Links: UBI and hard money

Max Sawicky offers a response to the post he inspired on the political economy of a universal basic income. See also a related post by Josh Mason, and a typically thoughtful thread by interfluidity‘s commenters.

I’m going to use this post to make space for some links worth remembering, both on UBI and hard money (see two posts back). The selection will be arbitrary and eclectic with unforgivable omissions, things I happen to have encountered recently. Please feel encouraged to scold me for what I’ve missed in the comments.

With UBI, I’m not including links to “helicopter money” proposals (even though I like them!). “Helicopter money” refers to using variable money transfers as a high frequency demand stabilization tool. UBI refers to steady, reliable money transfers as a means of stabilizing incomes, reducing poverty, compressing the income distribution, and changing the baseline around which other tools might stabilize demand. I’ve blurred the distinction in the past. Now I’ll try not to.

The hard money links include posts that came after the original flurry of conversation, posts you may have missed and ought not to have.

A note — Max Sawicky has a second post that mentions me, but really critiques Morgan Warstler’s GICYB plan, which you should read if you haven’t. Warstler’s ideas are creative and interesting, and I enjoy tussling with him on Twitter, but his views are not mine.

Anyway, links.


41 Responses to “Links: UBI and hard money”

  1. Dave Timoney writes:

    A discussion of the political differences between the UBI and the job guarantee: Scoubidou and the Protestant Work Ethic.

  2. Dave Timoney writes:
  3. Tim Young writes:

    ““Helicopter money” refers to using variable money transfers as a high frequency demand stabilization tool.”

    Not really. It is not much good if the stabilisation need is to restrain demand. Admit it, Steve, helicopter money is practically an inflationary ratchet, and since you (wrongly in my view) see the holders of money as undeserving risk-shy rentiers, you don’t really care if it is.

  4. Tim. No on every count. Perhaps I write poorly, or you might take a bit more time to read with care.

    The rate of distribution of helicopter money can be adjusted downward. That’s precisely why it’s important to distinguish it from a basic income, because it offers disbursements that cannot be relied upon to continue.

    Helicopter money suffers a zero lower bound, ie when the rate of disbursements falls to none. Fortunately, there is no upper bound on interest rate policy (and in extremis taxation) so with the addition of helicopter money both expansionary and contractionary policy are unbounded. There are issues if you want to add other targets besides aggregate demand management (see eg Mason and Jayadev on targeting both an output gap and a debt ratio), but there is no permanent ratchet.

    I do not see all holders of money as undeserving risk-shy rentiers. I see inequality of security and insurance to be a major problem, and I do think insurance is misallocated if we are stabilizing the real position of large holders of money while not offering meaningful insurance to those is much more fragile circumstances. I would be very glad if we could supply wealth and security to all — I really don’t care who is deserving — but if we cannot, I’d rather we improve the security of the fragile even if it comes at the expense of the relatively secure. Unfortunately, offering unlimited, purchasing power stabilized, zero-risk instruments requires allocating risk and costs to those less able to bear it. So we should find a better policy.

  5. Peter K. writes:

    @3 I don’t see why you believe restraining demand has been a problem. The problem is when the holders of money disengenously argue their preferred policies will bring about general prosperity when that is not the case. It’s a problem when people believe what they are saying is the truth. Since they are holding money, they have money to burn on propaganda.

  6. Tim Young writes:

    I guess I got the impression that you do not have much sympathy for savers in nominal assets from posts like these, Steve: , and

    I suppose I was assuming that helicopter money would only be used as a last resort in a slump bad enough to use up all capacity for fiscal stimulus as well as monetary stimulus. The central bank could issue a certain amount of base money even in normal economic conditions by crediting it directly to the government with no asset backing, but since that would leave the central bank without independent means to fully contract the base money supply, my point about an inflationary ratchet applies.

    As I said in commenting on your previous post, I am favour of basic income, but would fund it by looking the asset-rich in the eye and taxing them on their wealth that has been enhanced by QE, rather than levying some stealth tax on the small unsophisticated savers who tend to rely most on saving in simple nominal assets like bank accounts.

  7. Tim,

    If you want to accuse me of being mean to savers, you’ve missed the smoking gun!

    And I am mean, and I intend to be mean, to large savers. But in nearly all of my mean-to-savers pieces I include the caveat that we absolutely should provide purchasing-power-insured savings for modest buffers and nest eggs. Purchasing power insurance (which is what inflation targeting provides for nominal assets) is an expensive social insurance program whose provision we should ration. In fact, I’ve explicitly proposed schemes for providing and rationing that insurance:

    Elsewhere I’ve offered support for the other way we insure money, deposit insurance, but again only with strict, actually binding limits. (Actual FDIC has limits that are trivial to circumvent — cf CDARS — and spottily enforced — cf financial crisis of 2008.

    Purchasing power insurance on indefinitely large nest eggs is the worst welfare program ever. But that doesn’t imply a policy of continual ratcheting debasement, as you suggest. It just means imposing some purchasing power risk — symmetrical even! — on nominal asset holders. NGDP targeting, which I support by via helicopter money, would do that.

    You’ve misunderstood re helimoney. Do read

    I don’t support it as a last resort at the zero bound. I think interest rate management is a deeply problematic macro stabilization tool. I prefer helimoney as the tool of first resort, with interest rate management and taxation kept in the back pocket for the helimoney zero bound. There’s no ratcheting debasement, but modulation in both directions to track an NGDP path. (In the piece I still presumed an inflation target, but I now prefer NGDP.)

  8. Tim Young writes:

    I think you made my original point for me, Steve. You write that the Fed should “fake” an asset to offset its base money issuance, in which case the Fed would not have the independent means to contract the base money supply (because the fake asset could not be sold) if necessary, which is what I mean by an inflationary ratchet – suspending base money creation cannot contract the stock. And “Purchasing power insurance (which is what inflation targeting provides for nominal assets) is an expensive social insurance program whose provision we should ration.” implies that your desired norm is to allow occasional burst of inflation presumably averaging higher than at present, with a safe haven of limited size provided for the smallest savers.

    And let’s be honest, while you write as if you were proposing some kind of system reform, what you are really advocating is just an extension of the bailout for the holders of risky assets. If system reform was your primary concern, you would start by granting a real value guarantee to all existing nominal assets. And here is where I think your plan would fail. If you are open that holders of nominal contracts will face more risk, I would expect that, as after the 1970s inflation, the creditor-debtor bargain will simply reset to higher interest rates, and debtors will moan about that too.

    As I keep saying, my views are generally a bit left-of-centre, but I am strongly against magic monetary solutions and cheating people. I support wealth taxes, but let them be explicit, precise (ie allowing you to penalise nominal contracts if they bother you that much) and democratically approved.

  9. Tim — The “faking” of an asset, as I think that piece fairly carefully explained, didn’t represent any kind of cheating people or magic monetary solution. The solution proposed reflects the fact that a central bank balance sheet is not analogous to any other entity’s balance sheet. Since a central bank is the one entity that can never be illiquid in its own currency, the primary function of a balance sheet — predicting risks of illiquidity — is eliminated, and the real function of a CB balance sheet is to track the cumulative policy choices of the bank.

