A Review of Keynes's Tract on Monetary Reform: Hoisted from the Archives

I wrote this eleven years ago. I still like it:

John Maynard Keynes, A Tract on Monetary Reform (London: Macmillan, 1924)

This may well be Keynes's best book. It is certainly the best monetarist economics book ever written.

What do I mean by monetarist? Consider the book's preface, where Keynes writes:

[The economy] cannot work properly if the money... assume[d] as a stable measuring rod, is undependable. Unemployment, the precarious life of the worker, the disappointment of expectation, the sudden loss of savings, the excessive windfalls to individuals, the speculator, the profiteer--all proceed, in large measure, from the instability of the standard of value.

It is often supposed that the costs of production are threefold... labor, enterprise, and accumulation. But there is a fourth cost, namely, risk; and the reward of risk-bearing is one of the heaviest, and perhaps the most avoidable, burden on production....[T]he adoption by this country and the world at large of sound monetary principles, would diminish the wastes of Risk, which consume at present too much of our estate.

The belief that monetary instability--inflation and deflation--is the principal, or at least a principal, cause of other economic evils; the hope that sound monetary principles can be identified and, when identified, would greatly diminish uncertainty and risk; the focus on the job of the public sector being to provide the private economy with a stable measuring-rod and a stable environment--all these are core ideas of whatever we choose to call monetarism. Keynes believed these ideas very, very strongly in the mid-1920s. And his Tract on Monetary Reform is a review of economic theory and a look at the economic problems of post-WWI Europe through this set of monetarist spectacles.

The first chapter--"The Consequences to Society of Changes in the Value of Money"--may still be the best summary of the many and varied effects of deflation and inflation--on the distribution of income, on economic activity, on attitudes toward risk and reward--ever written. From our present-day standpoint, it could use a little more focus on the differing effects of "anticipated" and "unanticipated" inflation and deflation. But a great deal is packed into a short space.

The second chapter--"Public Finance and Changes in the Value of Money"--may also be the best of its class. It provides an extremely lucid introduction to the idea of the "inflation tax"--that inflation is most importantly seen as a way for governments to levy a hidden tax on holdings of real money balances, and that governments almost inevitably find themselves resorting to this tax, whether by accident or by design.

The third chapter--"The Theory of Money and the Foreign Exchanges" is in its firsts part a rapid introduction to the so-called "Quantity Theory of Money". It contains what must be Keynes's most famous line--in the long run we are all dead--which is embedded in the following discussion:

It would follow... that an arbitrary doubling of [the money stock], since this in itself is assumed not to affect [the velocity of money or the real volume of transactions] ... must have the effect of raising [the price level] to double what it would have been otherwise. The Quantity Theory is often stated in this, or a similar, form.

Now "in the long run" this is probably true. If, after the American Civil War, the American dollar had been stabilized... ten per cent below its present value ... [the money stock] and [the price level] would now be just ten per cent greater than they actually are.... But this long run is a misleading guide to current affairs. In the long run we are all dead. Economists set themselves too easy, too useless a task if in tempestuous seasons they can only tell us that when the strom is long past the ocean is flat again.

In actual experience, a change in [the money stock] is liable to have a reaction both on [the velocity of money] and on [the real volume of transactions]...

The second part of the chapter is a rapid introduction to exchange-rate determination--the purchasing-power-parity theory of exchange rate movements, and why there might be substantial and persistent deviations from what purchasing-power-parity would suggest.

Chapter four--"Alternative Aims in Monetary Policy"--sees Keynes shift from analyst to advocate: he comes down, in the context of Western Europe in the 1920s, on the side of devaluation to bring official currency values in line with relative national price levels rather than of deflation to force national price levels into consistency with pre-WWI exchange rate parities. He argues that when you are forced to choose between maintaining a stable exchange rate and maintaining a stable internal price level, choose the second. For avoiding fluctuations in your internal price level avoids a host of evils:

We see, therefore, that rising prices and falling prices each have their characteristic disadvantage.... Inflation is unjust and Deflation is inexpedient.... [I]t is not necessary that we should weigh one evil against the other. It is easier to agree that both are evisl to be shunned. The Individualistic Capitalism of today, precisely because it entrusts saving to the individual investor and production to the individual employer, presumes a stable measuring-rod of value, and cannot be efficient--perhaps cannot survive--without one...

He argues against return to the gold standard, on the grounds that modern central banks run by clever people like him can do a better job of maintaining price stability if they are not tied to gold. Keynes's arguments in chapter four look very good: current opinion among economic historians, exemplified by Barry Eichengreen's Golden Fetters: The Gold Standard and the Great Depression, is that attachment to gold did a large part of the work in preventing central banks from stemming the Great Depression of the 1930s.

