Archive for the 'Uncategorized' Category

Keynes and Accounting Identities

In a post earlier this week, Michael Pettis was kind enough to refer to a passage from Ralph Hawtrey’s review of Keynes’s General Theory, which I had quoted in an earlier post, criticizing Keynes’s reliance on accounting identities to refute the neoclassical proposition that it is the rate of interest which equilibrates savings and investment. Here’s what Pettis wrote:

Keynes, who besides being one of the most intelligent people of the 20th century was also so ferociously logical (and these two qualities do not necessarily overlap) that he was almost certainly incapable of making a logical mistake or of forgetting accounting identities. Not everyone appreciated his logic. For example his also-brilliant contemporary (but perhaps less than absolutely logical), Ralph Hawtrey, was “sharply critical of Keynes’s tendency to argue from definitions rather than from causal relationships”, according to FTC economist David Glasner, whose gem of a blog, Uneasy Money, is dedicated to reviving interest in the work of Ralph Hawtrey. In a recent entry Glasner quotes Hawtrey:

[A]n essential step in [Keynes’s] train of reasoning is the proposition that investment and saving are necessarily equal. That proposition Mr. Keynes never really establishes; he evades the necessity doing so by defining investment and saving as different names for the same thing. He so defines income to be the same thing as output, and therefore, if investment is the excess of output over consumption, and saving is the excess of income over consumption, the two are identical. Identity so established cannot prove anything. The idea that a tendency for investment and saving to become different has to be counteracted by an expansion or contraction of the total of incomes is an absurdity; such a tendency cannot strain the economic system, it can only strain Mr. Keynes’s vocabulary.

This is a very typical criticism of certain kinds of logical thinking in economics, and of course it misses the point because Keynes is not arguing from definition. It is certainly true that “identity so established cannot prove anything”, if by that we mean creating or supporting a hypothesis, but Keynes does not use identities to prove any creation. He uses them for at least two reasons. First, because accounting identities cannot be violated, any model or hypothesis whose logical corollaries or conclusions implicitly violate an accounting identity is automatically wrong, and the model can be safely ignored. Second, and much more usefully, even when accounting identities have not been explicitly violated, by identifying the relevant identities we can make explicit the sometimes very fuzzy assumptions that are implicit to the model an analyst is using, and focus the discussion, appropriately, on these assumptions.

I agree with Pettis that Keynes had an extraordinary mind, but even great minds are capable of making mistakes, and I don’t think Keynes was an exception. And on the specific topic of Keynes’s use of the accounting identity that expenditure must equal income and savings must equal investment, I think that the context of Keynes’s discussion of that identity makes it clear that Keynes was not simply invoking the identity to prevent some logical slipup, as Pettis suggests, but was using it to deny the neoclassical Fisherian theory of interest which says that the rate of interest represents the intertemporal rate of substitution between present and future goods in consumption and the rate of transformation between present and future goods in production. Or, in less rigorous terminology, the rate of interest reflects the marginal rate of time preference and the marginal rate of productivity of capital. In its place, Keynes wanted to substitute a pure monetary or liquidity-preference theory of the rate of interest.

Keynes tried to show that the neoclassical theory could not possibly be right, inasmuch as, according to the theory, the equilibrium rate of interest is the rate that equilibrates the supply of with the demand for loanable funds. Keynes argued that because investment and savings are identically equal, savings and investment could not determine the rate of interest. But Keynes then turned right around and said that actually the equality of savings and investment determines the level of income. Well, if savings and investment are identically equal, so that the rate of interest can’t be determined by equilibrating the market for loanable funds, it is equally impossible for savings and investment to determine the level of income.

Keynes was unable to distinguish the necessary accounting identity of savings and investment from the contingent equality of savings and investment as an equilibrium condition. For savings and investment to determine the level of income, there must be some alternative definition of savings and investment that allows them to be unequal except at equilibrium. But if there are alternative definitions of savings and investment that allow those magnitudes to be unequal out of equilibrium — and there must be such alternative definitions if the equality of savings and investment determines the level of income — there is no reason why the equality of savings and investment could not be an equilibrium condition for the rate of interest. So Keynes’s attempt to refute the neoclassical theory of interest failed. That was Hawtrey’s criticism of Keynes’s use of the savings-investment accounting identity.

Pettis goes on to cite Keynes’s criticism of the Versailles Treaty in The Economic Consequences of the Peace as another example of Keynes’s adroit use of accounting identities to expose fallacious thinking.

A case in point is The Economic Consequences of the Peace, the heart of whose argument rests on one of those accounting identities that are both obvious and easily ignored. When Keynes wrote the book, several members of the Entente – dominated by England, France, and the United States – were determined to force Germany to make reparations payments that were extraordinarily high relative to the economy’s productive capacity. They also demanded, especially France, conditions that would protect them from Germany’s export prowess (including the expropriation of coal mines, trains, rails, and capital equipment) while they rebuilt their shattered manufacturing capacity and infrastructure.

The argument Keynes made in objecting to these policies demands was based on a very simple accounting identity, namely that the balance of payments for any country must balance, i.e. it must always add to zero. The various demands made by France, Belgium, England and the other countries that had been ravaged by war were mutually contradictory when expressed in balance of payments terms, and if this wasn’t obvious to the former belligerents, it should be once they were reminded of the identity that required outflows to be perfectly matched by inflows.

In principle, I have no problem with such a use of accounting identities. There’s nothing wrong with pointing out the logical inconsistency between wanting Germany to pay reparations and being unwilling to accept payment in anything but gold. Using an accounting identity in this way is akin to using the law of conservation of energy to point out that perpetual motion is impossible. However, essentially the same argument could be made using an equilibrium condition for the balance of payments instead of an identity. The difference is that the accounting identity tells you nothing about how the system evolves over time. For that you need a behavioral theory that explains how the system adjusts when the equilibrium conditions are not satisfied. Accounting identities and conservation laws don’t give you any information about how the system adjusts when it is out of equilibrium. So as Pettis goes on to elaborate on Keynes’s analysis of the reparations issue, one or more behavioral theories must be tacitly called upon to explain how the international system would adjust to a balance-of-payments disequilibrium.