    The term goodwill was very carefully chosen, because there is one good analogy with private sector balance sheets. Defining a “goodwill” asset against helicopter drops would be a clear way of marking a historical cost value to the CBs demand stimulus program while emphasizing that the value is entirely intangible. The terms goodwill is used in a very similar way on private sector balance sheets. Do you object to the fact that when one firm acquires another, its cost almost always is higher than book value, yet it marks that surplus as “goodwill” on its balance sheet to avoid an immediate loss? You shouldn’t — the alternative would be forcing a useless and uninformative mark of a loss on the acquiring firm, which would not have acquired if it thought the operation an actual economic loss. For a central bank, the goodwill would reflect the fact that the CB perceives its operations as improving the real value of its assets, because the purpose of those operations would be to maximize real output in the economy (i.e. eliminate an output gap due to demand deficiency). “Goodwill”, for private firms as for central banks, represents an artificial, intangible category that is necessary to capture important realities lost by simple mechanical approaches to the balance sheet.

    If you think I advocate bailing out holders of risky assets, you’ve not read interfluidity for very long. I was and am adamantly opposed, and one of the many reasons I dislike interest rates as an instrument of monetary policy is because stimulus tends to raise long duration asset values long before it (maybe) helps people who don’t hold such assets. There is absolutely no aspect of helidrops that are intended to bail risky assets. On the contrary, they are motivated precisely by a desire to make it possible to bail out the macroeconomy while allowing risky assets to fail.

    When you write let’s be honest, you frankly piss me the fuck off. Because I’m an asshole and I may be wrong, but I am an exceedingly honest writer, and if you think I’m hiding some Svengali agenda that I’m not putting forth with painful candor, you are very much mistaken. Why on earth would I want to grant a real value guarantee to existing nominal assets when I think the current skein of policies that do largely guarantee the real value of nominal assets are the source of most macroeconomic evil? What you have been criticizing me for is precisely the fact that I do not, would not, would never, want to guarantee the real value of nominal (or not nominal) assets, except for small nest-eggs as explicitly rationed social insurance.

    If we do eliminate guarantees on nominal assets, yes, extra risk premia will become built into the price of nominal debt. That’s a feature not a bug, as you should have read if you had carefully read the piece from which you enjoyed pulling out the word “fake”. Again, helidrops mean we don’t need to depend on expanding credit to support the macroeconomy, and it is my view that we should work towards a dramatic retrenchment of the scope of consumer credit, by replacing borrowing with income, derived from a mix of increased labor bargaining power, helicopter drops when aggregate incomes falter, and a basic income. Again, if you have read my oeuvre with any degree of attention, absolutely none of this would be news to you, I drone on and on about the same themes so often I find my own crap too boring to write very often.

    But if you caricature a perhaps mistaken but pretty relentlessly and carefully expressed constellation of views as “magic monetary solutions” and “cheating people”, if you suggest I’m trying to pull the wool over anyone’s eyes and circumvent democratic approval, really, all I can ask is that you learn to read more attentively. I could care less whether you are left or right, if we’ve matters to make common cause about, awesome, I’ll stand by your side. But really. Criticize all you want, but don’t make up dumb views and “let’s be honest” them into being mine. Sheesh.

  10. Tim Young writes:

    A central bank can effectively be illiquid in its own currency if it is constrained by its inflation target, as I believe it should be. And of course unanticipated inflation bails out risky assets, because it floats many boats. Imagine what a 20% burst of inflation would have done for the sub-prime mortgage market in 2007. Even the most marginal borrowers would have been able to make their mortgage payments, the most reckless purchasers of opaque CDOs – think Wing Chau in the Big Short – would have been vindicated, and the market, probably joined by frustrated prudent treasury buyers, would have climbed to the next diving board up. In the UK, we’ve had the same process work out with house prices. The BoE compromised its inflation target, house prices fell little in nominal terms from the already stretched pre-financial-crisis levels, and now they are rising to even more dangerous levels, being sustained by a near-record current account deficit.

  11. A central bank’s balance sheet is not a meaningful predictor of inflation, so even under the very broad definition of “illiquidity” you prefer, it is incoherent to treat a CB balance sheet as analogous to that of a firm’s, whose primary purpose is to predict a much more brutal notion of illiquidity. The US Fed’s balance sheet has now effectively quadrupled in size. Yet the period has been disinflationary. When a central bank can vary the rate of interest on reserves, there is no simple relationship between balance sheet size and price level effects. When a central bank holds the IOR rate at or above the rate paid on general government debt, CB balance sheet size has little effect on the overall price level, though it does affect the price of the financial assets it purchases and that of other assets whose prices are related to the assets it purchases.

    Any effective macro policy “floats many boats” (and creates losers too). A policy regime less focused on inflation constraint would indeed help some asset-holders: those whose assets would otherwise be severely impaired by defaults. It also hurts many other asset holders: those who hold currency, government debt, or low credit risk debt. Interest rate policy (the current status quo) very mildly helps asset-holders at risk of default (some, not most, risky assets are made more valuable by reducing the interest burden faced by their issuer), but very immediately helps the vast majority of financial asset holders via reduction of the discount rate on both term debt and equities). If you dislike bailing out asset-holders, you should detest status quo interest rate management. Helidrops, you should concede, would be an improvement. Indirectly, they would bail out some otherwise impaired assets (extra income assists debt servicing, extra demand assists equities), but in a way that is far less targeted to asset-holders and far more direct in its stimulatory effect than interest rate policy. The question of helidrops is unrelated to the question of whether CB action should be disciplined by an inflation target. You’ll note that my original helidrop proposal presumed that it would be, although I currently favor discipline by an NGDP target.

    I agree with you that macro policy via goosed housing prices is shitty macro policy. Note that helidrops in the UK during the period you complain about would have had a much milder effect on housing prices for the same NGDP or employment stimulatory effect, and could have been disciplined by an inflation target. So why would you prefer status quo interest rate policy?

    When you are not putting words into my mouth (and I am grateful that you have not in your latest installment), it seems to me that our core disagreement is over whether macro policy should be disciplined by an inflation target or an NGDP path target. Do you have some independent reason to disprefer helidrops, or do you dislike them because you presumed, erroneously, that they must necessarily attach to an inflationary ratchet of some sort? Do we have some other disagreement I am not seeing?

  12. Tim Young writes:

    I think our difference, Steve, comes down to what the central bank needs to do to be able to independently meet its inflation target at all times. In order to do that, it must have assets to sell, so a helicopter drop leaves it, as you Americans say, naked as a jaybird. You can’t sell goodwill, but in the corporate world, valuing an acquisition somewhat above book value can be justified for a while at least, because until the acquisition becomes integrated with the acquiring firm, the acquisition could be easily spun off again. If anything, completely the opposite is true of helicopter drops – you do not want people to consider the possibility that their gift could be taxed back in the near future. Paying interest on reserves is no solution, because it just makes the central bank’s balance sheet hole bigger, so amounts to a Ponzi game.

    I did not say that there is a simple relationship between central bank balance sheet size and the price level. Besides IOER, there is also the question of the demand for base money. During a financial crisis, the demand for base money is augmented by demand for money as a safe asset, so it should come as no surprise that the central bank can expand its balance sheet in such conditions without raising inflation – indeed it should do so, to avoid 1930s-style deflation. But the several-fold expansions of base money in the US or UK by QE represents a risk that will not be tested until the time comes to withdraw QE. And that test cannot even be taken if the central bank has no assets to sell, and the government is not prepared to increase taxes to, say, service bonds issued gratis to the central bank and sold to reabsorb base money. Again, that is what I mean by an inflationary ratchet; not that inflation is inevitable – if financial risks are perceived to remain, the demand for base money as a safe asset may remain high indefinitely, like Japan and now US/UK – but that the bridge to independently reducing it has been burned.