The last chapter contains Keynes's "Positive Suggestions for the Future Regulation of Money". Keynes's suggested policies are the same as Irving Fisher, or indeed as Milton Friedman: spend money to construct a good price index, and then tune monetary policy so as to stabilize internal prices. As Keynes wrote in his preface:

We leave Saving to the private investor.... We leave the responsibility for setting Production in motion to the business man.... [T]hese arrangements, being in accord with human nature, have great advantages. But they cannot work properly if the [value of] money, which the assume as a stable measuring-rod, is undependable...

The implicit point of view is that if the value of money is dependable then leaving saving to the private investors and investment to business will work well. The magnitude of the Great Depression of the 1930s would destroy Keynes's faith in the proposition that stable internal prices implied a well-functioning macroeconomy and small business cycles. But from our perspective today--in which the Great Depression is seen as a unique disaster brought on by an unprecedented collapse in financial intermediation and in world trade, rather than as the largest species of the genus of business cycles--it is far from clear that Keynes of 1936 is to be preferred to Keynes of 1924.

Besides, Keynes of 1924 writes better: his prose is clearer, less academic, less formal; his argument is more straightforward, linear, easier to follow; his style is as witty.

Comments

[sa]

wow, i can't believe delong wrote the last para. he's actually moving from center-left to -gasp- the center.

[marcel]

"But from our perspective today--in which the Great Depression is seen as a unique disaster brought on by an unprecedented collapse in financial intermediation and in world trade, rather than as the largest species of the genus of business cycles--it is far from clear that Keynes of 1936 is to be preferred to Keynes of 1924."

It is important to distinguish your goals here, policy or theory. Accepting this argument about the GD, Keynes (1924) is more appropriate for policy. However, Keynes (1924) could not explain how the GD could persist (not occur, but persist). So something new was needed. Keynes (1936) may not be that new thing, but it was an attempt.

Consider the following analogy, with physics of course, this being economics. Newtonian mechanics is appropriate for policy, for understanding things we run into in everyday life. There are things that can happen, that have happened, that are not part of daily life and which Newtonian mechanics has no explanation. Here, Einsteinian mechanics, in both its incommensurable forms, relativity and quantum mechanics. is necessary,.

[Randolph Fritz]

1924 was the period of the German hyper-inflation, whereas 1936 was the period of the Great Depression. Different medicines, I think, for different diseases.

[andres]

Yes indeed, Brad's last paragraph is definitely deserving of a smackdown:

"The implicit point of view is that if the value of money is dependable then leaving saving to the private investors and investment to business will work well. The magnitude of the Great Depression of the 1930s would destroy Keynes's faith in the proposition that stable internal prices implied a well-functioning macroeconomy and small business cycles. But from our perspective today--in which the Great Depression is seen as a unique disaster brought on by an unprecedented collapse in financial intermediation and in world trade, rather than as the largest species of the genus of business cycles--it is far from clear that Keynes of 1936 is to be preferred to Keynes of 1924."

The Great Depression is _not_ a unique disaster. Just ask Argentinians, or Chileans who were around in the early 1980's for that matter. And now that I think about it, just ask any Fed veteran what might have happened if they had continued to target monetary aggregates (with resulting high interest rates) after 1982.

A Great Depression on either a national or a global scale is what happens when you have any monetary system that fanatically places price stability above the macro-needs of the economy. After all, exchange rate targeting in the form of gold standards or currency boards is simply the most direct and crudest way to target a 0% rate of inflation.

Practically every competent central bank today realizes that the road to both price and output stability is not through the control of exchange rates or monetary aggregates, but by the creation of stable future expectations guided by stable (and low) real interest rates, by secondary financial markets growing at a stable rate in real terms, and by real wages that have a slight downward trend that can only be corrected at the discretion of employers--and that's a more complicated and richer story than any form of monetarism can come up with.

There are other parts of the orthodox story of the Great Depression that are at least partially mistaken. For example the Smoot-Hawley tariff didn't take effect until after the Great Depression was underway and banks had already started to collapse, and the reduction in world trade is exaggerated, especially when one look at trade as a proportion of world GDP rather than as an absolute quantity, ie one should separate the reduction in world trade caused by collapsing aggregate demand from that caused by higher tariffs.

All in all, Keynes circa 1936 had his problems, but he definitely has a better story of the Great Depression than do most mainstream narratives today.

[mpowell]

"real wages that have a slight downward trend that can only be corrected at the discretion of employers"

Uh, excuse me?

[AS]

mpowell, I believe he is referring to the tendency for the real value of wages to erode as long as employers do not change their nominal values. This is guaranteed under a regime with inflation and provides some security against the most painful recession.