If Germany had to make substantial reparation payments, Keynes explained, Germany’s capital account would tend towards a massive deficit. The accounting identity made clear that there were only three possible ways that together could resolve the capital account imbalance. First, Germany could draw down against its gold supply, liquidate its foreign assets, and sell domestic assets to foreigners, including art, real estate, and factories. The problem here was that Germany simply did not have anywhere near enough gold or transferable assets left after it had paid for the war, and it was hard to imagine any sustainable way of liquidating real estate. This option was always a non-starter.

Second, Germany could run massive current account surpluses to match the reparations payments. The obvious problem here, of course, was that this was unacceptable to the belligerents, especially France, because it meant that German manufacturing would displace their own, both at home and among their export clients. Finally, Germany could borrow every year an amount equal to its annual capital and current account deficits. For a few years during the heyday of the 1920s bubble, Germany was able to do just this, borrowing more than half of its reparation payments from the US markets, but much of this borrowing occurred because the great hyperinflation of the early 1920s had wiped out the country’s debt burden. But as German debt grew once again after the hyperinflation, so did the reluctance to continue to fund reparations payments. It should have been obvious anyway that American banks would never accept funding the full amount of the reparations bill.

What the Entente wanted, in other words, required an unrealistic resolution of the need to balance inflows and outflows. Keynes resorted to accounting identities not to generate a model of reparations, but rather to show that the existing model implicit in the negotiations was contradictory. The identity should have made it clear that because of assumptions about what Germany could and couldn’t do, the global economy in the 1920s was being built around a set of imbalances whose smooth resolution required a set of circumstances that were either logically inconsistent or unsustainable. For that reason they would necessarily be resolved in a very disruptive way, one that required out of arithmetical necessity a substantial number of sovereign defaults. Of course this is what happened.

Actually, if it had not been for the insane Bank of France and the misguided attempt by the Fed to burst the supposed stock-market bubble, the international system could have continued for a long time, perhaps indefinitely, with US banks lending enough to Germany to prevent default until rapid economic growth in the US and western Europe enabled the Germans to service their debt and persuaded the French to allow the Germans to do so via an export surplus. Instead, the insane Bank of France, with the unwitting cooperation of the clueless (following Benjamin Strong’s untimely demise) Federal Reserve precipitated a worldwide deflation that triggered that debt-deflationary downward spiral that we call the Great Depression.

The Pot Calls the Kettle Black

I had not planned to post anything today, but after coming across an article (“What the Fed Really Wants Is to Reduce Real Wages”) by Alex Pollock of AEI on Real Clear Markets this morning, I decided that I could not pass up this opportunity to expose a) a basic, but common and well-entrenched, error in macroeconomic reasoning, and b) the disturbingly hypocritical and deceptive argument in the service of which the faulty reasoning was deployed.

I start by quoting from Pollock’s article.

To achieve economic growth over time, prices have to change in order to adjust resource allocation to changing circumstances. This includes the price of work, or wages. Everybody does or should know this, and the Federal Reserve definitely knows it.

The classic argument for why central banks should create inflation as needed is that this causes real wages to fall, thus allowing the necessary downward adjustment, even while nominal wages don’t fall. Specifically, the argument goes like this: For employment and growth, wages sometimes have to adjust downward; people and politicians don’t like to have nominal wages fall– they are “sticky.” People are subject to Money Illusion and they don’t think in inflation-adjusted terms. Therefore create inflation to make real wages fall.

In an instructive meeting of the Federal Reserve Open Market Committee in July, 1996, the transcript of which has been released, the Fed took up the issue of “long-term inflation goals.” Promoting the cause of what ultimately became the Fed’s goal of 2% inflation forever, then-Fed Governor Janet Yellen made exactly the classic argument. “To my mind,” she said, “the most important argument for some low inflation rate is…that a little inflation lowers unemployment by facilitating adjustments in relative pay”-in other words, by lowering real wages. This reflects “a world where individuals deeply dislike nominal pay cuts,” she continued. “I think we are dealing here with a very deep-rooted property of the human psyche”-that is, Money Illusion.

In sum, since “workers resist and firms are unwilling to impose nominal pay cuts,” the Fed has to be able to reduce real wages instead by inflation.

But somehow the Fed never mentions that this is what it does. It apparently considers it a secret too deep for voters and members of Congress to understand. Perhaps it would be bad PR?

This summary of why some low rate of inflation may promote labor-0market flexibility is not far from the truth, but it does require some disambiguation. The first distinction to make is that while Janet Yellen was talking about adjustments in relative pay, presumably adjustments in wages both across different occupations and also across different geographic areas — a necessity even if the overall level of real wages is stable — Pollock simply talks about reducing real wages in general.

But there is a second, more subtle distinction to make here as well, and that distinction makes a big difference in how we understand what the Fed is trying to do. Suppose a reduction in real wages in general, or in the relative wages of some workers is necessary for labor-market equilibrium. To suggest that only reason to use inflation to reduce the need for nominal wage cuts is a belief in “Money Illusion” is deeply misleading. The concept of “Money Illusion” is only meaningful when applied to equilibrium states of the economy. Thus the absence of “Money Illusion” means that the equilibrium of the economy (under the assumption that the economy has a unique equilibrium — itself, a very questionable assumption, but let’s not get diverted from this discussion to an even messier and more complicated one) is the same regardless of how nominal prices are scaled up or down. It is entirely possible to accept that proposition (which seems to follow from fairly basic rationality assumptions) without also accepting that it is irrelevant whether real-wage reductions in response to changing circumstances are brought about by inflation or by nominal-wage cuts.

Since any discussion of changes in relative wages presumes that a transition from one equilibrium state to another equilibrium state is occurring, the absence of Money Illusion, being a property of equilibrium,  can’t tell us anything about whether the transition from one equilibrium state to another is more easily accomplished by way of nominal-wage cuts or by way of inflation. If, as a wide range of historical evidence suggests, real-wage reductions are more easily effected by way of inflation than by way of nominal-wage cuts, it is plausible to assume that minimizing nominal-wage cuts will ease the transition from the previous equilibrium to the new one.

