James Tobin

R. Dimand

• Publisher: Palgrave Macmillan
• Year: 2014
• Print ISBN: 9781137431943
• eBook ISBN: 9781137431950
• Topic: Economics
• Subtopic: Econometrics
• Index: Economics

James Tobin, 1981 Nobel laureate in economics, was the outstanding monetary economist among American Keynesian economists. This book, the first written about James Tobin, examines his leading role as a Keynesian macroeconomist and monetary economist, and considers the continuing relevance of his ideas.

1

An American Keynesian

Introduction

In September 1936, when James Tobin was an 18-year-old sophomore taking “Principles of Economics” (Ec A) at Harvard, his tutor, Spencer Pollard (a graduate student who was also the instructor of Tobin’s Ec A section) “decided that for tutorial he and I, mainly I, should read ‘this new book from England. They say it may be important.’ So I plunged in, being too young and ignorant to know that I was too young and ignorant” to begin the study of economics by reading Keynes’s The General Theory of Employment, Interest and Money (Tobin 1988, 662). Pollard was right: the book did turn out to be important, not least for its lasting role in shaping Tobin’s intellectual development. Tobin (1992, 1993) remained proud to call himself an “Old Keynesian” in contrast to New Keynesian, New Classical, and Post Keynesian, and, when Harcourt and Riach (1997) edited A “Second Edition” of The General Theory, it was fitting that they invited Tobin (1997) to contribute the overview chapter, with the first part of the chapter written “as J. M. Keynes.”¹ Although Sir John Hicks (1935, 1937, 1939) and Irving Fisher also influenced Tobin², his approach to economics was always most deeply shaped by Keynes and by Tobin’s experience growing up during the Great Depression of the 1930s.

Throughout his career, Tobin was concerned with developing macroeconomic theory that would be relevant for stabilization policy – to prevent another depression and to improve people’s lives by promoting growth and stability – rather than with analytical problem-solving for its own sake. The Great Depression was associated with the breakdown of the US banking system and with Keynes’s argument that depression due to inadequate effective demand was a distinctive problem of a monetary economy as opposed to a barter economy. More than any of the other leading American Keynesians of his generation, such as Nobel laureates Paul Samuelson, Robert Solow, Franco Modigliani, and Lawrence Klein, Tobin concerned himself with the functioning and malfunctioning of the monetary system, telling David Colander (1999, 121) “I differed from that group [American Keynesians in the 1950s] in that I taught that monetary policy was a possible tool of macroeconomic policy and that to neglect it was a mistake.”

Tobin set himself apart from Keynes’s disciples at Cambridge University (such as Joan Robinson, Richard Kahn, and Nicholas Kaldor) and their post Keynesian allies in the United States because he objected to “throwing away the insights of neoclassical economics” (in Colander 1999, 121). Even his late-career mellowing toward the British side of the Cambridge capital controversies was subtitled, “A Neoclassical Kaldor-Robinson Exercise” (Tobin 1989b). But he also stood aside from New Keynesians: ‘If it means people like Greg Mankiw, I don’t regard them as Keynesians. I don’t think they have involuntary unemployment or absence of market clearing,’ just nominal wage and price stickiness, in contrast to Keynes’s insistence that nominal wage and price flexibility could not be relied upon to eliminate unemployment.” (Tobin in Colander 1999, 124; Keynes 1936, Chapter 19).

Tobin thus staked a distinctive claim to Keynes’s contested heritage. He reiterated this claim to be a Keynesian throughout his career, using Keynes’s term “liquidity preference” in the title of his article on demand for money as an asset (Tobin 1958a), linking the proposed Tobin tax to restrain international currency speculation to Keynes’s proposed turnover tax to curb stock market speculation (Keynes 1936, 160; Tobin 1984a), and building his theory of investment around Tobin’s q (Brainard and Tobin 1968, Tobin and Brainard 1977), a concept closely related in both substance and notation to the Q of Keynes’s Treatise on Money (1930), notation that Keynes had chosen because of Alfred Marshall’s quasi-rents.

