I was taught to dismiss the Pigou effect. Arthur Cecil Pigou (or “Pig” if you believe spell-check) was a great Cambridge economic theorist, known to most of us as the object of Keynes’s repeated ridicule in the General Theory.
The effect that bears his name evolved in response to Keynes, and in particular the idea of a liquidity trap. Keynes argued that not only do depressions occur in the real world of sticky prices and wages, but even in a world of fully flexible prices there could be an equilibrium of high unemployment, something sadly forgotten by much of modern macroeconomics (Roger Farmer is a notable exception). Pigou challenged this claim in the abstract – and with the preface that his argument had little practical relevance – arguing that if the stock of money is fixed in nominal terms, deflation could generate positive real demand growth through a real wealth effect.
If students of economics hear anything about the Pigou effect these days, it is likely to be written-off as “irrelevant”. Money is a small share of total wealth; and as Kalecki counter-argued, deflation also causes the real stock of debt to rise.
Now whenever a monetary aggregate is involved, it matters how you define “money” – and theorists are rarely clear on this. If we are talking about deposits (a significant component of household wealth), their nominal value is not at all stable in a deflation – if defaults are widespread and banks fail, deposits (which are loans to banks) can also be defaulted upon. For these reasons among others, Pigou typically gets short shrift.
The last time I thought about the Pigou effect was in analysing QE in Japan in the early 2000s. It struck me that if we limit the discussion to the monetary base – the real value of which definitively does rise during deflation – its effect on demand depends on how much is created and who holds it. My reasoning was as follows: If banks hold reserves at the central bank, the real value of which goes up during deflation, this will hardly impact consumer spending. Consumers might hold some equity in banks, but in deflation all the other assets of the banking sector will be impaired, so overall bank equity values will decline – as they did in Japan. There is no beneficial wealth effect for households.
But what if households held all the base money, and furthermore the amount was increased dramatically? Surely then a Pigou effect would be large. This line of reasoning in fact pointed me towards cash transfers (as I argued in the FT in 2002).
I was reminded of this line of reasoning when recently re-reading Milton Friedman’s 1968 AER presidential address. It is famous for its critique of the Philips curve. Reading it today, it is far more interesting for its perspective on Keynes – presenting a much more interesting perspective than that of today’s “New Keynesians”.
How contemporary does this sound:
“My topic for tonight is the role of … monetary policy. What can it contribute? And how should it be conducted to contribute the most? Opinion on these questions has fluctuated widely. In the first flush of enthusiasm about the newly created Federal Reserve System, many observers attributed the relative stability of the 1920s to the System’s capacity for fine tuning – to apply an apt modern term. It came to be widely believed that a new era had arrived in which business cycles had been rendered obsolete by advances in monetary technology [sounds familiar]. This opinion was shared by economist and layman alike, though, of course, there were some dissonant voices. The Great Contraction destroyed this naive attitude. Opinion swung to the other extreme. Monetary policy was a string. You could pull on it to stop inflation but you could not push on it to halt recession. You could lead a horse to water but you could not make him drink. Such theory by aphorism was soon replaced by Keynes’ rigorous and sophisticated analysis [Yes, Milton Friedman thought the General Theory was “rigorous and sophisticated”].
Keynes offered simultaneously an explanation for the presumed impotence of monetary policy to stem the depression, a non-monetary interpretation of the depression, and an alternative to monetary policy for meeting the depression and his offering was avidly accepted. If liquidity preference is absolute or nearly so – as Keynes believed likely in times of heavy unemployment – interest rates cannot be lowered by monetary measures. If investment and consumption are little affected by interest rates – as Hansen and many of Keynes’ other American disciples came to believe – lower interest rates, even if they could be achieved, would do little good. Monetary policy is twice damned. The contraction, set in train, on this view, by a collapse of investment or by a shortage of investment opportunities or by stubborn thriftiness, could not, it was argued, have been stopped by monetary measures. But there was available an alternative – fiscal policy. Government spending could make up for insufficient private investment. Tax reductions could undermine stubborn thriftiness.”
Right there we have recent history succinctly described – forty years before it happened, and with more flair than Larry Summers.