Why is that? Here’s one way to think about it. The resistance to nominal-wage cuts implies that more workers will be unemployed initially if nominal wages are cut than if there is an inflationary strategy. It’s true that the unemployment is transitory (in some sense), but the transitory unemployment will be with reduced demand for other products, so the effect of unemployment of some workers is felt by other sectors adn other workers. This implication is not simply the multiplier effect of Keynesian economics, it is also a direct implication of the widely misunderstood Say’s Law, which says that supply creates its own demand. So if workers are more likely to become unemployed in the transition to an equilibrium with reduced real wages if the real-wage reduction is accomplished via a cut in nominal wages than if accomplished by inflation, then inflation reduces the reduction in demand associated with resistance to nominal-wage cuts. The point is simply that we have to consider not just the final destination, but also the route by which we get there. Sometimes the route to a destination may be so difficult and so dangerous, that we are better off not taking it and looking for an alternate route. Nominal wage cuts are very often a bad route by which to get to a new equilibrium.

That takes care of the error in macroeconomic reasoning, but let’s follow Pollock a bit further to get to the hypocrisy and deception.

This classic argument for inflation is of course a very old one. As Ludwig von Mises discussed clearly in 1949, the first reason for “the engineering of inflation” is: “To preserve the height of nominal wage rates…while real wage rates should rather sink.” But, he added pointedly, “neither the governments nor the literary champions of their policy were frank enough to admit openly that one of the main purposes of devaluation was a reduction in the height of real wage rates.” The current Fed is not frank enough to admit this fact either. Indeed, said von Mises, “they were anxious not to mention” this. So is the current Fed.

Nonetheless, the Fed feels it can pontificate on “inequality” and how real middle class incomes are not rising. Sure enough, with nominal wages going up 2% a year, if the Fed achieves its wish for 2% inflation, then indeed real wages will be flat. But Federal Reserve discussions of why they are flat at the very least can be described as disingenuous.

Actually, it is Pollock who is being disingenuous here. The Fed does not have a policy on real wages. Real wages are determined for the most part in free and competitive labor markets. In free and competitive labor markets, the equilibrium real wage is determined independently of the rate of inflation. Remember, there’s no Money Illusion. Minimizing nominal-wage cuts is not a policy aimed at altering equilibrium real wages, which are whatever market forces dictate, but of minimizing the costs associated with the adjustments in real wages in response to changing economic conditions.

I know that it’s always fun to quote Ludwig von Mises on inflation, but if you are going to quote Mises about how inflation is just a scheme designed to reduce real wages, you ought to at least be frank enough to acknowledge that what Mises was advocating was cutting nominal wages instead.

And it is worth recalling that even Mises recognized that nominal wages could not be reduced without limit to achieve equilibrium. In fact, Mises agreed with Keynes that it was a mistake for England in 1925 to restore sterling convertibility into gold at the prewar parity, because doing so required further painful deflation and nominal wage cuts. In other words, even Mises could understand that the path toward equilibrium mattered. Did that mean that Mises was guilty of believing in Money Illusion? Obviously not. And if the rate of deflation can matter to employment in the transition from one equilibrium to another, as Mises obviously conceded, why is it inconceivable that the rate of inflation might also matter?

So Pollock is trying to have his cake and eat it. He condemns the Fed for using inflation as a tool by which to reduce real wages. Actually, that is not what the Fed is doing, but, let us suppose that that’s what the Fed is doing, what alternative does Pollock have in mind? He won’t say. In other words, he’s the pot.

Optimistic Thoughts about Productivity Growth and Secular Stagnation

The world, for all kinds of reasons, seems to be in a rather deplorable state, and the trendlines are not necessarily pointed in the right direction either. I won’t go through the whole, or even a partial, list of what is depressing me these days, but there is one topic that comes up a lot in the economics blogosphere, secular stagnation, about which I can conjure up some reasons for optimism.

What a lot of people are depressed about is the unusually low rate of growth in labor productivity that we have seen in the current recovery from the Little Depression. Although the rate of job creation has picked up over the past 15 months or so, averaging over 200,000 new jobs a month, the minimal increases in labor productivity mean that output growth has been much slower than usually observed in previous recoveries from steep downturns. The slow increase in labor productivity and the corresponding weakness of the recovery have been widely cited as evidence that we have entered into an era of secular stagnation.

I don’t deny that secular stagnation is a reasonable inference to be drawn from the persistently low increases in labor productivity during this recovery, but it does seem to me that a less depressing, though perhaps partial, explanation for low productivity growth may be available. My suggestion is that the 2008-09 downturn was associated with major sectoral shifts that caused an unusually large reallocation of labor from industries like construction and finance to other industries so that an unusually large number of workers have had to find new jobs doing work different from what they were doing previously. In many recessions, laid-off workers are either re-employed at their old jobs or find new jobs doing basically the same work that they had been doing at their old jobs. When workers transfer from one job to another similar job, there is little reason to expect a decline in their productivity after they are re-employed, but when workers are re-employed doing something very different from what they did before, a significant drop in their productivity in their new jobs is likely, though there may instances when, as workers gain new skills and experience in their new jobs, their productivity will rise rapidly.

In addition, the number of long-term unemployed (27 weeks or more) since the 2000-09 downturn has been unusually high. Workers who remain unemployed for an extended period of time tend to suffer an erosion of skills, causing their productivity to drop when they are re-employed even if they are able to find a new job in their old occupation. It seems likely that the percentage of long-term unemployed workers that switch occupations is larger than the percentage of short-term unemployed workers that switch occupations, so the unusually high rate of long-term unemployment has probably had a doubly negative effect on labor productivity.

I would even hazard a guess that the low rate of productivity growth in the 1970s may have also, at least in part, been skills-related, though for somewhat different reasons. The 1970s were a period of rapid growth in the labor force because of an influx of women and baby boomers, which continued throughout the 1980s. The influx of first-time workers into the labor force may have had something to do with slowdown in productivity growth in the 1970s and 1980s compared to the 1950s and 1960s, though the increase in the absolute size of the labor force was also a factor in holding down the rate of productivity growth. When the labor-force participation rate peaked in the 1990s, the rate of productivity growth increased.

