Friday, 12 October 2012

What do people mean by helicopter money?


Following a speech by one of the front runners to replace Mervyn King as Governor of the Bank of England, there has been renewed talk about helicopter money. Helicopter money involves the central bank printing money, but that in effect is what Quantitative Easing (QE) does, so what is different about helicopter money? There seem to be two rather different things that people might have in mind.

The first difference is where the money goes. QE, in the UK and to some extent in the US, involves the central bank printing money to buy government debt. Helicopter money is like the central bank sending a cheque to everyone in the economy. The second difference is whether the creation of new money is permanent or temporary. QE, if you ask central bankers, is temporary: when the economy picks up and there is the first sign of inflation, QE will be put into reverse (except, just maybe, in the US). Helicopter money is thought to be permanent: the central bank is sending out cheques, not loans.

Let’s take the permanent/temporary issue first. As I have argued before, calls for money creation to be permanent are in effect calls to increase inflation above target at some time for some period. The reason why most people believe QE is temporary is because (with the possible recent exception of the US Fed) central banks are sticking firmly to their inflation targets. There may be very good reasons why central banks should instead allow inflation to exceed those targets as the economy recovers. But if this is the issue, why not just call for higher inflation? Surely it makes sense to be explicit about what is trying to be achieved, particularly as the benefit involved in higher inflation for the real economy comes from increasing inflation expectations. We gave up money targets long ago, and quite rightly so. For the central bank to destroy some proportion of the government debt they now own and just hope this gave them the amount of extra inflation they desire would be like going back to money targets.(1)

The first issue, of where the money being printed is going, is more interesting. It reflects an understandable view that it would be better to print money and give it to consumers who would spend it (helicopter money), rather than using it to buy government debt (QE) which may reduce long term interest rates which may help stimulate the economy. The second route has been tried and has not been that successful, so why not try the first route?

It is useful to think about the circumstances in which the two routes are exactly equivalent. To focus on this, let us assume helicopter money is temporary: the central bank sends out cheques, but the government says it will get the money back in a few years time by raising a poll tax. (This is like the proposal from Miles Kimball.) If consumers are Ricardian, they will save all the amount of the cheque, because only by doing so can they pay the future poll tax without cutting their consumption. How will they save the money? Let us suppose they buy government debt. Then this is exactly the same as QE, except that consumers hold government debt temporarily instead of the central bank. This seems to be what David Miles had in mind in the speech that Stephanie Flanders refers to here, when he says: “If helicopter drops of money are reversed when their impact shows up very largely in prices and not in activity, the economic difference with conventional QE largely evaporates.”

Yet we can now see why in reality the two may not be equivalent, because consumers may not be Ricardian. In particular, some may have been asking their local bank for a loan to buy a car, and the bank had refused because it has become very risk averse since the crisis. For these credit constrained folk, the central bank’s cheque is just like the loan they couldn’t get. So they use the cheque to buy the car, and reduce their future consumption to pay the poll tax later. Instead of buying government debt, they have bought something real, which will increase aggregate demand for sure.

If this is the reason people call for helicopter money, then I have a lot of sympathy but only one problem:  what difference is this from an expansionary fiscal policy combined with further QE? Instead of the central bank sending people cheques, the government can send the cheques using money borrowed by selling debt, and the central bank can buy the debt by printing money (i.e. QE). In this sense, helicopter money is just another name for a fiscal stimulus combined with QE. We have the QE, so why not call for fiscal stimulus rather than helicopter money?

 (1) There may be a case for announcing (in some form) higher future inflation, and then destroying the debt, because that commits the central bank to higher inflation. What does not make sense is to destroy the debt and pretend you are not going to increase inflation at some point.

17 comments:

  1. Even without credit constraints direct transfers can have large efficiency effects. For example, if a transfer allows someone who could not otherwise make their mortgage payments to remain in their home, the deadweight losses from foreclosure (which are substantial, as much as half the value of the loan) are avoided. This can even benefit creditors - in fact helicopter money can facilitate implicit loan modification. Fiscal policy cannot do this except in a very patchy and piecemeal way. Infrastructure spending, for instance, will be targeted at some locations and industries and leave most debtors unaffected until the effects of the policy gradually permeate through the system. Unless, of course, the fiscal policy itself involves direct transfers.

    More generally, no two policies with different distributive implications can be equivalent with respect to their macroeconomic effects. Highly aggregative models assume away these distributional effects.

    About the permanent/temporary issue, I don't see it. One could maintain an inflation target, and finance direct transfers by the sale of bonds (or the interest from bonds). You tolerate higher interest rates to maintain the given inflation target but counteract this with expansionary transfers. Why does the inflation target need to change?

    ReplyDelete
  2. > We have the QE, so why not call for fiscal stimulus rather than helicopter money?