But it is a later, almost throw-away remark of Friedman’s that really caught my attention:
“This revival [of belief in the potency of monetary policy] was strongly fostered among economists by the theoretical developments initiated by Haberler but named for Pigou that pointed out a channel – namely, changes in wealth – whereby changes in the real quantity of money can affect aggregate demand even if they do not alter interest rates. These theoretical developments did not undermine Keynes’ argument against the potency of orthodox monetary measures when liquidity preference is absolute since under such circumstances the usual monetary operations involve simply substituting money for other assets without changing total wealth [i.e. QE]. But they did show how changes in the quantity of money produced in other ways could affect total spending even under such circumstances. [italics added]”
This statement is fascinating for host of reasons. Firstly, Haberler, not Pigou, is the interesting person to read on the “Pigou effect”. Secondly, Friedman has correctly dismissed conventional QE – “the usual monetary operations involve simply substituting money for other assets without changing total wealth”. Thirdly, Friedman was no fool, and he believed that not only was the Pigou effect as described by Haberler significant, he thought that it shows how “changes in the quantity of money produced in other ways [ways other than open market operations or QE] could affect total spending even under such circumstances [a liquidity trap]”. Now what does “produced in other ways” mean?
I’m embarrassed to admit that I think I have only once read anything by Haberler, in a collection of articles on Austrian economics, and I can’t actually remember what it was about. So I decided to follow up Friedman’s lead. Haberler does indeed have a fascinating take on the Pigou effect, which are similar to my thoughts on Japan. In his brilliant article, written in 1952, Haberler is crystal clear on the fact that wages and price in the real world are sticky, but that this is not the crux of the argument. More relevantly, and in the section Friedman was clearly referring to, he says:
“The importance of the wealth-saving relation goes beyond the case usually designated by the Pigou effect, viz., beyond the effect of an increase in the real value of cash balances and government bonds due to falling prices. Suppose the quantity of money is increased by tax reduction or government transfer payments, government expenditures remaining unchanged and the resulting deficit being financed by borrowing from the central bank or simply printing money [he adds a footnote, which Friedman lifted without direct attribution: ‘Open market operations are different, because they result merely in a substitution of one type of asset for another.’]”
He goes on to say, “… consumption and investment expenditure will increase when the quantity of money grows. I find it difficult to believe that this might not be so.”
So do I. And I think its pretty clear that so did Milton Friedman – he says as much in his reference to Haberler’s argument in his 1968 presidential address. What Haberler is describing is either tax cuts or cash transfers financed by base money – and Mark Blyth and I thought we were on to something new! The only unaddressed issue is an institutional one: does it require coordination with fiscal authorities, or – as Mark and I recommend – let the central bank do it directly.
I concluded in a previous post that Keynes had already answered all the major problems in macro policy-making, but maybe today’s challenges require new thinking. I’m not so sure, now. I am increasingly convinced that the academic discussion of monetary and fiscal policy from 1930 to 1970 was far more useful, interesting and subtle than most of what I read currently. That’s not in-of-itself a bad thing – it’s publicly available and by-and-large better written too.
What we need to do faced with a shortfall in global demand is in fact blindingly obvious – global tax cuts, or better still cash transfers, financed by central banks. Friedman thought so, and so it appears did Gottfried Haberler – an Austrian economist. Could someone please whisper this in Ben Bernanke’s ear … he would have something much more interesting to blog about.
A banal confusion has marred discussion of the Pigou effect. No one has ever believed that a combination of deflation and an unchanged stock of base money would cause a recovery of demand in a high unemployment deflation. That straw man has unfortunately been the object of many “refutations”. I should be clear, however, that as I am using the term, the “Pigou effect” does not assume that the stock of money is constant, and all the real wealth increase is due to deflation. The Pigou effect, as I am using the term, is simply the hypothesis that the real value of money rises under deflation, and an increase of real money balances under deflation – if sufficiently large – could cause higher demand growth.
