James Forder is mad as hell, and he's not going to take it anymore. Turns out that everybody—absolutely everybody—has the history of macro in the 1960s and 1970s totally screwed up. Rather than being at the center of the story, Professor Phillips and his curve were at best a sideshow. Seems that in 1958, when Phillips used inconsistent data and laughably crude statistical techniques to slap together a paper on the relationship between wage inflation and unemployment in the United Kingdom, no one took the paper seriously—not even Phillips (“The Relationship between Unemployment and the Rate of Change of Money Wages in the United Kingdom 1861–1957.” Economica 25 (1958): 283–299). A couple of years later Lipsey used better data and better techniques to write the paper Phillips should have written. But nobody paid much attention to Lipsey's paper either. Then in 1960 Samuelson and Solow wrote a paper looking at the relationship between price inflation and unemployment in the United States. But neither they nor anyone else—even for a minute—believed that the relationship represented a structural tradeoff between unemployment and inflation that could be exploited by policymakers. Until … Friedman, in his 1968 AEA president address and 1977 Nobel address, in a feat of rhetorical legerdemain never equaled before or since, convinced everybody that they had considered the Phillips curve to represent an exploitable tradeoff between unemployment and inflation, even though in reality, they had not. Moreover, Friedman's key observation in his AEA address that because the expected rate of inflation converges to the actual rate of inflation, changes in nominal demand cannot affect real variables in the long run, was neither interesting nor original because everybody dating back to John Stuart Mill in 1844 already understood it.

Not often do you read a book about relatively recent events that claims nobody but the author understands what really happened. Here is Forder's argument more fleshed out. At the time Phillips published his paper, most economists writing about wage inflation focused on institutional factors, such as the strength of labor unions, the existence of government income programs, the competitive structure of industries, perceptions of fairness, and so on. With respect to strictly economic variables, corporate profits and productivity were thought to be particularly important. The level of unemployment was seen as playing a secondary role—or perhaps no role at all—in explaining the rate of wage change. Because institutional factors were critical, the determinants of wage inflation change over time, and it was unlikely that a stable structural relationship between wage inflation and any set of variables would ever be discovered. What distinguished Phillips's 1958 paper was his focus on the level of unemployment as the key determinant of wage inflation and his claim to have found in the U.K. data a stable relationship between these two variables over a span of decades. Because Phillips's approach was so at odds with the literature of the time and because there were problems with both his data and his statistical techniques, his paper had very little effect on research during the immediately following years.

Samuelson and Solow's 1960 AER paper (“Analytical Aspects of Anti-Inflation Policy.” American Economic Review: Papers and Proceedings 50 (1960): 177–94) also has assumed a much greater importance in retrospect than it did at the time. Their observation that the inflation-unemployment schedule they plotted from U.S. data represented a “menu of choice” (presumably it was policymakers who would be choosing from the menu) was not the main focus of the paper, which was instead meant as a discussion of the conditions necessary for price stability. Very little—at times Forder seems to want to say “none”—of the subsequent macro literature up through the time of Friedman's AEA address focused on analyzing why there might be a trade-off between unemployment and inflation, and almost no one advocated buying a lower unemployment rate at the cost of higher inflation.

Friedman himself in his 1968 address (“The Role of Monetary Policy.” American Economic Review 58 (1968): 1–17) did not directly accuse policymakers of having assumed that they could indefinitely buy a lower unemployment rate with a higher inflation rate; that claim came later, in his 1977 Nobel address. In the intervening years, the tectonic plates of macro had shifted and the institutionally oriented approach, focused on understanding short-run movements in macro variables and—secondarily—providing policy guidance had begun its decent into professional oblivion. The new wave, dependent more on Phelps and Lucas than on Friedman, was concerned with the determinants of long-run equilibrium and working out the implications of the assumption that long-run equilibrium resulted from the actions of rational, forward-thinking actors.

Forder argues that Friedman in his Nobel address was able to pass off an essentially phony history of the previous 20 years. That Friedman's version of history has been almost universally accepted is partly due to its harmony with modern ways of thinking about macro and partly due to an odd failure of other macroeconomists to point out Friedman's errors before his story became embedded in the hard concrete of textbooks.

You cannot read Forder's book without agreeing with a good part of his argument: Phillips did not have much influence on macro theory and policy during the 1960s; Samuelson and Solow, and most other prominent Keynesians, did not spend the 1960s advocating that the Fed buy lower unemployment at the cost of higher inflation; few economists in the 1960s (or before) were confused about the distinction between nominal and real variables or ignored the role of the expected inflation rate on the determination of nominal wages; and Friedman's 1977 Nobel address is far from a reliable account of the preceding 20 years.

But I also think that Forder has overstated his case. First, while Samuelson and Solow may not have advocated buying lower unemployment with higher inflation, the idea of a tradeoff—“a menu of choice”—is in their 1960 AER article, which contains no real recognition that the inflation-unemployment schedule they present is a short-run phenomenon dependent, at least in part, on workers' and firms' expectation that inflation would remain low. Exploring the nature of the inflation-unemployment tradeoff was not entirely absent from 1960s macro, even if it was not its focus. Second, a key part of the traditional story is that policymakers in the United States and United Kingdom (and perhaps elsewhere) during the 1960s and 1970s were influenced to at least some extent by the idea that a stable tradeoff existed. But Forder spends very little time on this point. He provides next to no discussion of the basis of Fed policymaking during these years. He acknowledges that Tobin mentioned in 1972 that in the early 1960s the Council of Economic Advisers, of which Tobin was a member, had considered the Phillips curve when formulating policy, only to dismiss Tobin's recollection (of then fairly recent events) as mistaken. To my mind, Forder should have devoted more space to what policymakers understood the relationship between unemployment and inflation to be, and much less space to summarizing scores of journal articles.

Finally, the immediate reaction to Friedman's 1968 AEA address by Solow and at least some other Keynesians was to dispute Friedman's contention that the short-run tradeoff between unemployment and inflation eventually breaks down as the expected inflation rate converges to the actual inflation rate. If Friedman's address consisted of not much more than a restatement of a truism about the distinction between real and nominal variables and the observation that Fed policy is unable to affect real variables in the long run, it's difficult to understand why the address was immediately seen as important or why it attracted the kinds of responses that it did. If the idea of a short-run tradeoff was unimportant for policy and if everyone knew that changes in nominal demand cannot affect real variables in the long-run, then why was there a raft of papers meant to “test” Friedman's story? The answer is that Friedman's address was not seen as a restatement of obvious points that were unrelated to real-world policymaking, but rather the opposite: Friedman was seen as saying something new; something that was contrary to what at least some prominent economists saw as one of the underpinnings of macro policy.

So, while I do not buy the whole of Forder's story, he's written an important book that should be read by anyone interested in the history of macroeconomics.