    That's easy! Because consumers spend money on things that they want, whereas the government spends money on bridges to nowhere. Sure, from an aggregate demand viewpoint, these are equivalent, but from an aggregate utility viewpoint, there is a big difference. You may as well ask why US consumers don't donate their money to the government to spend for them.

    ReplyDelete
    Replies
    1. Re-read the article. Wren-Lewis says the government would send everyone a check, not build a bridge to nowhere.

      Delete
  3. 1) QE buys current debt, helping holders of those bonds. Helicopter money is QE wth the fresh deficit, distributively neutral.

    2) Fiscal stimulus is targeted by the govt and faces public choice issues. Helicopter money is person-blind.

    3) Though it ultimately comes down to what the monetary policy target of the central bank is, permanence of monetary increase is the default assumption in a helicopter drop. Reversal is the default assumption in pure QE.

    Infrastructure spending etc. are RoI and 'role of the state' decisions and must proceed in a normal fashion, without AD management as an additional consideration. AD management best left to PAYE/ consumption taxes/ bank transfers, or something equally wide-reaching and distributively neutral.

    ReplyDelete
    Replies
    1. PierGiorgio Gawronski15 October 2012 02:22

      1) Don't see large implications of this
      2) Fiscal stimulus could also be person-blind. See the comment above: "Wren-Lewis says the government would send everyone a check, not build a bridge to nowhere". More importantly, you can't have 'helicopter money' unless through the government (fiscal authorities). So the alternative described by SWL is not about who gets the money, but about how it is financed. Should fiscal authorities issue new bonds which the FED would buy back on the secondary market? (QE + fiscal stimulus). Or should the fiscal authority have a current account at the FED, where the FED would credit positive money balances in return for nothing? SWL asks if it make any difference.

      Delete
  4. Cash is not the form of money that matters in our modern economy. Credit - i.e.: numbers in bank accounts - is what matters.

    To stimulate the economy, we need to restore credit growth - so that the stock of credit keeps up with population growth, growth of economic activity and a low level of inflation.

    Credit grows when borrowing exceeds repayment. Borrowing is normally stimulated by low interest rates.

    However, low interest rates have not been sufficient to restore credit growth in the current environment of private sector deleveraging (the aftermath of a bubble in the private sector debt:income ratio).

    That is why the government must step in with additional fiscal stimulus to restore robust credit growth. Government borrowing creates both bonds (debt) and bank deposits (credit).

    The credit growth (a rise in the stock of money) will restore economic growth (a rise in the flow of transactions) and hence raise the governments tax revenue in the future - sufficient to cover interest payments on bonds.

    It is a political decision as to whether the fiscal stimulus is spent on public spending or on a tax cut (or a mix of both).

    ReplyDelete
  5. PierGiorgio Gawronski13 October 2012 03:43

    The difference between Strategy (1) Helicopter Money, and Strategy (2) QE + dG=dB (where B = Gov. Bonds), is that the latter implies a higher public debt (although bought back by the central bank), and this would have an impact on expectations. And expectations are crucial.

    First, markets participants will observe an increase of the debt/GDP ratio; the bond vigilantes) will not like it. Do bond vigilantes exist? Well, in the Eurozone they do.

    Second, if you believe in non-Keynesian effects, the greater risks of your central bank turning like the BCE in the future, and thus the greater risks of future bond vigilantes action, macro instability and disruptions, may reduce aggregate demand today, even in countries such as US, UK, Japan.

    Third: an increase in the supply of B will look like a temporary expansion, so Ricardian consumers will save a lot. On the contrary, an increase in M will look more like a permanent choice, a permanent expansion: similar to the first strategy plus the central bank burning its bonds (see http://worthwhile.typepad.com/worthwhile_canadian_initi/2010/06/why-its-a-really-good-thing-that-the-ecb-has-overpaid-for-greek-junk-bonds.html ): thus Ricardian consumers will spend more now. And yes, inflation expectations will be higher. I know: in QE + dG the supply your ideal government announced that the current fiscal expansion (whether dB or dM) will be financed by a future poll tax (not by a future increase of the inflation tax). But I believe it is an unrealistic, ad hoc assumption. More likely, a real government would not explain how the current expansion will be financed, and anyway there’s uncertainty: so dM (as opposed to dB) would increase the probability that in the future the financing would come from inflation. Yes, I see that in Strategy (2) the dB element is only apparent, since it is held by the central bank, which is a public institution. But public perceptions count.

    So in the end Mr. Helicopter Money would be much more powerful than Ms QE+dB in the Eurozone, and only moderately more powerful elsewhere.

    ReplyDelete
  6. I absolutely agree that helicopter money is QE plus fiscal stimulus.

    I think the reason some people talk about helicopter money is because they live in the USA. They know that Congress will not approve any more fiscal stimulus (just the opposite) so they want the Federal Reserve to implement fiscal stimulus.

