Tuesday, August 13, 2013

Friedman's Legacy

I'm not sure why, but there has been a lot of blogosphere writing on Milton Friedman recently, including posts by Paul Krugman and Tyler Cowen, among the usual suspects. Randy Wright once convinced me that we should call ourselves New Monetarists, and we wrote a couple of papers (this one, and this one) in which we try to get a grip on what that means. As New Monetarists, we think we have something to say about Friedman.

We can find plenty of faults in Friedman's ideas, but those ideas - reflected in Friedman's theoretical and empirical work - are deeply embedded in much of what we do as economists in the 21st century. By modern standards, Friedman was a crude economic theorist, but he used the simple tools he had available to develop deep ideas that were later fleshed out in fully-articulated economic models. His empirical work was highly influential and serves as a key reference point for some sub-fields in economics. Some examples:

1. Permanent Income Theory: The key idea here is that wealth is the primary determinant of consumption behavior. This serves as the basis for modern consumption-smoothing theories of consumption/savings decisions, modified to take account of various credit market frictions that we need to fit the data. The permanent income logic is a basic piece of intuition that macroeconomists use all the time - one of those ideas that are embedded in what we do.

2. The Friedman rule: Don't confuse this with the constant money growth rule, which comes from "The Role for Monetary Policy." The "Friedman rule" is the policy rule in the "Optimum Quantity of Money" essay. Basically, the nominal interest rate reflects a distortion. Eliminating that distortion requires reducing the nominal interest rate to zero in all states of the world, and that's what monetary policy should be aimed at doing, according to Friedman's logic in his paper. That logic is watertight, as it turns out there exists a wide class of conventional monetary models that yield Friedman's conclusion. We can think of plenty of good reasons why optimal monetary policy could take us away from the Friedman rule in practice, but whenever someone makes an argument for some monetary policy rule, we have to first ask the question: why isn't that rule the Friedman rule? The Friedman rule is fundamental in monetary theory.

3. Monetary history: Friedman and Schwartz's "Monetary History of the United States" was monumental. Part of the project involved data collection, including the construction of time series for monetary measures going back to the 19th century. Friedman wanted to make the case that monetary phenomena were important for economic activity, and that monetary policy matters. We may have good reasons to object to Friedman's view of how monetary policy works (if he even had a view that differed from conventional Old Keynesian ideas), or his monetarist black box, but the Monetary History is a touchstone, particularly for people who work on the Great Depression.

4. Policy rules: The rule that Friedman wanted central banks to follow was not the Friedman rule, but a constant-money-growth rule - that's in his 1968 AER paper. Friedman was successful in getting the rule adopted by central banks in the 1970s and 1980s, but the rule was a practical failure, for reasons that are well-understood. But Friedman got macroeconomists and policymakers thinking about policy rules and how they work. Out of that thinking came ideas about central bank commitment, Taylor rules, inflation targeting, nominal GDP targeting, thresholds, etc., that form the basis for modern analysis of central bank policy.

5. Money and Inflation: People seem to forget this, but in the 1970s there were many people, including prominent members of the economics profession such as Robert Solow, who thought that the way to control inflation was by way of "incomes policy." By that they meant wage and price controls. Friedman played a key role in convincing economists and policymakers that central banks could, and should, control inflation. That seems as natural today as saying that rain falls from the sky, and that's part of Friedman's influence.

6. Narrow banking: I tend to think this was one of Friedman's bad ideas, but it's been very influential. Friedman advocated a 100% reserve requirement in "A Program for Monetary Stability." Friedman thought of money as the stuff that is used in transactions - currency, transactions deposits at banks, etc. He also thought it was important to control the nominal quantity of that stuff. So, since the central bank controls the quantity of outside money (currency and reserves), if transactions deposits at banks are backed one-for-one with reserves, then the central bank controls the quantity of money (inside plus outside). Three problems with this: (i) moneyness is only a matter of degree; it's useless to try to define some stuff as "money" and other stuff as "not money." (ii) there are incentives to get around the 100% reserve requirement by starting another type of financial intermediary (a shadow bank) that isn't a bank in terms of regulations, but for all intents and purposes acts like one. (iii) financial intermediation performs a useful social role - the 100% reserve requirement is then a tax on something useful. Friedman's ideas in this respect are reflected in modern proposals for narrow banking. You can find some of that in Gary Gorton's work, for example, and I don't agree with it.

6. Counterpoint to Keynesian economics: Some people seem to think that Friedman was actually a Keynesian at heart, but he sure got on Tobin's nerves. Criticism is important - it helps to prevent and root out lazy science. Old Keynesian economics was probably much better - e.g. there would have been no "neoclassical synthesis" - because of Friedman.

If anyone wants to argue that Friedman is now unimportant for modern economics, that's like saying Bob Dylan is unimportant for modern music. Today, Bob Dylan is quite willing to climb on a stage and perform with a world-class group of musicians - but it's truly pathetic. Nevertheless, Bob Dylan doesn't get booed off the stage today, because people recognize his importance. In the 1960s, he got people riled up, everyone paid attention, and the world is much different today than it would have been if he had not done the work he did.


  1. I was persuaded, until you called Dylan's performances pathetic! Now I question everything you wrote!

    1. The recent records are OK. I even find something to like in his Christmas record - part of that is a weird joke, but in another way I think he is totally serious. The performances are another thing altogether. Ten years ago, he was one of the best things going, but now I think it's a little embarrassing.

  2. Note also that he was ahead of his time (in 1950!) when he began to make the case for flexible exchange rates.

  3. i sense some contradictions in your posting - you agree with Friedman that inflation is caused by excess money growth but also hold that there is no meaningful asset we can call money, so how can the central bank control inflation ?