    When I wrote that unanticipated inflation bails out the holders of risky assets, I should have added something like “even if this inflation comes about by gifting money to the public rather than buying assets”. I am sure you are right that how the added base money is distributed matters, but I have to say that in the UK, people are so obsessed with rising real estate prices (viz the plethora of house-trading TV programmes), that I seriously doubt that helidrops would have had a milder effect on house prices than QE.

  13. Steve Randy Waldman writes:

    Tim — A central bank’s balance sheet is no operational constraint whatsoever on its ability to withdraw base money. Central banks can and do issue and sell their own debt. PBoC has issued “sterilization bonds” for many years. The US Fed is now ramping up its own debt issues in the form of term deposits and reverse repos. By law or custom it is possible that such independent debt issues might be restrained. Indeed I once argued that the Fed should be, when I wanted the US Fed to be prevented from taking private sector risk onto its books, as it was determined to do. (That was before it had hit its ZLB, so it needed to sterilize its bailout purchases to maintain its rate target.) But that’s an optional constraint on CB behavior that has already disappeared in the United States. I don’t know whether the Bank of England is experimenting with its own debt issues (bonds, interest paying term deposits, reverse repos, whatever.) This is a much less radical change than the introduction of IOR, which, when combined with the expansion of CB balance sheets, leaves base money a close substitute to government debt anyway.

    In any case, your critique is mistaken. To the degree that a central bank wishes to replace circulating base money with government debt, as long as no law prevents it and the market considers the CBs debt equivalent to that of the Treasury, the CB may do so. No meaningful flexibility in terms of quantity of base money vs government debt is lost in a helidrop regime.

    You did not say anything particular at all about the price level and the CB balance sheet, but you did define (a bit histrionically) missing the CB’s macro target as “insolvency”, and implied (or so I thought) that meant CB balance sheets could be understood as analogous to firm balance sheets, as predictors of potential insolvency. As I said and you agree, even under your odd definition of insolvency, CB balance sheets cannot be so interpreted, the relationship between a CB balance sheet and macro targets is too loose once you introduce the extra degree of freedom that is IOR.

    I think you are quite mistaken to suggest that, in a counterfactual world where BoE held rates at some “conventional” level (say 4% or 5%) and relied upon helidrops for stimulus, that the British housing market would have become similarly frothy. We cannot, of course, run the counterfactual. But we do know from decades of experience that the things most directly and immediately affected by interest rate policy is the value of real estate and financial assets, and we understand perfectly well why this should be. Financial assets and rentable real estate appreciates by virtue of the reduction of discount rates. Even real estate that is unlikely ever to be rented appreciates by virtue of the incremental “affordability” of high prices when mortgage rates are low. With low risk mortgage rates at 6% or 7%, Britain would not have seen such a rise in housing prices, or any rise at all. Helicopter drops would have made very little difference: the quantity of money per capita that would be injected to counter macro fluctuations would not be large enough to make a big difference in home prices, and the vast majority of recipients would use the funds for other purposes. Holding effective demand constant while trading increased interest rates for helidrop money is terrible for real estate and financial assets (imagining a counterfactual rather than a transition in time, when we can expect an uncomfortable crisis from any increase in interest rates unless several years of on-targer inflation are allowed to help ratify prices). Wealth effects (i.e. the effect of unrealized capital gains on spending behavior) are an extremely inefficient means of generating demand, on the order of 5% of the capital appreciation is spent (dimly recalling old papers on the subject). Lots of asset appreciation could be foregone while still hitting macro targets under a helidrop regime.

    Your talk of “inflationary ratchets” assumes a particular kind of political dysfunction. High interest rates with IOR are not a Ponzi game. They just represent another form of government debt, and we’ve seen eras where interest rates have been in the high double digits. Helicopter drops are just another form of fiscal policy, no more or less problematic from a public debt perspective than any other kind of spending. The rate at which government debt can be financed, including IOR, depends on the NGDP growth rate of the economy. There is no hazard of a debt spiral unless the interest on government debt (including reserves) is higher than NGDP growth for a prolonged period. There is a natural feedback loop that pushes against — though does not necessarily pervent! — the kind of dynamic that frightens you. IOR would be raised, under a helidrop regime or an interest rate regime, only if there were inflationary pressure, that is, if NGDP were running hot. The high NGDP growth degrades the debt even as the increased interest rate accelerates it.

    That’s helpful, but you would be absolutely right in saying it’s no more than a “guard rail”, if we push it, we could generate spiraling public debt dynamics. We have to worry about that less than you think, I’d argue, but we do have to worry about it some. (The main thing we should worry about with public debt is the very unequal distribution of government guaranteed assets that inevitably results.) Ultimately, taxation has to come into the mix, whether you wish to describe it MMT-ishly as a tool for managing inflation without the interest cost, or conventionally as a means of directly replacing or reducing debt.

    Ultimately your suspicion of helidrops derives from an assumption of a certain political dysfunction. If “baseline” taxation by the fisc is sufficiently severe, a helidropping CB will never face its zero bound, it will always be able to modulate a positive rate of transfer. My claim is that this is the macro-management regime we should strive for: general taxation at a level that would be unduly contractionary, offset by variable helicopter drops. Your worry, fair enough, is that we might do the helicopter drops without sufficient taxation, get to the point where zero helidrops is too expansionary and thus be forced into raising interest rates, a policy whose hazards you overstate, but with real hazards nonetheless.

    But the existing regime encounters a perfectly analogous possibility of political dysfunction. Interest rate policy is only a reliable tool when the baseline fiscal stance is sufficiently aggressive that hitting the macro target requires positive rates of interest to restrain demand. Given your fiscal concerns, you should absolutely detest the status quo, because it requires the fiscal burden of paying out interest and loose fiscal policy. (It is irrelevant whether that interest is IOR or interest on government debt, other than an inconsequentially small savings on IOR when reserves are scarce so IOR can be below T-bill rates. When the quantity of reserves is enough to matter, IOR must be greater than or equal to the short-term T-bill rate, or banks will have a happy, easy arbitrage.) Under the status quo, the hazard that might prevent hitting the macro target is too conservative a fiscal stance. This is precisely the inverse of a helidrop regime, where the hazard is too loose a fiscal stance.

    So, we have a political risk management decision to make. In either regime, the CB will be able to hit its macro target reliably as long as it is not pressed against the regime specific zero-bound. In either regime, if parliament/congress/treasury fails to set the fiscal baseline appropriately, the CB will miss its macro target, leading to some form of depression. In an interest rate management regime, too-tight fiscal force the CB to miss on the downside, inflation or NGDP too low. In a helidrop regime, too loose fiscal means the CB misses on the upside, unwanted inflation or too-high NGDP. Perhaps a place where we disagree is this: which risk is more tolerable?

    In my view, inflationary (but not hyperinflationary) depressions are generally preferable to deflationary depressions. Deflationary depressions are much more disruptive, as assets lose value through the unpredictable and sometimes cascading process of bankruptcy, default, and restructuring, which overwhelms court and legal systems and causes a lot of unpredictable conflict. Deflationary depressions are historically much harder on employment than inflationary depression as well. If the parliament/congress/treasury is going to screw up and set a baseline that the CB will be unable to technocrat its way out of, I’d rather the CB be stuck with unwanted inflation than with unwanted inflation. Perhaps you take the opposite side, and think that if there must be a miss, a deflationary or disinflationary miss would be superior. I think that’s a mistaken view, but there’s a case to be made there.