    This would clearly be illegal under current law. Also unconstitutional under any law. I think the point is that those who advocate helicopter money wish the USA had a different fiscal authority, because the one we have is controlled by Republicans not named Bernanke.

    ReplyDelete
    Replies
    1. You're mistaken. Singapore implemented a helicopter drop throughd transfers to household bank accounts, monetizing the resultant debt, about 2 years ago. One major central bank to discuss it was BoE, which released a discussion paper considering monetizing PAYE tax cuts, and finally concluded against it.

      There's a strong intellectual foundation behind the call for helicopter drops, which has nothing to do with the shenanigans of the US political system.

      Helicopter drop is indeed QE + fiscal stimulus, but both are motherhood terms because of aggregation. As Rajiv Sethi points out, distribution matters non-trivially.

      Delete
  7. "the government can send the cheques using money borrowed by selling debt, and the central bank can buy the debt by printing money (i.e. QE)"

    In the U.S., the fiscal deficit is running about 8% of gdp. Growth in government ransfer payments have accounted for a large percentage of Disposable Income growth. Meanwhile, the Fed is buying the bulk of term Treasuries. How is this different than what Wren-Lewis is proposing?

    ReplyDelete

  8. Here is how to do a helicopter drop in the U. S. context. The President requires the Treasury to create 4 trillion dollars of coin signorage, delivers it to the Fed, and has the Fed mark up the Social Security Trust Fund by 4 trillion, giving the the Trust Fund a surplus of 6.6 trillion dollars. The President then goes to the Congress and asks "Why are we taxing employers and employees 10.4 % of every paycheck to bolster a retirement fund that already has 6.6 trillion dollars? Give workers and employers back their payroll tax. They earned it, let them spend it or save it, or pay down debt with it." Require that enabling legislation specify a target for re-establishing the payroll tax, perhaps in stages, as the U. S. reaches targetted NGDP growth. The pressure on Congress to do this would be strong.

    One can look at this as debt forgiveness on obligations going forward. Of course, Bernanke could sterilize this move by raising interest rates, but he has been calling for help from Congress for some time, and republicans do like tax breaks.

    ReplyDelete
  9. It's only stimulus to the extent it circulates. So long as the Fed pays interest (>0) on Excess Reserves* while we're in a ZLB situation(r=0), you're not going to get a sustained multiplier effect.

    You can only throw money at the problem to the extent that money can solve the problem: A buys from B who buys from C who buys from D who buys from A, et seq. When the path is A buys from B who pays Jaime Dimon or Vikram Pandit's compensation for being Such A Great Leader, you accomplish less than the value of the coinage.


    *I'll quibble paying interest on reserves, too--it's rewarding a bank for follwing the rules that enable it to be bailed out with impunity by idiots like Timmeh--but paying interest to encourage "intermediaries" not to intermediate is something that reeks of crony capitalism and/or the inability to do math. So long as the Fed insists on doing that, I'm applying Occam's Razor and declaring that the major U.S. banks are all insolvent.

    ReplyDelete
  10. This discussion detracts attention from the main macroeconomic question - what did induce the economic slowdown? Likely, the financial crisis was just one of many features of the economic fall. Currently, the treatment seems more dangerous than the desease, which in turn, is not clear.

    ReplyDelete
  11. Arent unemployment checks an example of helicopter money? The unemployed are more likely to be Ricardian. The bigger challenge is to prevent moral hazards from popping up.

    ReplyDelete
  12. Yes, in theory, HeliMoney and QE+dB are the same. The difference between those macroeconomic tools does not come from their function (which is the same), but from the people who hold them.

    HeliMoney is printed by the central bank (an independent body with supposedly some economic knowledge).
    In the example of SWL, HeliMoney can be reversed (if there is inflation) with a poll tax imposed by the Government (a political body, prone to electoral considerations, and whose economic knowledge is variable).

    Fiscal policy is decided by the Government and may, or may not, be supported by the central bank with QE.
    If there is no inflation, or if the economic situation is too bad, it is imaginable that the central bank simply cancels the gov debt.

    In both cases, the outcomes may be the same (either a permanent or a temporary monetary expansion). But their likelihood is very much different.
    The permanence of the monetary expansion looks much more likely in the case of HeliMoney than with QE + dB.

    (Note that it depends on who sits respectively on the central bank and the government. It's long been assumed that central bankers are more conservative than governments, but that's an hypothesis that's been proved wrong in a very recent past.)

    In practice, small nuances may have big effects, and this is why economic theory can never be taken without a pinch of salt.

    ReplyDelete
  13. Thanks for the explanation. I guess the term helicopter money has a different meaning than I thought.

    ReplyDelete
  14. Parabens pela estrutura e conteudo de seu blog, Forte abraço Renato Artesanato em MDF

    ReplyDelete