 Having written this blog, I subsequently discovered, via Simon Wren-Lewis, a post by Paul Krugman, which gets close. Krugman points out that he has careful re-read his Samuelson and Solow 1960. He was less careful re-reading Friedman 1968. Krugman assumes that Friedman is talking about open-market operations – when he expressly is not, as I have detailed above. Krugman also assumes the only alternative is fiscal policy. He talks vaguely about helicopter drops in this context, saying “it depends very much who gets the windfall and who pays the taxes, and we’re basically talking about fiscal rather than monetary policy.” Basically? What’s fiscal and what’s monetary is never well-defined by macro theorists. I have attempted here. Whatever one calls it, what Friedman and Haberler are clearly arguing is that cash transfers to households financed by base money would stimulate demand under deflation and the transmission mechanism is rising real balances. Base money is not a liability, so Ricardian equivalence – if you care about it – is not relevant. And neither if one thinks about it, is the issue of “permanence”.
I fully agree with you about the discussion of 1930-1970. We have a lot to learn from Fisher, Simons and Keynes, and though I often disagree with Friedman, he expresses himself in a clear way and has taught me a lot.
I might be missing something, but to me it sounds that what you and, for instance, Richard Werner seem to be suggesting doesn’t really take into account investor/human psychology? Why shouldn’t we expect more volatile inflation, very much like in the 1970s, as a consequence of these increased deficits? I don’t see anything new in the idea that government can spend without creating a corresponding liability. What would be new, should this path be chosen, is if the price of gold would not shoot up and people wouldn’t rush to buy real assets with leverage.
Am I missing something?
I’ve written about a related issue here: http://clumsystatements.blogspot.com/2015/03/inflation-cannot-be-micromanaged-or.html
Despite my skepticism, I really appreciate your writing and the way you open-mindedly consider the options we have. Thanks!
Thanks Peter. I think we have to work out what causes inflation. The microeconomic structure matters. Today’s world is radically different to the 1960s and 1970s because it is highly de-regualted and fiercely competitive. No one has pricing power. Combine that with a problem generating demand globally and you get no inflation.
Read: Ambrose Evans-Pritchard, Why Paul Krugman is Wrong, 15 December 2014
When the channel from second-hand trade in finance at Wall Street to first-hand trade in operational production at Main Street is jammed, pumping into the former will not have much effect on the latter. Injecting directly into the latter will ignite the multiplier-process. Taking politics and central bank as two legs or arms of administration, we see Evans-Pritchard arguing that the other limb can take over if the one limb does not do its job, like Keynes had judged. There is a constitutional thing, but the economics is trivial.
Hicks, The Crisis in Keynesian Economics, 1974 shows recovery taking time to gather momentum because excess-inventories are dismantled first. After a dragging slump the economy does not get heated that quickly. Hicks also considered micro: a flexprice sector – dominated by speculative arbitrage – and a fixprice sector. Crudely, relative to fixprice “the” flexprice is “interest-bound”. Crudely, relative to flexprice, the fixprice is wage-bound. It takes wide tolerance bounds. Ricardo was a tolerant man. Hicks had sound ideas on wages in 20th century Britain. You best can assess it yourselves. After a dragging slump wage demands may take a while, and by the time …
Thanks Joop – I’ll read
Thanks for the reply! I agree, we need to work out what causes inflation. What you suggest is, to me, the cause for us having not seen much inflation despite our enormous credit growth since the 1980s.
When credit grows, so do liabilities (debt). Now, if government would spend deposits into existence without adding to taxpayer liabilities (= would run a larger deficit), and even make it explicit that this deficit is not going to be covered later, this would necessarily affect people’s trust in the value of the currency. If not, people would be irrational 🙂 And what I’m trying to say is that should we try this, the consequences would not be neatly linear — trust is not linear — but unpredictable and hard to manage. Inflation is very much a psychological phenomenon, having much (not all) to do with people’s trust in the currency. Supply constraints are just one part of the inflation story, another part being people’s willingness to hold their savings in the form of deposits (vs. real assets).
You might agree if I say that we don’t understand inflation well enough? I think one problem is that many people still think of inflation in terms of the Quantity Theory of Money, which paints a very mechanical picture of the economy. It ignores the often decisive role expectations, and psychology in general, play in the economy.
Re Bernanke, see his para starting “A possible arrangement…” here:
Basically he advocates helicoptering which (strikes me) works in part because of the Pigou effect: i.e. everyone’s stock of money rises.