    In any case, stories about an “inflationary ratchet” are far too simple. Indeed, I could claim that interest rate targeting is a deflationary ratchet, if I presume that the fisc fails to set a reasonable baseline as you need to to make the opposite case. Current events (and at least in the US three decades of secular disinflation) seem to bear out that view! But I’ll be the first to admit its a caricature: nothing would prevent easily managing any inflation or NGDP target under interest rate management if the baseline fiscal stance were loose enough (and egalitarian enough, given propensity to spend effects) to give interest rates something to pull back against. In the same way, nothing would cause helidrops to be lead to perpetual rises in inflation, unless the fiscal stance is left so loose that zero is too big a drop. And in a helidrop universe, there is always the additional tool of interest rates as a backstop. (High interest rates may persuade the fisc to return to a tight stance so that the appropriate helidrop rate becmes positive. Historically, high interest rates are a catalyst for fiscal retrenchment.)

    When the interest rate is our only instrument, we are pretty much screwed at the zero bound. We have no backstop but the fisc, which has failed once if it let us get here, and is inappropriate to use for fine-grained feedback and targeting. We can have our usual left-right expansion-austerity debates, and resort to QE and other mostly symbolic/expectational forms of macro policy, but we are pretty stuck. In a regime where helidrops were the tool of first resort, and interest rate increases a “penalty” backstop, we’d never be so helpless, we’d have a tool that would help goad the fisc to appropriate (appropriately tight!) policy, and if we are in a bind, we err on the side of inflationary rather than deflationary depression. In every respect, then, I’d argue that helidrops + interest rate as last resort backstop is a superior macro regime.

  14. Tim Young writes:

    Steve, you are missing the wood for the trees. For routine policy purposes, an independent central bank IS like that of a firm; ie it is not part of government. Yes, governments generally appropriate any surplus from the central bank and retain powers to direct it in emergencies, but outside of a genuine emergency like a war, using such emergency powers would likely provoke a constitutional crisis that would probably mortally wound the government. In this light, an independent central bank creating and issuing its own debt is another form of Ponzi game – it is just swapping one form of liability for another which emits more of the first form of liability. You are right that the PBoC issues sterilisation bills to mop up its currency intervention, but (a) it does so after acquiring assets of similar value in the form of US Treasury bonds, and (b) it is not independent of government.

    As far as I am concerned, a central bank should have only a minor role in macroeconomic management beyond the maintenance of price stability. Monetary policy has some short-lived power to stimulate or restrain real economic activity, and provided that this is used even-handedly, then the central bank may, subject to its inflation target being met, assist a little with demand management. This view was more or less (Greenspan was already on the slippery slope) the norm until the financial crisis, since when an unholy alliance of governments desperate to avoid saddling their electorates with the real cost of cleaning up after the crisis, and attention-loving central bankers, began to subvert this hard-won understanding. As we have seen in the last week or so, the insurgents are now besieging the last major refuge of this truth at the Bundesbank.

  15. Steve Randy Waldman writes:

    Well, if you don’t think we need some arm of government to do high-frequency, fine-grained effective demand management. there’s the source of our disagreement. To me, that is the role that central banks have evolved to play. It cannot be separated from the role you ascribe to CBs, price stability, because demand management sometimes implies making tradeoffs between price stability and other macroeconomic goals, like output growth and employment.

    I view a central bank to be an arm of government, with its “independence” a matter of institutional choices designed to insulate CBs from some but not all of the so-called “political” incentives that affect some other parts of government. Fundamentally, I think of a CB as not especially different from a regulatory agency, with a mandate and powers delegated by the legislature (in notionally democratic countries) but enabled to use those powers in pursuit of its mandate without cumbersome legislative action. Within any government, there are agencies with a wide variety of differing degrees and characters of “independence”, from agencies whose directors serve at the pleasure of ruling politicians to courts with judges appointed for life and (to a certain degree) “independent” of politics. A CB is an institution within that continuum. You may dislike this. There are institutional differences across CBs, and about some of them we can squabble about whether they are “really” private banks or government agencies. But in all major economies with national central banks, central banks are certainly very different from ordinary private firms. Indeed, banks in general are very different from ordinary private firms. In my view, modern banking systems are more public institutions than private, but enshrouded with quasiprivate institutional forms in order to encourage decisionmaking on a commercial basis that is difficult (we know this historically) when the public nature of banks is made very explicit. There is a continuum in my view (this is a positive view, not a normative one, an is not and ought) between a regulatory agency like the US EPA, a central bank like the US Federal Reserve, and a “private” but publicly backstopped and regulated bank like JP Morgan.

    It is, I think, pretty indisputable though that the balance sheet of central banks in stable developed countries are not treated by markets like the balance sheets of firms, and in fact are not at all like the balance sheets of banks or firms so long as the coordination equilibrium surrounding its liabilities remains entrenched. During the “taper tantrum”, when long-term interest rates spiked in the US, the US Federal Reserve was almost certainly “insolvent” on a mark-to-market basis, as it suffered losses on its increasingly high duration portfolio. (Those losses would have easily overwhelmed the tiny formal equity of the institution.) This fact mattered not at all, and could not matter, because for a central bank, balance sheet insolvency is not a predictor of illiquidity, there is no USD obligation the US central bank could not meet. Whether that “should be” or “should not be”, whether a CB ought to face constraints that it does not, is a separate question. We could, for example, impose legal constraints and consequences that required CBs to be balance sheet solvent (as I think you might prefer). But we haven’t chosen to, and nothing requires us to do so. There is, of course, the possibility of the coordination equilibrium surrounding the CBs liabilities failing, a run on the central bank or equivalently a hyperinflation. But historically. again, that has little to do with the formal balance sheet position of a central bank and almost always follows a major political upheaval or imbalance that encompasses the larger government. The US Fed and the Bank of England are safe from runs so long as the US and British governments are deemed to have economies productive enough and taxation powers strong enough that they could force redemption of CB liabilities for goods and services in the economy at a decent price. I think, again as a positive not normative matter, you are making an error in viewing the CB balance sheet as being informative unconsolidated with the fiscal position and taxing power of the larger government. National CBs are the ultimate to big to fail financial institution, they enjoy a firm guarantee from the governments they relate to.

    ECB is an interesting case, since it is not attached to a single government. Still, I’d argue, the ECB’s balance sheet is mostly irrelevant as an analytical document. The strength of the ECB’s liabilities is pretty much entirely a matter of the strength of the market’s belief that governments of economically strong members support the bank. If, for example, there was a serious risk that (as many have suggested), strong Northern European economies might secede from the Eurozone into their own currency zone, the Euro would collapse, regardless of whether its balance sheet at the time was full of safe German bunds or trash ABS discounted to it by banks in the periphery.

    You mischaracterize the formal role of the central bank, at least in the United States, prior to the crisis. Formally, the US central bank is and has been since the 1970s charged with three goals, “maximum employment, stable prices, and moderate long-term interest rates”. Your view captures what is in fact a novelty since the 1980s, a view macroeconomists refer to as “divine coincidence”, that argues pursuit of price stability would achieve the other goals as a side effect, so price stability could be the sole goal of a central bank. That view, which derives from very stylized New Keynesian DSGE models, is I think no longer viewed as realistic. (I always considered it risible. But some economists obviously took it seriously. It’s reflected in the charter of the ECB, for example.) As a matter of historical fact in the United States (I should know more about the Bank of England, but I don’t), the price stability uber alles view you propose is legally incorrect and as a matter of practice was not pursued except arguably for a short period of time between the mid-1980s and the early 2000s. You may dislike this, but there it is.

