is both I and S from the perspective of the income identities.

    But to the degree that what we are talking about is savings in the form of government paper, we have no disagreement. We agree that in aggregate it’s constrained by the stock of government paper, whose change is a function of the deficit/surplus of the consolidated USG/Fed/insured-banking-system.

    I think our main disagreement comes down to which form of savings “matters” most, “unrealized” savings in the form of capital goods and claims on private sector entities, or “realized” savings in the form of government money. I think that “unrealized” savings matters more, and that when we think about savings we should mostly be thinking about the quality and quantity of real capital held. You focus on “realized” savings. I think there’s a case for both choices: the productive capacity of the real economy is ultimately a function of real capital — “unrealized” savings. But the perceived health of households’ individual balance sheets, and therefore households’ willingness of households to consume or engage in new investment, may be dependent on “realized” savings, how much peace-of-mind “money in the bank” they have, particularly if their “unrealized” savings is in an illiquid (and recently downside volatile) for of capital like a residence.

    I think that last intuition explains why we tend to disagree in practice even when we disagree in theory. We are both very aware that households want comfort capital now, want top believe they hold savings in a form that won’t suddenly “poof”. For better or for worse, government scrip is the form of savings people believe to be real, “money in the bank”. In ordinary times, I don’t think this is the most important form of savings. As I’ve said, in ordinary times I want this form of savings to be rationed, because the comfort it provides involves shifting risk to those who don’t whole “guaranteed” assets or creating systemic risk by putting stress on the price stability commitment.

    But I think we both agree that the biggest constraint on actual economic activity now is the health of household balance sheets (as perceived both by households and their potential counterparties), and the easiest way to expand economic activity would be to supply them with the “realized savings” that despite (in my view) its nonrelationship with real investment is the conventional sine qua non of individual balance sheet health. There is a tension between this policy and aggregate economic health: we can make every individual balance sheet strong while leaving the quality of real investment shoddy, and eventually the real will trump the nominal. I would prefer that we use this crisis as an opportunity to put our real capital allocation house in order, and I would prefer to the extent we engage in nominal expansion we do so in a manner that is “flat”, rather than one that compounds patterns of wealth that correlated poorly with actual production. But one way or another, I think that you are right that the economy will be stalled until the demand for “healthy” balance sheets is accommodated, whether by transfers (my preference), tax cuts, fiscal spending, or “on the back of unemployment” via automatic stabilizers.

    btw i’m not sure i understand your last point — i wasn’t arguing about the balance of cash in circulation and cash in the bank, i think that the guaranteed banking sector is basically consolidated with government for the purposes of this analysis, except that the banking sector usually runs deficits mostly by borrowing against loans, which creates a nominal liability against a nominal asset and so fails to supply very much net balance sheet joy, although the maturity mismatch helps. but to the degree the banking sector creates deposits against real assets, say by nonrecourse lending against homes, its effect on “realized savings” is similar to the government running a deficit and purchasing assets. Ironically, when the banking system creates deposits, purchases assets, and then transfers those to insiders inclined to consume, it both creates perceived balance sheet health (“realized savings”) and increases present consumption even while it puts a stake in the heart of the quality of aggregate investment. but this is the same as the government borrowing to purchase goods and then distributing them.

    yes. very good conversations.