I am not a labor economist, but a quick internet search seems to indicate that a strong empirical connection between labor-force composition and productivity growth has not been found. But another factor, the increase in energy prices, also had a skills aspect, the adjustment to increased energy prices having led to the adoption of new technologies and new methods of production that presumably required new skills different from those that workers had previously acquired. And the new technologies adopted in response to a sudden increase in energy prices had to go through a period of implementation and debugging and improvement during which measured productivity likely declined. Technological transformations require many workers to learn new skills and adapt to new technologies. The time spent learning how to do new things in new ways and figuring out how to do them better and more efficiently may have caused the measured productivity of such workers to drop, thereby slowing down the overall rate of productivity growth.

At any rate, if the anomalous decline in productivity in the expansion following the 2008-09 downturn has been caused, even if only in part, by a loss of skills by workers who were induced to change jobs or occupations or who experienced long periods of unemployment, that decline in unlikely to be permanent, so that the rate of productivity growth and the rate of growth in real GDP can be expected to pick up somewhat over time. Stagnation may therefore not be as long-lasting as some people fear.

Maybe that’s not such a great reason for optimism, but for now at least, it’s the best that I can muster.

Hey, Look at Me; I Turned Brad Delong into an Apologist for Milton Friedman

It’s always nice to be noticed, so I can hardly complain if Brad Delong wants to defend Milton Friedman on his blog against my criticism of his paper “Real and Pseudo Gold Standards.” I just find it a little bit rich to see Friedman being defended against my criticism by the arch-Keynesian Brad Delong.

But in the spirit of friendly disagreement in which Brad criticizes my criticism, I shall return the compliment and offer some criticisms of my own of Brad’s valiant effort to defend the indefensible.

So let me try to parse what Brad is saying and see if Brad can help me find sense where before I could find none.

I think that Friedman’s paper has somewhat more coherence than David does. From Milton Friedman’s standpoint (and from John Maynard Keynes’s) you need microeconomic [I think Brad meant to say macroeconomic] stability in order for private laissez-faire to be for the best in the best of possible worlds. Macroeconomic stability is:

  1. stable and predictable paths for total spending, the price level, and interest rates; hence
  2. a stable and predictable path for the velocity of money; hence
  3. (1) then achieved by a stable and predictable path for the money stock; and
  4. if (3) is secured by institutions, then expectations of (3) will generate the possibility of (1) and (2) so that if (3) is actually carried out then eppur si muove

I agree with Brad that macroeconomic stability can be described as a persistent circumstance in which the paths for total spending, the price level, and (perhaps) interest rates are stable and predictable. I also agree that a stable and predictable path for the velocity of money is conducive to macroeconomic stability. But note the difference between saying that the time paths for total spending, the price level and (perhaps) interest rates are stable and predictable and that the time path for the velocity of money is stable and predictable. It is, at least possibly the case, that it is within the power of an enlightened monetary authority to provide, or that it would be possible to construct a monetary regime that could provide, stable and predictable paths for total spending and the price level. Whether it is also possible for a monetary authority or a monetary regime to provide a stable and predictable path for interest rates would depend on the inherent variability in the real rate of interest. It may be that variations in the real rate are triggered by avoidable variations in nominal rates, so that if nominal rates are stabilized, real rates will be stabilized, too. But it may be that real rates are inherently variable and unpredictable. But it is at least plausible to argue that the appropriate monetary policy or monetary regime would result in a stable and predictable path of real and nominal interest rates. However, I find it highly implausible to think that it is within the power of any monetary authority or monetary regime to provide a stable and predictable path for the velocity of money. On the contrary, it seems much more likely that in order to provide stability and predictability in the paths for total spending, the price level, and interest rates, the monetary authority or the monetary regime would have to tolerate substantial variations in the velocity of money associated with changes in the public’s demand to hold money. So the notion that a stable and predictable path for the money stock is a characteristic of macroeconomic stability, much less a condition for monetary stability, strikes me as a complete misconception, a misconception propagated, more than anyone else, by Milton Friedman, himself.

Thus, contrary to Brad’s assertion, a stable and predictable path for the money stock is more likely than not to be a condition not for macroeconomic stability, but of macroeconomic instability. And to support my contention that a stable and predictable path for the money stock is macroeconomically destabilizing, let me quote none other than F. A. Hayek. I quote Hayek not because I think he is more authoritative than Friedman – Hayek having made more than his share of bad macroeconomic policy calls (e.g. his 1932 defense of the insane Bank of France) – but because in his own polite way he simply demolished the fallacy underlying Friedman’s fetish with a fixed rate of growth in the money stock (Full Empoyment at Any Price).

I wish I could share the confidence of my friend Milton Friedman who thinks that one could deprive the monetary authorities, in order to prevent the abuse of their powers for political purposes, of all discretionary powers by prescribing the amount of money they may and should add to circulation in any one year. It seems to me that he regards this as practicable because he has become used for statistical purposes to draw a sharp distinction between what is to be regarded as money and what is not. This distinction does not exist in the real world. I believe that, to ensure the convertibility of all kinds of near-money into real money, which is necessary if we are to avoid severe liquidity crises or panics, the monetary authorities must be given some discretion. But I agree with Friedman that we will have to try and get back to a more or less automatic system for regulating the quantity of money in ordinary times. The necessity of “suspending” Sir Robert Peel’s Bank Act of 1844 three times within 25 years after it was passed ought to have taught us this once and for all.

He was briefer and more pointed in a later comment (Denationalization of Money).

As regards Professor Friedman’s proposal of a legal limit on the rate at which a monopolistic issuer of money was to be allowed to increase the quantity in circulation, I can only say that I would not like to see what would happen if it ever became known that the amount of cash in circulation was approaching the upper limit and that therefore a need for increased liquidity could not be met.

And for good measure, Hayek added this footnote quoting Bagehot:

To such a situation the classic account of Walter Bagehot . . . would apply: “In a sensitive state of the English money market the near approach to the legal limit of reserve would be a sure incentive to panic; if one-third were fixed by law, the moment the banks were close to one-third, alarm would begin and would run like magic.