    We can argue about whether governments ought to be in the business of trying to manage effective demand, or whether price stability should be pursued as a lexicographic priority over other goals. We began because you dislike that I have taken the opposite side of this argument than you would. I believe that price stability should not be the overriding goal of a central bank, and have argued quite extensively about why. (I believe that there are times when maximizing real output and growth is inconsistent with price stability. I dislike the distributional implications of prioritizing price stability over other goals.) But that is what we are arguing over, not details of central bank balance sheets or Ponzi games or anything else. My position could be wrong (lead to worse outcomes than the one you prefer for nearly everyone), near universally right (lead to better outcomes for almost everyone at least in the medium term), or most likely contingently right (lead to better outcomes for some and worse for others). The same is true of your position. I think it is very likely that we are both contingently right, but place different weights on different interests when we consider the tradeoffs that would be made. (I don’t think hard money is a mistake, I think it is an interest.) You may think my position is wrong for nearly everyone, but again, I think it would be more fruitful to argue that in terms of the costs and benefits of price stability rather than details of CB balance sheets that mean very much less than people pretend.

  16. Tim Young writes:

    “if you don’t think we need some arm of government to do high-frequency, fine-grained effective demand management”: What I wrote was actually the opposite (although I am not sure what you mean by “fine-grained”). I said myself that it is reasonable for central banks to use their short-lived influence over economic activity to stabilise output, which can be done in a practically lexicographic objective framework if this involves an acceptable range around the point inflation target, like the BoE. This became known as “flexible inflation targeting”. But it is vital to resist the temptation to try to stretch out the central bank’s influence for too long, because there does not seem to be much long-run trade-off between price stability and real economic activity. This is both what you would expect from common sense (quantity theory ideas going back to Aristotle) and was the realised experience of the 1970s, long before we had DSGE models. The Fed’s dual/triple mandate is a politicians’ fantasy that should be changed, but when the Fed was at the height of its moral power after Volcker, they probably did not think it worth the effort. And, while today’s gutless politicians would no doubt like it forgotten, there is fiscal policy too.

    Central bank “independence” may be seen as another frequency domain issue. As is right in a democratic society, the central bank is ultimately under the control of government, but, for stability, this control is constitutionally organised to march to a different, slower, beat than the electoral cycle, rather like, in the US, appointments to the supreme court. The trouble is, however, that a new breed of central bankers, in a desire to ingratiate themselves with the politicians who are their patrons, is subverting this scheme.

  17. stone writes:

    I’m struck by how Tim Young @6 did kick qualify his argument by stating clearly that there should be an asset tax as a way to redistribute. This Tim Young/SRW argument doesn’t seem to be about whether there is big a need for redistribution. It is purely about whether it should be conducted by way of explicit asset taxation or by way of inflation. I must admit, I’m also in the camp of favoring price stability so long as it is accompanied by a sufficiently aggressive asset tax.

    My impression is that a crucial backdrop to this whole hard money debate is how the economies of different nations interact. I think we need to explore and face up to how the 1970s did lead to something of slide in the fortunes of the developed world and that the 1980s did reverse that. Crudely, in the 1970s, in the developed world labor gained a greater share of economic output and capital relinquished that. In the 1980s, that was reversed. In the 1970s, there was much investment in developing countries. In the 1980s, the developed world instead became the ideal place to store wealth. By making the developed world an ideal home to store wealth, we were able to ensure massive capital flows out of the developing world to the developed world. The consequent massive currency exchange rate shifts meant we were able to transform the terms of trade so that someone in say Nigeria had to do many many times as much work/ provide far more natural resources to buy a share of global output and so we got it all instead.
    I had a try posting about that:

    I think it is a really big mistake to pretend that “wealth friendly policies” do not provide a free lunch for the developed world. We just need to unpick it and expose how it works. No-one really wants to perpetuate our current global annual three million deaths from childhood hunger.

  18. Peter K. writes:

    Is Brazil’s conditional cash transfer program (Bolsa Familia) somewhat like a UBI?

    “his return to growth, plus the government’s use of increased revenues to boost social spending, has reduced Brazil’s poverty rate by 55 percent and extreme poverty by 65 percent. For those in extreme poverty, the government’s internationally renowned conditional cash transfer program (Bolsa Familia) provided 60 percent of their income in 2011, up from 10 percent in 2003. A hefty increase in the minimum wage – 84 percent since 2003 after adjusting for inflation – also helped quite a bit.

    Unemployment has fallen to a record low of 4.9 percent; it was 12.3 percent when Lula da Silva took office in 2003. The quality of jobs has also increased: the percentage of workers stuck in the informal sector of the economy shrank from 22 to 13 percent.”

  19. Tim — I think we may have very different views of the CB’s frequency domain! If I’m not misreading you, you’re suggesting that CB’s are on a low frequency domain, slower then the election cycle. I have the opposite view.

    The reason why I think we need the CB to act as demand manager is that it can act, in pursuit of its mandate, on a month-to-month basis, much faster than and (partially) insulated from election cycles, and without the cumbersome poliicking and ad hoc legislation of conventional fiscal policy. You talk of “gutless politicians”, but that’s not a helpful way to put the problem. If we want to use fiscal policy to manage demand, we need to define institutions to make that possible. Helicopter drops are precisely a proposal to make fiscal rather than interest rate policy workable for high frequency macro demand management.

    We disagree very fundamentally on whether there are (sometimes) tradeoffs between price stability and real economic activity, both the quality and quantity thereof. I am quite sure there are, you believe there are not. You are certainly with the consensus that existed among orthodox economists from the mid-1990s until 2008, but it’s important to note that even among that rather compromised group, your view is the consensus no longer. Neither is my view — there is no consensus. I’d argue that invoking the New Keynesian DSGE case for “divine coincidence” is no longer tenable: even among the mainstream macroeconomists who have not jettisoned DSGE as a methodology, the stylized, no financial sector, single representative household, models in which divine coincidence appeared are discredited. The profession is moving towards DSGE models with financial “frictions” and heterogeneous agents grafted on, for better or for worse. Perhaps some of these will resuscitate divine coincidence. I suspect that many of them will not, or will at least identify divergences of interest among heterogeneous agents beneath whatever happens to aggregate output. (It’s very easy to come up with Pareto optimality in a world where only a single consumer is represented! In order to go from models with heterogeneous consumers to policy recommendations, DSGE macro will have to start thinking about social welfare functions, a value-laden exercise practitioners of this style of economics have been loathe to consider.)

    Stone — If there were some sort of redistributive taxation so great that interests of the relatively wealthy and relatively poor didn’t diverge much, your synthesis of my views and Tim’s would be reasonable. But I think that’s a bridge too far to hope for (and, for the record, it is not what I hope for). Short of that, our disagreement is pretty hard to reconcile I think. I believe that prioritizing price stability over other objectives is extremely costly for some social groups and beneficial for other social groups. That policy amounts, in my view, to an expensive transfer program whose costs are borne by those who can least bear them and whose benefits accrue to those who are most comfortable, and which creates deadweight costs in terms of overall output loss as well. Tim thinks, if I am not mischaracterizing him, that prioritizing price stability maximizes overall output, and whatever distributional issues might result from the policy are overwhelmed by the aggregate wealth it enables. To his credit, he favors transfers to ensure the maximized wealth he envisions addresses some of those distributional issues, but — again, I hope I am not mischaracterizing — he does not favor holding prioritization of price stability hostage to the policies he might prefer to address any distributional costs. Both our views on the positive economics and our normative priorities are very divergent, I’m afraid.

