Winner of the New Statesman SPERI Prize in Political Economy 2016


Wednesday, 12 September 2012

Why not finance fiscal stimulus by printing money?


This is a friendlier version of an earlier post that was marked for economists.

Q: Everyone keeps saying that the government cannot boost the economy by increasing its spending because we need to reduce the amount of government debt. Now I know we have talked a lot about why this view is mistaken, but what is stopping the government paying for additional spending by just printing money?

A: Do you know the technical name for that idea?

Q: Money financed fiscal expansion rather than bond financed fiscal expansion? And didn’t Friedman have this idea involving a helicopter?

A: Good. And what is meant by Quantitative Easing, or QE for short?

Q: The central bank buying government debt by creating bank reserves?

A: That will do. And what does the term ‘monetary base’ or ‘high powered money’ mean?

Q: That is the amount of money created directly by the central bank, which is either cash held by the public, or reserves held by banks.

A: So if we have a policy that involves both bond financed fiscal expansion, and QE, what does that amount to?

Q: Ah! Is it the same as money financed fiscal expansion?

A: It looks that way. But to properly answer that question, we need consider another. Now what happens if a government spends like there is no tomorrow, and finances it all by printing money?

Q: We get inflation of course. But that is only because the government would be increasing demand even when we have got to full employment. Today when there is deficient demand and high unemployment, adding to demand should not cause an inflation problem.

A:  I agree. But suppose money financed fiscal expansion, or bond financed expansion plus QE, works, and we get back to what you call full employment. What happens to all that money the central bank has created?

Q: Well now there would be too much money chasing too few goods, so the central bank would have to put QE into reverse. Otherwise we would get inflation.

A: OK. Would the world look any different after QE had been reversed, compared to a policy that had just involved bond financed fiscal expansion in the first place?

Q: So what you are saying is that bond financed fiscal expansion with or without QE looks just the same in the long run, as long as QE is reversed.

A: Well you have just said it, having gone down the path I have subtly led you down. Now what have I always told you about the time frame involved with issues involving government debt?

Q: I know I asked the first question, but now I’m doing all the answers. You really should be using different letters from Q and A, like maybe S and T.

A: Ah, the famous dialogs between Socrates and Theaetetus. How nice it is to teach Students in Tutorials who also study philosophy. But what is the answer to my question.

Q: That issues involving government debt are long term, not short term. So if using money to finance extra government spending just involves a temporary increase in money, and no permanent reduction in debt, what is the point?

A: Indeed. Now there might be some point if the government or central bank wanted to signal with QE that it did intend to raise inflation above the target level for some time, which is the same as saying that it would not reverse all of the QE.

Q: But haven’t both the Bank of England and Federal Reserve said they remain totally committed to their inflation targets?

A: Effectively yes, and recently they seem to be content to see inflation below target, so it would be very odd if they were using QE to signal the opposite.

Q: So does that make QE a complete waste of time? And why do you say there would be some point in having higher inflation in the future?

A: I think this is an excellent point at which to end things with some reading for next week. I suggest this paper that Michael Woodford has just written. After that have a look at these blogs (BruegelKimball and DeLongSerlinHamilton, Gagnon) on the effectiveness of QE, and this on nominal GDP targets and fiscal policy. (Try and keep the issues involving instruments, mechanisms and targets separate if you can.)

Q: Not another paper by Woodford! The last time you made us read one of his papers it went into maths just when it was getting interesting.

A: I’m tempted to say that without maths there can be no true knowledge, and this is just the exception that proves that rule.  

9 comments:

  1. If I read you correctly, you say that QE can be reversed in order to avoid inflation, but helicopter money can't.
    Am I wrong to believe that it is not entirely true? Inflation can be avoided if the government conducts some reverse fiscal policy once full employment is reached, by raising taxes and withdrawing the extra money.
    Of course, we know that even a German government would not do that.
    So, the real difference between QE and helicopter money is that QE keeps the policy in the hands of the central bank, and this is why central agree to QE, but won't even consider helicopter money.

    ReplyDelete
  2. I have two questions: if we're in a depressed economy at the zero-bound, and the fiscal authority is undertaking bond-financed deficit spending while the central bank is undertaking quantitive easing, then the nominal value of increased gov't spending will be reflected in an increase in excess reserves. But if this policy is effective in getting the economy back to full employment, why would be assume that the level of excess reserves would be unchanged (and hence that QE would need to be "reversed")? And second, even if the level of reserves is unchanged, what is the basis for believing that the level of reserves, rather than their rate of change, affects aggregate spending and inflation?