But I take issue with andres' argument. Monetary over-orthodoxy in the form of currency boards or the gold standard is good at preventing hyperinflations, but is not necessarily good at promoting stable inflation. These monetary systems all too often fall into deflation, which is destabilizing and perhaps the most salient feature of the worst recessions. And even present day advocates of inflation targeting set their target sufficiently above 0% so as to make deflation avoidable.

So it is hard to imagine a Great Depression happening under an inflation rule. This is not to say that an inflation rule is optimal, but it does mean that there is room to fight recessions while maintaining stable inflation.

[andres]

mpowell: to clarify, the sentence you point to is one of several reasons why some economists have argued that a positive but low rate of inflation is beneficial to the economy in that it gives firms the option of reducing real labor costs without taking the aggressive and demoralizing step of cutting nominal wages. Most of the time, of course, firms allow real wages to stay level by increasing nominal wages at approximately the rate of inflation, or even higher if labor productivity has increased.

[andres]

AS: Good point, but that's precisely the issue--modern central banks don't set 0% inflation targets because they know that the high interest rates/tight credit that are necessary to at least initially achieve 0% inflation also run a serious risk of causing the financial system to collapse. The only reason why gold standards/currency boards have a tendency to lead to deflation is because they often lead to financial distress in the wake of capital flight.

Policy errors like that happen only when central banks are suffering from PTHD (post-traumatic hyperinflation disorder). A good case can be made that Keynes himself, looking at central Europe's situation, was suffering from PTHD when he wrote the _Treatise on Monetary Reform_. More reality-grounded central banks recognize that as a rule they can put up with low single-digit inflation rates if the alternative is financial fragility and potential unemployment.

[JonF]

Re: The Great Depression is _not_ a unique disaster. Just ask Argentinians, or Chileans who were around in the early 1980's for that matter.

No doubt they experienced some bad times, but nothing like the Great Depression, probably more like the 19th century panics. To find anything that equals or exceeds the Depression you need to look at economic calamities that arose directly from other catastrophes, wars, plagues and famines. The Depression was unique both for its severity and it duration, and the fact that it was not directly caused by some larger disaster in society.

[roublen]

actually, I was wondering what m3 money supply growth had been these past, say 10-15 years, but googling m3 tends to get you only austrians, arguing m3 is super-important and ignored by the corrupt Keynesocracy .

I've been toying with the idea of Uncle Milton's revenge. big-M Monetarism was discredited in the 80's and 90's, but even though m3 growth is not the most important thing in the world, is it okay to completely ignore it, like we're doing now?

I have a feeling that m3 growth has has been quite high, ever since 1996. What does it mean when 1) money supply growth has been high for at least 10 years 2) but high m3 growth has not led to increased inflation?

Or have I just got the facts wrong?

[Scott Martens]

JonF, when it's you who loses his job and house and ends up living in a car for several years, it is very little comfort to know that what you're experiencing is merely like the 19th century panics rather than the Great Depression.

If I am to understand Professor Delong, the Great Depression is to be seen primarily as a political failure - that had a better functioning system of financial intermediation been in place, it might all have been avoided. Or perhaps I am misunderstanding the meaning of the words "financial intermediation". However, if this is the case, then ignoring the lessons of the Great Depression because it's an exceptional case is the height of folly. I see no reason to believe that the political mechanisms associated with the dawn of the Great Depression have meaningfully improved, and even if they had, the Bush Administration makes it clear that even if we could rely on a consistent measure of economic value to provide accurate signals about risk, we have no comparable yardstick to measure political risks.

I should think the lessons from the Depression that are being rejected are very important ones because they apply to the all too easy to envision case of failure in the political machinery of trade and credit managment.

[Matt]

Our central bank loses, monetarily, some 3% of its transactions, integrated over the business cycle. Effectively a negative 3% fee interst subsidized by us to member banks.

If banks see increased risk, they effectively lobby for higher inflation taxes. This increases the monetary base, raises long term interest rates; and forces shorter term financing and higher velocity rates.

The gold standard does not work for it effectively blocks banks from increasing or decreasing the inflation fee in periods of increased or decreased risk.

A central bank can only account for international currency rates if the trading partners have the same risk management scheme, as the Europeans have done, and as we have done in federalizing money over the states.

So, when a large economy, like China, with secret, command banking holds huge dollar reserves, the American banking risk increases and bankers raise the inflation tax.

Foreign holders of dollars are, in turn, motivated to manage the risk of foreign dollar reserves for fear of a higher inflation tax by American bankers.

It all balances out, I think! The fact that we are having this discussion implies that large foreign dollars will move cautiously.