    The interactions between intra-nation-state policies that try to stimulate demand or address national inequality and international effects on, say, less-developed countries are important but complicated. I think we need to talk very specifically to make headway on those questions. For example, some intra-national demand-stimulating policies would clearly be harmful to both rich and poor in developing countries (say, imposing tariffs on Chinese exports to advantage domestic production). Other policies might be beneficial to actors in developing countries (say helidrops that spur import demand in developed countries, as well as demand for domestically produced goods), but that benefit to developing countries may limit their effectiveness in developed countries. There are short-term vs long-term questions to consider as well. One might argue, in the spirit of Bretton Woods, that an international system should strive to maintain overall trade balance, even when that has clear short-term costs to some countries that would otherwise run a production-stimulating trade surplus. One might argue the opposite too, and point to salient examples of countries that developed rapidly precisely by running large surpluses and using the most financialized developed economies as “piggy banks”, in your words. I think we need to consider policy regimes (both shorter-term demand management tools and longer-term balance of trade management systems) in specific detail to make much headway adjudicating these (hard) questions. My tentative allegiances are to Bretton-Woods-like arrangements that permit some development-enhancing or demand-stimulating imbalance, but create incentives to prevent those imbalances from cumulating too large or too long. There are demographic issues to consider, and important differences between small and large economies on these questions, however.

    Peter K — I don’t think that Brazil’s Bolsa Familia is very close to a UBI, but I could be mistaken, I’ve done little to research it. I believe it is a means-tested transfer program. That’s not to say I don’t support it, it seems to have been very successful, and I’d hate to let some monomaniacal preference on my part cast aspersions on a program that has proven achievable and effective. But the U in UBI is a very important part of that policy, and of why I think a UBI would be effective in say a US context.

  20. stone writes:

    Steve, thanks for the clarifications.

    One thing though -I do think it is so important to make a clear distinction between on the one hand the national monetary authorities of countries accumulating US treasury bonds as a way to stabilize exchange rates (as done by China and Japan) and on the other hand, individuals syphoning vast sums out of developing countries to private accounts in UK, Switzerland, USA etc. It is the macro effects of that latter phenomenon that I think is so crucial for understanding “the great moderation” and the way that it turned around what was happening in the 1970s. The 1980’s runaway currency slide in developing countries most definately did not spur economic growth there. As an example, from the time of independence in 1960, Nigerian GDP increased 135% during the 1960s and then 283% during the 1970s. By contrast it shrank by 66% in the 1980s. The Nigerian Naira / USD exchange rate went from 0.78 in 1980 to 2.83 in 1985 to 8.94 in 1990 to 102.24 in 2000.

    If we don’t have “wealth friendly” policies, we won’t attract hot money inflows from capital flight. So we screw up our own economy with “wealth friendly” policies so as to screw over the global economy to our advantage.

  21. stone writes:

    I’m no fan of above inflation interest rates (my pipe dream would be for price stability to be acheived by instead using an asset tax to moderate credit expansion). BUT I’m really struck by Steve’s confidence (@15) that it is a trivial matter for the fed or Bank of England to reverse QE to such an extent so as to revert to a pre-2008 fairly typical situation with say interest rates 3% above inflation. I take the point about the fed not needing to worry about mark to market insolvancy but I’m just trying to imagine how a QE reversal could pan out. Imagine, like you say, the fed sold enough “one year fed sterilization bonds” by auction to mop up so much bank reserves as to make bank reserves once again scarce enough to garner an interest rate 3% above inflation. After the year was up, the fed would need to roll over those “fed sterilization bonds” with a fresh issuance of that amount plus 3% and so on every year for perpetuity ???? Would that not rapidly get out of hand ????

    I had rationalized to myself the apparent effect of QE on asset markets by thinking that QE was a sort of bridge burning demonstration by the central banks to assure holders of interest rate sensitive assets that the central banks would not really raise interest rates appreciably in the forseable future. Am I in a total muddle about that?

  22. Stone — “Reversing QE”, if it happens at all, would take a long time. But central banks that pay interest on reserves never need to. As I’ve written about before, I think we (at least in the US) are on the “floor” system of monetary policy for the indefinite future. The CB will not (at least not quickly) “mop up” its QE created reserves. It will just pay interest on them. The fiscal difference between this “floor” policy and traditional monetary policy is negligible small, since in traditional monetary policy banks held minimal reserves. (The effect on CB balance sheets might be nonnegligible if no procedural changes are made, but I’m confident we’ll either find ways of letting accounting follow reality, or else remedy the “ratchet” in which routine transfers between the CB and Treasury only go one way. (CB distributes profits to Treasury, but has no routine or automatic way at present to call on Treasury to cover losses. But substantively, the govt can’t afford to treat the CB like a limited liability company, upon which it may exercise a bankruptcy option via failure to support.)

    Term deposits, reverse repos, and CB bond issues are entirely unnecessary for interest rate policy. Interest rates can be raised at will without those tools. However, some people argue that the split between base money and other government securities is important even under a floor system, where base money pays the same or better interest than equivalent-term Treasury debt. As far as I’m concerned, that case has not bee strongly made, but there are interesting (and often paradoxical wrt ordinary intuitions) conjectures, e.g. related to the institutional difference that only banks can hold interest-paying reserves while anyone can hold Treasury debt.

    Term deposits, reverse repos, and CB bond issues permit a CB to manage the split between reserves and marketable debt at will, in a manner entirely independent of both interest rate policy and the quantity of salable securities on the CB balance sheet. Note that the different instruments have different institutional details, with outright bond issuance the “purest”
    substitute to selling marketable securities from the CB balance sheet. But ultimately, neither a central bank’s policy rate nor it’s ability to manage the quantity of base money outstanding are dependent upon the contents of it net equity on its balance sheet.

  23. (All that said, I don’t think we’ll ever let CB net equity positions go formally negative. We’ll modify accounting rules, as necessary and quite reasonably, to prevent that. Money is a coordination equilibrium, and a coordination equilibrium is a self-fulfilling confidence game. A CB balance sheet is not remotely analogous to a firm balance sheet, but lots of people mistakenly think it is, and that mistaken view could become self-fulfilling if a “weak” balance sheet led to a flight from the CBs money. CB accounting rules already defy firm norms to prevent this — in general, holdings are not marked to market even though CB’s routinely sell assets rather than hold to maturity, for example. My “goodwill” proposal for a helidrop counterasset might sound radical, but it is quite in the spirit of existing CB accounting, which resists showing equity losses not because the accounting is fraudulent, but because “equity losses” to a CB just don’t mean what they mean in any other context. If there is something misleading, it is the attempt to graft conventional financial accounting forms onto CBs that don’t resemble conventional firms at all. The US Fed understands this. For example, while they do publish a financial balance sheet weekly, it is buried beneath a CB-specific “factors affecting reserve balances” statement, by which the Fed emphasizes that the most meaningful position of the bank has to do with the quantity and character of the liabilities it has issued. The asset side of a CB balance sheet is most interesting wrt the character of assets — maturity, whether they carry pvt sector risk — than wrt a quantitative comparison of value between the asset and liability sides.)

  24. stone writes:

    Steve, if the method for raising interest rates relies on paying interest on reserves (rather than on scarcity of reserves), and the source of funding for paying that interest on reserves ends up being transfers from the Treasury to the central bank (because the returns from the fed’s portfolio aren’t enough), then I find it so hard to see how previous QE doesn’t really set a limit on achievable future interest rates. If there are now $2.7T in bank reserves, then paying say 5% on that would be $135B wouldn’t it? Would the Treasury really unflinchingly transfer that sort of amount to the fed each year? What is more that amount would be increasing year after year by the compounding interest rate.
    Hard as I try, I still can’t help thinking that QE “bakes in the cake” low interest rates for a long period ahead.