    ReplyDelete
  3. Note that in the case of Spain, for example, the government cannot count on its central bank to do some much needed QE. It can only beg the ECB to do so. It can't print euros either. But it can print IOUs...
    IOUs are usually weak, because the institutions that emit them are extremely weak. That's not exactly the case of Spain. It might pledge to withdraw the IOUs once the Spanish economy returns to full employment (well, a drop from 25% to 12% would already do), and retain its credibility in doing so.
    This proposition is a bit more developed here http://www.98economics.com/2012/05/old-fashioned-remedy-for-euro-disease.html). Jonathan Portes half-seriously developed a similar argument (http://notthetreasuryview.blogspot.ch/2012/07/easymoney-could-save-eurozone.html), but I agree it sounds weird, and there's definitely no maths backing the argument.

    ReplyDelete
    Replies
    1. Spain can't print euros? Maybe it could, by using TARGET2.
      Let's say a Spanish government bond held by Deutsche bank is maturing today.
      Step 1 - the Spanish government sells a new bond to a government-owned bank (it instructs its bank to open a new deposit).
      Step 2 - it tranfers the new deposit to pay off Deutsche Bank
      Step 3 - at the end of the day the Bundesbank will have a corresponding credit position versus the ESCB while the Spanish Central Bank will have a negative postion towards the ESCB. In TARGET2 there are no limits to such balances.
      Step 4 - repeat the process tomorrow, after tomorrow, on and on.
      In practice, Spain has started printing euros.

      Delete
  4. A. ... But suppose money financed fiscal expansion, or bond financed expansion plus QE, works, and we get back to what you call full employment. What happens to all that money the central bank has created?

    Q: Well now there would be too much money chasing too few goods, so the central bank would have to put QE into reverse. Otherwise we would get inflation.

    How do we know there would be too much money? Perhaps there would be just the right amount. After all, the new money is just replacing bubble money for the most part.

    ReplyDelete
  5. "A: I agree. But suppose money financed fiscal expansion, or bond financed expansion plus QE, works, and we get back to what you call full employment. What happens to all that money the central bank has created?"

    Q: How is that different from the times of metal backed currency, when there was a buffer of money that did not depend upon loans?

    Q2: Would a buffer of non-loan money act as a stabilizer?

    ReplyDelete
  6. "A: I agree. But suppose money financed fiscal expansion, or bond financed expansion plus QE, works, and we get back to what you call full employment. What happens to all that money the central bank has created?

    Q: Well now there would be too much money chasing too few goods, so the central bank would have to put QE into reverse. Otherwise we would get inflation."

    If the fiscal expansion 'works', it produces enough goods so that there won't be too much money chasing too few goods. If there is too much money chasing too few goods, it means the the fiscal expansion was too expansionary. In which case the inflation would result even if it was not money-financed.

    The reverse QE needed to mop up any excess liquidity will not be any different from the OMOs that central banks do day in and day out. It won't involve a full scale back, so you can't write it off by saying that if *any* reverse QE is needed, no initial QE is required. Socrates wouldn't approve of such binary worlds. Nor would Keynes or Marshall.

    Your model is precisely the kind of Old Keynesian comparative statics that was punched all around by people on the left and the right. You think money goes into prices, but government spending goes right into the real economy. How the two immaculately manage to separate out so well in a world where everyone spends nominally -you don't tell us. Your model has no nominal anchor. It is Solow-Samuelson at peak ad-hockery.

    It is possible, of course, to turn this around and say that my argument implies that the helicopter drop's success and failure are independent of how it is financed. So a drop of bonds is as likely to succeed or fail as a drop of money. Government is not Modigliani Miller constrained in creating net private wealth through helicopter drops, but the mode of financing is subject to MM type neutrality.

    Buiter and Bernanke think they have the answer of why money may be special (perpetual liability). You can agree or disagree. I am agnostic. But the assumption of your model that fiscal transfers can work on the real economy without a nominal anchor is a discredited one.

    ReplyDelete
  7. “Well now there would be too much money chasing too few goods, so the central bank would have to put QE into reverse.” That might be the case, but I don’t see any good reason to assume it. If the increased stock of money is needed to get back to full employment, that will presumably be because of the private sector’s desire for an increased stock of cash (e.g. because of the borrowing binge in recent years, burned fingers, etc), and the effect of the latter salutary lesson won’t suddenly disappear just because we have full employment. Thus I’d guess that desire for an increased stock of cash will stay for a year or two yet.

    “So if using money to finance extra government spending just involves a temporary increase in money, and no permanent reduction in debt, what is the point?” As implied above, the point is to meet the private sector’s desire for a larger stock of very liquid assets. And it’s impossible to say how long that desire will last.

    But given that monetary base and government debt are very similar in nature, I regard QE as much like the Bank of England offering everyone two £10 notes for £20 notes.

    And if I’m wrong, I’ll go and eat some Hemlock.

    ReplyDelete

Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.