The central propositions of Keynes’s General Theory according to Tobin

In “How Dead is Keynes?” Tobin (1977a) summarized the central message of Keynes’s General Theory in four propositions and argued that reports of the death of Keynes, like those of the demise of Mark Twain, were much exaggerated:

None of the four central Keynesian propositions is inconsistent with the contemporary economic scene here or in other advanced democratic capitalist countries. At least the first three fit the facts extremely well. Indeed the middle 70s follow the Keynesian script better than any post-war period except the early 60s. It hardly seems the time for a funeral. (460)

What Keynes meant to convey as his central message is hotly contested, with a huge literature emphasizing nominal wage stickiness, or discretionary policy, or fundamental uncertainty, or rejection of Say’s Law, or the spending multiplier. But while Tobin’s four central Keynesian propositions do not settle the controversies over what Keynes meant (let alone the controversies over whether Keynes was, in some sense, right), they show what being a Keynesian meant to Tobin.

Tobin’s first central Keynesian proposition was that “In modern industrial capitalist societies, wages respond slowly to excess demand or supply, especially slowly to excess supply,” so that over “a long short run” fluctuations in aggregate demand affect real output, not just prices. A corollary of this was the second proposition: “the vulnerability of economies like ours to lengthy bouts of involuntary unemployment.” The only distinctively Keynesian aspect of Tobin’s first two central Keynesian propositions was the insistence on the phenomenon of involuntary unemployment, an excess supply of labor in a non-clearing labor market. Replace “involuntary unemployment” with “high unemployment” in the second proposition, and the two propositions would be acceptable to David Hume in 1752, Henry Thornton in 1802, Alfred Marshall in 1887, or Milton Friedman (1968). Tobin (1977a, 459–60) pointed to the high unemployment since 1974 as supporting evidence, insisting that the increased unemployment was indeed involuntary: “People willing to work at or below prevailing real wages cannot find jobs. They have no effective way to signal their availability.” By contrast, in Friedman (1968) with adaptive expectations and the expectations-augmented Phillips curve, and in Lucas (1981a) with the monetary-misperceptions version of New Classical economics, the labor market clears, but the labor demand curve shifts as workers are fooled by monetary shocks into misperceiving the real wage.

Tobin’s first two Keynesian propositions summarized widely shared views (although New Classical economists would be troubled by the very idea of involuntary behavior), and came to textbook Keynesianism from Chapter 2 of Keynes (1936), in which Keynes discussed the two classical postulates of the labor market. Keynes accepted the first classical postulate – that the real wage is equal to the marginal product of labor (that is, firms are competitive and on the labor demand curve) – but rejected the second one that the utility of the real wage is equal to the marginal disutility of labor (that is, labor is on the labor supply curve). Although Keynes’s Chapter 2 provided an account of why staggered contracts and concern of workers with relative wages could make nominal wages sticky downwards without any money illusion (a precursor of the more formal modeling of Taylor 1980³), the textbook version and Tobin’s first two Keynesian propositions were consistent with the claim that Keynesian analysis, however practically important, was theoretically trivial: just a classical system with a sticky nominal wage rate. Emphasizing slow adjustment of prices and money wages implied viewing Keynesian unemployment as a disequilibrium situation, a short-run phenomenon of transition periods, rather than accepting Keynes’s claim to have shown the possibility of equilibrium with involuntary unemployment (excess supply of labor).

Writing as J. M. Keynes for A “Second Edition” of The General Theory, Tobin (1997, 7) held that Keynes (1936, Chapter 2)

leaned too far to the classical side, as I learned shortly after the book was published, thanks to the empirical studies of [John] Dunlop and [Lorie] Tarshis. If the first classical postulate were correct, then we would expect real wages – measured in terms of labour’s product rather than workers’ consumption – to move counter-cyclically. However, Dunlop and Tarshis found that product-wages were, if anything, pro-cyclical. This is not a fatal flaw in the general theory; quite the contrary: my essential propositions remain unscathed. ... If increases in aggregate demand can raise employment and output without diminishing real wages, so much the better! ... Nothing is lost by recognizing that imperfect competition and sluggish price adjustment may result in departures from marginal cost pricing, especially in short runs.