  25. stone writes:

    Thinking about what I wrote @24, I guess if the Treasury were transferring say $135B a year to the fed to pay for interest on reserves, then that would be funded by issuance of extra Treasury debt which would drain some of the bank reserves and sort of shift things back over to being a direct Treasury liability. I guess the overall interest burden would be no different than what it would have been had no QE occurred. The only sticking point is whether there is some institutional issue with such large transfers from the Treasury to the fed -and you seem confident that there isn’t.

  26. Tim Young writes:

    Steve, you may have misunderstood my point about the frequency domain. The point is that its mandate is set in a democratic way, but over a longer timescale than the life of a government, so that, providing that the central bankers care about their mandate as envisaged (which I think they have not been doing, especially in the UK), the central bank and value of money is exogenous as far as the incumbent government is concerned. And I like it that. Besides money neutrality theory, my instinct is that money is a measuring rod, and changing that is more likely to be a matter of appearance than substance. I would for example, make the same point about grade inflation in schools (in the UK at least). You can argue that grade inflation increases utility by taking the hard edge off assessment for students, and parents and (our) politicians love the appearance (“the best results ever!”), but it makes it hard for employers to judge potential employees and may be leading to a reduction in the quality of education that is making our competitiveness worse.

    Given the previous paragraph, it will be no surprise to you that I dislike your attitude to accounting (I would argue that un-rigorous accounting has contributed both to the financial crisis and the slow recovery since – the response to the crisis being to relax MTM/fair value accounting). Modifying accounting rules changes nothing in reality. And I hope your friends have not read your next few sentences, which suggest that you approve of saying whatever is necessary to keep them happy! But you are wrong anyway; you have absorbed too much of the “money as a social construct” waffle promulgated by historians and sociologists who became popular in the post-crisis period because their writing was more accessible than getting to grips with the mechanics of money issuance. The whole point of central bank assets is that the value of money need not depend on a self-fulfilling confidence game. If the public loses confidence and starts to try to trade money for something else, the central bank can immediately resist that, perhaps by selling assets but ideally by not rolling over its short term refinancing. If you think about this, because the central bank typically receives more interest on its assets, at some point the central bank can make the real value of its currency practically infinite. In fact, “fiat” money can be harder than a gold standard. But only if the people running monetary policy have their ears plugged and their captain is tied to the mast.

  27. Tim Young writes:

    Stone @24. I totally agree. If you can read FT comments:

    Stone @25. Yes, if the US Treasury borrows $135bn a year to cover the Fed’s reserve interest outlay, that temporarily (until the treasuries mature) mops up the extra base money issued by the Fed as interest on reserves, but in the meantime those treasuries themselves pay interest which has to be funded from somewhere. This is extra interest, representing the cost of forestalling the problem of realising the losses on QE. It’s still a Ponzi game. And what US government wants to be seen to be raising about $450 dollars per year from every citizen to pay to “bankers”?

    In the teeth of the financial crisis, QE was a reasonable policy, to expand the stock of base money to meet the demand for base money as a safe asset, which would have otherwise interfered with its role as the medium of exchange, as in the 1930s. But extending QE to stimulate the economy was always based on dubious reasoning, and has pushed the US (and UK, not to mention Japan) towards a high inflation trap that is going to take some effort, manifested in hardship for citizens, to avoid.

  28. Tim — Central bank money is backed by nothing of intrinsic value other than a government’s capacity to act strategically or coercively to render it so. Modern CB money is backed by Treasury debt which is merely a promise to pay more modern CB money. Governments by taxation or establishment of a standard, network effects, a self-fulfilling confidence game ensure that money is sought after and therefore redeemable for pvt sector goods and services. Gold standards CBs also had no intrinsic value to offer: the monetized value of gold was always much greater than it could possibly have been if the metal were desired primarily for industrial or decorative purposes. The extra increment in value derived precisely from a self-fulfilling confidence game. As Mencius Moldbug used to put it (not a guy I share much political common ground with), money is the bubble that doesn’t burst.The more civilized the government, the more it relies upon money as a coordination equilibrium rather than money backed by the hard asset of escape from government coercion. (The simplistic tax-driven money story, where a govt imposes a head tax and people value money as a stay-out-of-jail card comes directly from systems of colonial exploitation. Money as a standard conjured and cajoled by governments only in small part by the threat of coercion for tax nonpayment is more descriptive of civilized societies.)

    I think we’ll have to leave it there, or I will. Of course feel free to have the last word. In my view, you rely a very great deal on your gut feelings, intuitions, and very pat ideologies about these matters, but trouble yourself very little to make a case. Those who disagree with you have been tainted by postmodern “waffle” or whatever. I suppose you may perceive my views as similarly unfounded and dogmatic. We’ll let others stand and judge, to the degree that others have the patience to read this back and forth.

  29. Tim Young writes:

    Steve, you are making a typical American’s mistake about central bank balance sheets. Of the developed country central banks, only in the US and Japan is base money routinely backed mainly by treasury bonds. In the UK and ECB, base money is supplied, distributed and backed by repo loans to banks. That gives them an iron grip on the value of money (which was why I wrote “ideally by not rolling over their refinancing”). All the same, unless faith in a central bank’s base money is lost very fast so that the value even of base-money-emitting bonds declines fast, it should still be possible for the central bank to contract the stock of its money fairly quickly by either selling treasuries or not buying more as they mature. Of course if the government refuses to repay its loans as its treasury debt matures, then all bets are off, but that probably represents a bigger crisis than a fall in the value of money anyway. Don’t you see, Steve; it is the very fact that modern base money is backed by debts in modern base money that makes it potentially so hard? Under a gold standard, if the real value of gold declines, so does the real value of base money, and there is not much that the central bank can do about it. Under a base money debt standard, however, once interest is due on the initial central bank loans of base money, the value of the currency unit can be driven to whatever the central bank can demand from the marginal debtors that they can hand over without going bust. No need for self-fulfilling confidence, coordination equilibrium, taxation or any of that fancy stuff.

    I too fondly recall Mencius Moldbug and his comments on Brad Setser’s blog.

  30. stone writes:

    Tim Young @27, I sort of recanted @25 what I wrote @24 because I think I belatedly grasped Steve’s point that the only difference between interest on reserves and Treasury debt interest is the need for a (perhaps meaningless) hand over between the Treasury and the central bank in the interest on reserves situation as opposed to a direct payment from the Treasury to the bond holders in the “normal” situation with treasury bonds in place of the excess reserves. It is all a wash so long as the Treasury is happy to see it that way (I hope I’m not in a muddle or stating the obvious).

    Imagine that there was never QE and so there was say $14.9T of outstanding Treasury debt and just $1B of bank reserves and the treasury paid $750B per year in interest. The alternative is that there is QE such that there is $12T of outstanding Treasury debt and $3T of bank reserves so the Treasury then pays $500B directly as Treasury debt interest and transfers $150B to the central bank so that the central bank can pay interest on reserves. So the Treasury is not any more out of pocket due to QE and it is all a wash?

    If this is true then does QE achieve anything at all????

    Perhaps it is really the size of the combined stock of Treasury debt and/or bank reserves that “bakes in the cake” the need for future low interest rates?