In other words if 3 is secured by institutions, all hell breaks loose.

But let us follow Brad a bit further in his quixotic quest to make Friedman seem sensible.

Now there are two different institutional setups that can produce (3):

  1. a monetarist central bank committed to targeting a k% growth rate of the money stock via open-market operations; or
  2. a gold standard in which a Humean price-specie flow mechanism leads inflating countries to lose and deflating countries to gain gold, tightly coupled to a banking system in which there is a reliable and stable money multiplier, and thus in which the money stock grows at the rate at which the world’s gold stock grows (plus the velocity trend).

Well, I have just – and not for the first time — disposed of 1, and in my previous post, I have disposed of 2. But having started to repeat myself, why not continue.

There are two points to make about the Humean price-specie-flow mechanism. First, it makes no sense, as Samuelson showed in his classic 1980 paper, inasmuch as it violates arbitrage conditions which do not allow the prices of tradable commodities to differ by more than the costs of transport. The Humean price-specie-flow mechanism presumes that the local domestic price levels are determined by local money supplies (either gold or convertible into gold), but that is simply not possible if arbitrage conditions obtain. There is no price-specie-flow mechanism under the gold standard, there is simply a movement of money sufficient to eliminate excess demands or supplies of money at the constant internationally determined price level. Domestic money supplies are endogenous and prices are (from the point of view of the monetary system) exogenously determined by the value of gold and the exchange rates of the local currencies in terms of gold. There is therefore no stable money multiplier at the level of a national currency (gold or convertible into gold). Friedman’s conception a pure [aka real] gold standard was predicated on a fallacy, namely the price-specie-flow mechanism. No gold standard in history ever operated as Friedman supposed that it operated. There were a few attempts to impose by statutory requirement a 100% (or sometime lower) marginal reserve requirement on banknotes, but that was statutory intervention, not a gold standard, which, at any operational level, is characterized by a fixed exchange rate between gold and the local currency with no restriction on the ability of economic agents to purchase gold at the going market price. the market price, under the gold standard, always equaling (or very closely approximating) the legal exchange rate between gold and the local currency.

Friedman calls (2) a “pure gold standard”. Anything else that claims to be a gold standard is and must be a “pseudo gold standard”. It might be a pseudo gold standard either because something disrupts the Humean price-specie flow mechanism–the “rules of the game” are not obeyed–so that deficit countries do not reliably lose and surplus countries do not reliably gain gold. It might be a pseudo gold standard because the money multiplier is not reliable and stable–because the banking system does not transparently and rapidly transmute a k% shift in the stock of gold into a k% shift in the money stock.

Friedman’s calling (2) “a pure [real] gold standard,” because it actualizes the Humean price-specie-flow-mechanism simply shows that Friedman understood neither the gold standard nor the price-specie-flow mechanism. The supposed rules of the game were designed to make the gold standard function in a particular way. In fact, the evidence shows that the classical gold standard in operation from roughly 1880 to 1914 operated with consistent departures from the “rules of the game.” What allows us to call the monetary regime in operation from 1880 to 1914 a gold standard is not that the rules of the game were observed but that the value of local currencies corresponded to the value of the gold with which they could be freely exchanged at the legal parities. No more and no less. And even Friedman was unwilling to call the gold standard in operation from 1880 to 1914 a pseudo gold standard, because if that was a pseudo-gold standard, there never was a real gold standard. So he was simply talking nonsense when he asserted that during the 1920s there a pseudo gold standard in operation even though gold was freely exchangeable for local currencies at the legal exchange rates.

Or, in short, to Friedman a gold standard is only a real gold standard if it produces a path for the money stock that is a k% rule. Anything else is a pseudo gold standard.

Yes! And that is what Friedman said, and it is absurd. And I am sure that Harry Johnson must have told him so.

The purpose of the paper, in short, is a Talmudic splitting-of-hairs. The point is to allow von Mises and Rueff and their not-so-deep-thinking latter-day followers (paging Paul Ryan! Paging Benn Steil! Paging Charles Koch! Paging Rand Paul!) to remain in their cloud-cuckoo-land of pledging allegiance to the gold standard as a golden calf while at the same time walling them off from and keeping them calm and supportive as the monetarist central bank does its job of keeping our fiat-money system stable by making Say’s Law true enough in practice.

As such, it succeeds admirably.

Or, at least, I think it does…

Have I just given an unconvincing Straussian reading of Friedman–that he knows what he is doing, and that what he is doing is leaving the theoretical husk to the fanatics von Mises and Rueff while keeping the rational kernel for himself, and making the point that a gold standard is a good monetary policy only if it turns out to mimic a good monetarist fiat-money standard policy? That his apparent confusion is simply a way of accomplishing those two tasks without splitting Mont Pelerin of the 1960s into yet more mutually-feuding camps?

I really sympathize with Brad’s effort to recruit Friedman into the worthy cause of combating nonsense. But you can’t combat nonsense with nonsense.


John Cochrane on the Failure of Macroeconomics

The state of modern macroeconomics is not good; John Cochrane, professor of finance at the University of Chicago, senior fellow of the Hoover Institution, and adjunct scholar of the Cato Institute, writing in Thursday’s Wall Street Journal, thinks macroeconomics is a failure. Perhaps so, but he has trouble explaining why.

The problem that Cochrane is chiefly focused on is slow growth.

Output per capita fell almost 10 percentage points below trend in the 2008 recession. It has since grown at less than 1.5%, and lost more ground relative to trend. Cumulative losses are many trillions of dollars, and growing. And the latest GDP report disappoints again, declining in the first quarter.

Sclerotic growth trumps every other economic problem. Without strong growth, our children and grandchildren will not see the great rise in health and living standards that we enjoy relative to our parents and grandparents. Without growth, our government’s already questionable ability to pay for health care, retirement and its debt evaporate. Without growth, the lot of the unfortunate will not improve. Without growth, U.S. military strength and our influence abroad must fade.