  31. Tim Young writes:

    Stone, you were wrong to recant. It is the fact that the Treasury may not “see it as wash” whether they stump up money to the central bank, in which case refusal would precipitate only a relatively arcane constitutional crisis, or to private sector treasury holders (because, eg, the central bank reduced its outstanding reserves by selling treasuries), in which case refusal would mean default, that is the problem. If the central bank is truly independent so that the value of its money is effectively exogenous to the government, whatever the government’s “need” for low interest rates, the government may well find that, like Greece, its debt yields rise as it runs up debt.

    QE is not without achievement. As I wrote @27, QE was reasonable during the financial crisis, and perhaps for a while afterwards to lower long-term interest rates in an anti-cyclical attempt to stimulate economic activity with short-term interest rates stuck at the lower bound. But there comes a time when a prudent central bank should accept that on-going economic weakness is probably structural rather than cyclic, and start looking to unwind its QE. I would agree with the BIS that inflated asset prices can pose a threat to price stability ( ), so my bias here would be to gradually reduce QE as long as inflation is not persistent significantly below target. Of course this will make the central bank unpopular, so the temptation for its decision-makers is not to do it.

  32. stone writes:

    Tim Young @31, I guess it all boils down to whether it matters whether or not the central bank is somewhat independent from the Treasury or really consolidated with it. If the central bank and the Treasury are really consolidated, then perhaps QE is a wash. But QE means that the central bank looses the capacity to impose high interest rates unless the Treasury is fully compliant in that effort. So perhaps QE might be thought of as a demonstration by the central bank that it is relinquinshing the independence of its capacity impose high interest rates?

    My impression was that QE has been ascribed some (mystical) property over and above ensuring low interest rates. Back in the 1940s, to help the WWII effort, the Bank of England and the Fed pegged both short and long term interest rates to very low levels BUT the QE we have had in recent years has not been a simple interest rate pegging exercise like that. The QE of recent years has been over and above what would be needed for that. Hasn’t it been done in the belief that QE has a channel of action somehow independent of interest rates? The only way I could rationalize that was as a bridge burning demonstration to reasure people that interest rates could not be easily raised in coming years. But Steve (and other people) say QE creates no impediment to subsequent rate rises.

  33. Tim Young writes:


    “If the central bank and the Treasury are really consolidated, then perhaps QE is a wash.”

    “But QE means that the central bank looses the capacity to impose high interest rates unless the Treasury is fully compliant in that effort.”
    Incorrect. As long as the central bank has assets that it is able to sell without government approval, it can maintain independent control of monetary policy. If QE was conducted via helicopter drops instead, the central bank would have no such marketable assets, in which case it would need the compliance of the treasury to tighten monetary policy, except by raising the interest rate paid on reserves, which, as I say, is a Ponzi game (ie makes the problem bigger down the track).

    In time order, central banks have argued that QE is expansionary because:
    (1) It increases base money supply
    (2) It increases broad money supply (if the central bank buys from the non-bank private sector)
    (3) It lowers long term rates and encourages investment
    (4) It boosts spending via asset prices and the wealth effect
    The fact that the rationale keeps changing as the policy is maintained is enough to make one suspicious!

  34. stone writes:

    Tim Young @33, I was under the impression that the current situation with the Bank of England or the Fed or the Bank of Japan, was that there is such a surfeit of bank reserves (and the central bank’s portfolio of assets are of sufficient duration) so that even if all of their assets were sold off, it would still leave enough excess reserves for short term interest rates to still be stuck at the zero lower bound. Basically forced selling by the central bank means it gets lower prices than it originally paid for those assets when it conducted QE.

    So the central bank has to rely on interest on reserves to get interest rates off the zero lower bound and the central bank only has limited income from its portfolio of assets which means that paying higher interest on reserves would entail transfers from the Treasury to the central bank. Please correct me if I’m in a muddle with this.

  35. Tim Young writes:

    In fact, Stone QE central banks have taken some steps to protect their independence. The BoE has been most careful not to get into the QE trap. In the UK, QE is actually on the balance sheet not of the central bank but an SPV called the Bank of England Asset Purchase Facility Fund Limited, legally indemnified against losses by HMT. If HMT refused to pay up under the indemnity, I would presume that that would mean that the UK government was in default.

    Since the Fed bought a lot of MBS and spreads have narrowed, I dare say that their assets are presently worth far more than their liabilities. But the Fed also has an arrangement in which they can stop remitting positive net interest to the Treasury if their balance sheet shows negative net value (I am not sure of the precise details of valuation etc): (note my comment there too).

    I should think that the BoJ is in the worst hole, but they have practically surrendered their independence anyway. Shame, and their QQE does not seem to be doing them any good anyway.

  36. Caleb writes:

    Goodd webb site you have here.. It’s hatd tto find ggood quality writing likke yours these days.
    I honestly appresciate peopole like you! Takke care!!

    Also visxit mmy web page: esy (Caleb)

  37. stone writes:

    Tim Young @35, as far as I could make out, that FT link simply said that if the Fed were to choose to pay high interest on reserves, then they would “fund” it by racking up an exponentially growing IOU to the Treasury whilst ever increasing the stock of bank reserves for which they needed to pay interest for. Does that really sound like a tidy solution?

  38. Tim Young writes:

    @Stone, it’s the other way round. The IOU is negative; in other words, the US Treasury accepts a debt to the Fed, to be paid down when (and if!) the normally positive net interest flow that the Fed enjoys (but remits to the US Treasury – ie seigniorage) resumes. It assumes that the Ponzi game will stop at some point short of collapse. What happens if it does not is not clear – ie whether the negative IOU has the status of a US government debt which they are obliged to pay. What is clear, however, is that if the Fed’s negative net interest does reverse, the US Treasury will be deprived of seigniorage income from the Fed for some time while the Fed draws on the negative IOU to rebuild its capital. Hence the remarks in the post about the effect of this rule on the creditworthiness of the US sovereign.

  39. stone writes:

    Tim Young@38, so currently because interest rates are low, the Fed is building up a sort of war chest that in principle it could draw upon if it needed to in order to pay high interest on reserves. It still stands that if and when the Fed were to decide to pay high interest on reserves it would quickly spend that stored up war chest and then start owing the Treasury and it would be a runaway exponentially growing cost with no end in sight?

    We are told that the central bank still has the capacity to use interest on reserves to achieve say 1990s style interest rates. What would things look like after say a couple of decades of doing that ????

  40. stone writes:

    I’m still left with the impression that debating the injustices of “hard money” is now of merely historical interest because QE has irreparably broken the hard money system anyway. We have a ZIRP future so campaigning against “hard money” is misdirected. Instead nowaydays other means are being used to acheive the injustices that are being blamed on hard money. The ire being directed against hard money policy might be better directed against regressive fiscal policies such as say consumer taxes and low pay income taxes paying for bank bailouts or whatever.

  41. Tim Young writes:

    Stone@38, Actually, no. The Fed is continuing to remit its positive net interest income to the US Treasury as long as it lasts, so the Fed would start with less net worth.

    I am not quite as pessimistic as your scenario, however. I dare say that the natural (ie non-inflationary short-term interest rate will be low for a long time, and anyway the Fed can operate policy normally, provided that it has the cooperation of the US government (meaning its willingness to meet the cost of doing so). Moreover, even if I deplore the rest of their politics, the US has enough politicians like Rick Perry to at least raise the alarm about inflation. But what does worry me, which is why I react to this kind of post, in this case from Steve, is that we might drift into a mess simply because influential people do not understand central bank balance sheets and deny that there could be a problem.