Macroeconomists offer two possible explanations for slow growth: a) too little demand — correctable through monetary or fiscal stimulus — and b) structural rigidities and impediments to growth, for which stimulus is no remedy. Cochrane is not a fan of the demand explanation.

The “demand” side initially cited New Keynesian macroeconomic models. In this view, the economy requires a sharply negative real (after inflation) rate of interest. But inflation is only 2%, and the Federal Reserve cannot lower interest rates below zero. Thus the current negative 2% real rate is too high, inducing people to save too much and spend too little.

New Keynesian models have also produced attractively magical policy predictions. Government spending, even if financed by taxes, and even if completely wasted, raises GDP. Larry Summers and Berkeley’s Brad DeLong write of a multiplier so large that spending generates enough taxes to pay for itself. Paul Krugman writes that even the “broken windows fallacy ceases to be a fallacy,” because replacing windows “can stimulate spending and raise employment.”

If you look hard at New-Keynesian models, however, this diagnosis and these policy predictions are fragile. There are many ways to generate the models’ predictions for GDP, employment and inflation from their underlying assumptions about how people behave. Some predict outsize multipliers and revive the broken-window fallacy. Others generate normal policy predictions—small multipliers and costly broken windows. None produces our steady low-inflation slump as a “demand” failure.

Cochrane’s characterization of what’s wrong with New Keynesian models is remarkably superficial. Slow growth, according to the New Keynesian model, is caused by the real interest rate being insufficiently negative, with the nominal rate at zero and inflation at (less than) 2%. So what is the problem? True, the nominal rate can’t go below zero, but where is it written that the upper bound on inflation is (or must be) 2%? Cochrane doesn’t say. Not only doesn’t he say, he doesn’t even seem interested. It might be that something really terrible would happen if the rate of inflation rose about 2%, but if so, Cochrane or somebody needs to explain why terrible calamities did not befall us during all those comparatively glorious bygone years when the rate of inflation consistently exceeded 2% while real economic growth was at least a percentage point higher than it is now. Perhaps, like Fischer Black, Cochrane believes that the rate of inflation has nothing to do with monetary or fiscal policy. But that is certainly not the standard interpretation of the New Keynesian model that he is using as the archetype for modern demand-management macroeconomic theories. And if Cochrane does believe that the rate of inflation is not determined by either monetary policy or fiscal policy, he ought to come out and say so.

Cochrane thinks that persistent low inflation and low growth together pose a problem for New Keynesian theories. Indeed it does, but it doesn’t seem that a radical revision of New Keynesian theory would be required to cope with that state of affairs. Cochrane thinks otherwise.

These problems [i.e., a steady low-inflation slump, aka “secular stagnation”] are recognized, and now academics such as Brown University’s Gauti Eggertsson and Neil Mehrotra are busy tweaking the models to address them. Good. But models that someone might get to work in the future are not ready to drive trillions of dollars of public expenditure.

In other words, unless the economic model has already been worked out before a particular economic problem arises, no economic policy conclusions may be deduced from that economic model. May I call  this Cochrane’s rule?

Cochrane the proceeds to accuse those who look to traditional Keynesian ideas of rejecting science.

The reaction in policy circles to these problems is instead a full-on retreat, not just from the admirable rigor of New Keynesian modeling, but from the attempt to make economics scientific at all.

Messrs. DeLong and Summers and Johns Hopkins’s Laurence Ball capture this feeling well, writing in a recent paper that “the appropriate new thinking is largely old thinking: traditional Keynesian ideas of the 1930s to 1960s.” That is, from before the 1960s when Keynesian thinking was quantified, fed into computers and checked against data; and before the 1970s, when that check failed, and other economists built new and more coherent models. Paul Krugman likewise rails against “generations of economists” who are “viewing the world through a haze of equations.”

Well, maybe they’re right. Social sciences can go off the rails for 50 years. I think Keynesian economics did just that. But if economics is as ephemeral as philosophy or literature, then it cannot don the mantle of scientific expertise to demand trillions of public expenditure.

This is political rhetoric wrapped in a cloak of scientific objectivity. We don’t have the luxury of knowing in advance what the consequences of our actions will be. The United States has spent trillions of dollars on all kinds of stuff over the past dozen years or so. A lot of it has not worked out well at all. So it is altogether fitting and proper for us to be skeptical about whether we will get our money’s worth for whatever the government proposes to spend on our behalf. But Cochrane’s implicit demand that money only be spent if there is some sort of scientific certainty that the money will be well spent can never be met. However, as Larry Summers has pointed out, there are certainly many worthwhile infrastructure projects that could be undertaken, so the risk of committing the “broken windows fallacy” is small. With the government able to borrow at negative real interest rates, the present value of funding such projects is almost certainly positive. So one wonders what is the scientific basis for not funding those projects?

Cochrane compares macroeconomics to climate science:

The climate policy establishment also wants to spend trillions of dollars, and cites scientific literature, imperfect and contentious as that literature may be. Imagine how much less persuasive they would be if they instead denied published climate science since 1975 and bemoaned climate models’ “haze of equations”; if they told us to go back to the complex writings of a weather guru from the 1930s Dustbowl, as they interpret his writings. That’s the current argument for fiscal stimulus.

Cochrane writes as if there were some important scientific breakthrough made by modern macroeconomics — “the new and more coherent models,” either the New Keynesian version of New Classical macroeconomics or Real Business Cycle Theory — that rendered traditional Keynesian economics obsolete or outdated. I have never been a devote of Keynesian economics, but the fact is that modern macroeconomics has achieved its ascendancy in academic circles almost entirely by way of a misguided methodological preference for axiomatized intertemporal optimization models for which a unique equilibrium solution can be found by imposing the empirically risible assumption of rational expectations. These models, whether in their New Keynesian or Real Business Cycle versions, do not generate better empirical predictions than the old fashioned Keynesian models, and, as Noah Smith has usefully pointed out, these models have been consistently rejected by private forecasters in favor of the traditional Keynesian models. It is only the dominant clique of ivory-tower intellectuals that cultivate and nurture these models. The notion that such models are entitled to any special authority or scientific status is based on nothing but the exaggerated self-esteem that is characteristic of almost every intellectual clique, particularly dominant ones.

Having rejected inadequate demand as a cause of slow growth, Cochrane, relying on no model and no evidence, makes a pitch for uncertainty as the source of slow growth.

Where, instead, are the problems? John Taylor, Stanford’s Nick Bloom and Chicago Booth’s Steve Davis see the uncertainty induced by seat-of-the-pants policy at fault. Who wants to hire, lend or invest when the next stroke of the presidential pen or Justice Department witch hunt can undo all the hard work? Ed Prescott emphasizes large distorting taxes and intrusive regulations. The University of Chicago’s Casey Mulligan deconstructs the unintended disincentives of social programs. And so forth. These problems did not cause the recession. But they are worse now, and they can impede recovery and retard growth.

Where, one wonders, is the science on which this sort of seat-of-the-pants speculation is based? Is there any evidence, for example, that the tax burden on businesses or individuals is greater now than it was let us say in 1983-85 when, under President Reagan, the economy, despite annual tax increases partially reversing the 1981 cuts enacted in Reagan’s first year, began recovering rapidly from the 1981-82 recession?

The Enchanted James Grant Expounds Eloquently on the Esthetics of the Gold Standard

One of the leading financial journalists of our time, James Grant is obviously a very smart, very well read, commentator on contemporary business and finance. He also has published several highly regarded historical studies, and according to the biographical tag on his review of a new book on the monetary role of gold in the weekend Wall Street Journal, he will soon publish a new historical study of the dearly beloved 1920-21 depression, a study that will certainly be worth reading, if not entirely worth believing. Grant reviewed a new book, War and Gold, by Kwasi Kwarteng, which provides a historical account of the role of gold in monetary affairs and in wartime finance since the 16th century. Despite his admiration for Kwarteng’s work, Grant betrays more than a little annoyance and exasperation with Kwarteng’s failure to appreciate what a many-splendored thing gold really is, deploring the impartial attitude to gold taken by Kwarteng.

Exasperatingly, the author, a University of Cambridge Ph. D. in history and a British parliamentarian, refuses to render historical judgment. He doesn’t exactly decry the world’s descent into “too big to fail” banking, occult-style central banking and tiny, government-issued interest rates. Neither does he precisely support those offenses against wholesome finance. He is neither for the dematerialized, non-gold dollar nor against it. He is a monetary Hamlet.

He does, at least, ask: “Why gold?” I would answer: “Because it’s money, or used to be money, and will likely one day become money again.” The value of gold is inherent, not conferred by governments. Its supply tends to grow by 1% to 2% a year, in line with growth in world population. It is nice to look at and self-evidently valuable.

Evidently, Mr. Grant’s enchantment with gold has led him into incoherence. Is gold money or isn’t it? Obviously not — at least not if you believe that definitions ought to correspond to reality rather than to Platonic ideal forms. Sensing that his grip on reality may be questionable, he tries to have it both ways. If gold isn’t money now, it likely will become money again — “one day.” For sure, gold used to be money, but so did cowerie shells, cattle, and at least a dozen other substances. How does that create any presumption that gold is likely to become money again?

Then we read: “The value of gold is inherent.” OMG! And this from a self-proclaimed Austrian! Has he ever heard of the “subjective theory of value?” Mr. Grant, meet Ludwig von Mises.

Value is not intrinsic, it is not in things. It is within us. (Human Action p. 96)

If value “is not in things,” how can anything be “self-evidently valuable?”

Grant, in his emotional attachment to gold, feels obligated to defend the metal against any charge that it may have been responsible for human suffering.

Shelley wrote lines of poetry to protest the deflation that attended Britain’s return to the gold standard after the Napoleonic wars. Mr. Kwarteng quotes them: “Let the Ghost of Gold / Take from Toil a thousandfold / More than e’er its substance could / In the tyrannies of old.” The author seems to agree with the poet.

Grant responds to this unfair slur against gold:

I myself hold the gold standard blameless. The source of the postwar depression was rather the decision of the British government to return to the level of prices and wages that prevailed before the war, a decision it enforced through monetary means (that is, by reimposing the prewar exchange rate). It was an error that Britain repeated after World War I.

This is a remarkable and fanciful defense, suggesting that the British government actually had a specific target level of prices and wages in mind when it restored the pound to its prewar gold parity. In fact, the idea of a price level was not yet even understood by most economists, let alone by the British government. Restoring the pound to its prewar parity was considered a matter of financial rectitude and honor, not a matter of economic fine-tuning. Nor was the choice of the prewar parity the only reason for the ruinous deflation that followed the postwar resumption of gold payments. The replacement of paper pounds with gold pounds implied a significant increase in the total demand for gold by the world’s leading economic power, which implied an increase in the total world demand for gold, and an increase in its value relative to other commodities, in other words deflation. David Ricardo foresaw the deflationary consequences of the resumption of gold payments, and tried to mitigate those consequences with his Proposals for an Economical and Secure Currency, designed to limit the increase in the monetary demand for gold. The real error after World War I, as Hawtrey and Cassel both pointed out in 1919, was that the resumption of an international gold standard after gold had been effectively demonetized during World War I would lead to an enormous increase in the monetary demand for gold, causing a worldwide deflationary collapse. After the Napoleonic wars, the gold standard was still a peculiarly British institution, the rest of the world then operating on a silver standard.

Grant makes further extravagant and unsupported claims on behalf of the gold standard:

The classical gold standard, in service roughly from 1815 to 1914, was certainly imperfect. What it did deliver was long-term price stability. What the politics of the gold-standard era delivered was modest levels of government borrowing.

The choice of 1815 as the start of the gold standard era is quite arbitrary, 1815 being the year that Britain defeated Napoleonic France, thereby setting the stage for the restoration of the golden pound at its prewar parity. But the very fact that 1815 marked the beginning of the restoration of the prewar gold parity with sterling shows that for Britain the gold standard began much earlier, actually 1717 when Isaac Newton, then master of the mint, established the gold parity at a level that overvalued gold, thereby driving silver out of circulation. So, if the gold standard somehow ensures that government borrowing levels are modest, one would think that borrowing by the British government would have been modest from 1717 to 1797 when the gold standard was suspended. But the chart below showing British government debt as a percentage of GDP from 1692 to 2010 shows that British government debt rose rapidly over most of the 18th century.

uk_national_debtGrant suggests that bad behavior by banks is mainly the result of abandonment of the gold standard.

Progress is the rule, the Whig theory of history teaches, but the old Whigs never met the new bankers. Ordinary people live longer and Olympians run faster than they did a century ago, but no such improvement is evident in our monetary and banking affairs. On the contrary, the dollar commands but 1/1,300th of an ounce of gold today, as compared with the 1/20th of an ounce on the eve of World War I. As for banking, the dismal record of 2007-09 would seem inexplicable to the financial leaders of the Model T era. One of these ancients, Comptroller of the Currency John Skelton Williams, predicted in 1920 that bank failures would soon be unimaginable. In 2008, it was solvency you almost couldn’t imagine.

Once again, the claims that Mr. Grant makes on behalf of the gold standard simply do not correspond to reality. The chart below shows the annual number of bank failures in every years since 1920.


Somehow, Mr. Grant somehow seems to have overlooked what happened between 1929 and 1932. John Skelton Williams obviously didn’t know what was going to happen in the following decade. Certainly no shame in that. I am guessing that Mr. Grant does know what happened; he just seems too bedazzled by the beauty of the gold standard to care.

Further Thoughts on Capital and Inequality

In a recent post, I criticized, perhaps without adequate understanding, some of Thomas Piketty’s arguments about capital in his best-selling book. My main criticism is that Piketty’s argument that. under capitalism, there is an inherent tendency toward increasing inequality, ignores the heterogeneity of capital and the tendency for new capital embodying new knowledge, new techniques, and new technologies to render older capital obsolete. Contrary to the simple model of accumulation on which Piketty relies, the accumulation of capital is not a smooth process; it is a very uneven process, generating very high returns to some owners of capital, but also imposing substantial losses on other owners of capital. The only way to avoid the risk of owning suddenly obsolescent capital is to own the market portfolio. But I conjecture that few, if any, great fortunes have been amassed by investing in the market portfolio, and (I further conjecture) great fortunes, once amassed, are usually not liquidated and reinvested in the market portfolio, but continue to be weighted heavily in fairly narrow portfolios of assets from which those great fortunes grew. Great fortunes, aside from being dissipated by deliberate capital consumption, also tend to be eroded by the loss of value through obsolescence, a process that can only be avoided by extreme diversification of holdings or by the exercise of entrepreneurial skill, a skill rarely bequeathed from generation to generation.

Applying this insight, Larry Summers pointed out in his review of Piketty’s book that the rate of turnover in the Forbes list of the 400 wealthiest individuals between 1982 and 2012 was much higher than the turnover predicted by Piketty’s simple accumulation model. Commenting on my post (in which I referred to Summers’s review), Kevin Donoghue objected that Piketty had criticized the Forbes 400 as a measure of wealth in his book, so that Piketty would not necessarily accept Summers’ criticism based on the Forbes 400. Well, as an alternative, let’s have a look at the S&P 500. I just found this study of the rate of turnover in the 500 firms making up the S&P 500, showing that the rate of turnover in the composition of the S&P 500 has been increased greatly over the past 50 years. See the chart below copied from that study showing that the average length of time for firms on the S&P 500 was over 60 years in 1958, but by 2011 had fallen to less than 20 years. The pace of creative destruction seems to be accelerating


From the same study here’s another chart showing the companies that were deleted from the index between 2001 and 2011 and those that were added.


But I would also add a cautionary note that, because the population of individuals and publicly held business firms is growing, comparing the composition of a fixed number (400) of wealthiest individuals or (500) most successful corporations over time may overstate the increase over time in the rate of turnover, any group of fixed numerical size becoming a smaller percentage of the population over time. Even with that caveat, however, what this tells me is that there is a lot of variability in the value of capital assets. Wealth grows, but it grows unevenly. Capital is accumulated, but it is also lost.

Does the process of capital accumulation necessarily lead to increasing inequality of wealth and income? Perhaps, but I don’t think that the answer is necessarily determined by the relationship between the real rate of interest and the rate of growth in GDP.

Many people have suggested that an important cause of rising inequality has been the increasing importance of winner-take-all markets in which a few top performers seem to be compensated at very much higher rates than other, only slightly less gifted, performers. This sort of inequality is reflected in widening gaps between the highest and lowest paid participants in a given occupation. In some cases at least, the differences between the highest and lowest paid don’t seem to correspond to the differences in skill, though admittedly skill is often difficult to measure.

This concentration of rewards is especially characteristic of competitive sports, winners gaining much larger rewards than losers. However, because the winner’s return comes, at least in part, at the expense of the loser, the private gain to winning exceeds the social gain. That’s why all organized professional sports engage in some form of revenue sharing and impose limits on spending on players. Without such measures, competitive sports would not be viable, because the private return to improve quality exceeds the collective return from improved quality. There are, of course, times when a superstar like Babe Ruth or Michael Jordan can actually increase the return to losers, but that seems to be the exception.

To what extent other sorts of winner-take-all markets share this intrinsic inefficiency is not immediately clear to me, but it does not seem implausible to think that there is an incentive to overinvest in skills that increase the expected return to participants in winner-take-all markets. If so, the source of inequality may also be a source of inefficiency.

About Me

David Glasner
Washington, DC

I am an economist at the Federal Trade Commission. Nothing that you read on this blog necessarily reflects the views of the FTC or the individual commissioners. Although I work at the FTC as an antitrust economist, most of my research and writing has been on monetary economics and policy and the history of monetary theory. In my book Free Banking and Monetary Reform, I argued for a non-Monetarist non-Keynesian approach to monetary policy, based on a theory of a competitive supply of money. Over the years, I have become increasingly impressed by the similarities between my approach and that of R. G. Hawtrey and hope to bring Hawtrey's unduly neglected contributions to the attention of a wider audience.

Enter your email address to follow this blog and receive notifications of new posts by email.

Join 374 other followers


Get every new post delivered to your Inbox.

Join 374 other followers