(See articles by Dunlop, Tarshis, Keynes, and Ruggles reprinted, together with Tobin 1941, in Dimand 2002, Volume VIII.)

Tobin’s third central Keynesian proposition was that:

Capital formation depends on long run appraisals of profit expectations and risks and on business attitudes toward bearing the risks. These are not simple predictable functions of current and recent economic events. Variations of the marginal efficiency of capital contain, for all practical purposes, important elements of autonomy and exogeneity. (1977a, 460, cf. Keynes 1936, Chapter 12, “The State of Long-Term Expectation”)

This emphasis on autonomous shifts of long-period expectations (Keynes’s “animal spirits”) rejected the rational expectations hypothesis introduced into macroeconomics in the 1970s by Robert Lucas (1981a) and Thomas Sargent and Neil Wallace (1976), as well as the endogenous, adaptive expectations of Friedman (1968). Tobin’s emphasis on fluctuations in long-period expectations of future profits fit with a view that the Wall Street crash of October 1929 mattered for investment and the Great Depression (the market value of equity, the numerator of Tobin’s q, is the present discounted value of expected future after-tax net earnings), in contrast to Friedman and Schwartz (1963), who reinterpreted the Great Depression as a Great Contraction of the money supply resulting from mistaken Federal Reserve policy. Tobin’s third central Keynesian proposition also undermined attempts (for instance by Minsky 1981 and Crotty 1990) to contrast an allegedly neoclassical Tobin’s q, supposedly based on a known probability distribution of underlying fundamental variables, with a more truly Keynesian approach that recognized fundamental uncertainty and exogenous shifts in long-period expectations.

The fourth central Keynesian proposition in Tobin (1977a), following Chapter 19 of The General Theory, held that “Even if money wages and prices were responsive to market excess demands and supplies, their flexibility would not necessarily stabilize monetary economies subject to demand and supply shocks.” This proposition, advanced vigorously by Tobin (1975, 1980a, 1992, 1993), placed the Keynesian challenge to what Keynes termed “classical economics” on a level of core theory. Keynesianism, as interpreted by Tobin, could not be dismissed as nothing more than the empirical observation (or arbitrary assumption) that money wage rates are sticky downwards. Even if prices and money wages responded promptly, the economy might fail to automatically readjust to potential output after a large negative demand shock and might require government intervention to restore full employment. Making money wages more flexible by eliminating trade unions, minimum wage laws, and unemployment compensation might just make things worse. Tobin’s fourth Keynesian proposition, and the emphasis on Chapter 19 as crucial to understanding the message of Keynes’s General Theory, were central to Tobin’s Keynesianism: involuntary unemployment might be a disequilibrium phenomenon, but the system might not have any mechanism to move it back to the full employment equilibrium after a sufficiently large negative demand shock.

Tobin stressed Keynes’s Chapter 19 on changes in money wages rather than Chapter 17 on the “peculiar properties of money”: its zero elasticity of production and zero elasticity of substitution (in contrast to Tobin’s view of money as an imperfect substitute for other assets, and of an endogenous money supply with a finite elasticity of supply). Tobin (1977a, 460) endorsed “Keynes’s challenge to accepted doctrine that market mechanisms are inherently self-correcting and stabilizing.” Unlike his first three central Keynesian propositions, Tobin did not claim empirical support for the fourth proposition: since money wages and prices did not in fact respond rapidly to excess demands and supplies, there could not be much direct evidence of what would happen in that counterfactual situation. The case for the fourth proposition had to be made, as in Tobin (1975), at a theoretical level. It was a case that he only made explicitly and formally from the 1970s onward, when Keynesianism was under challenge from natural rate of unemployment theories, first the monetarism of Friedman (1968), and then the New Classical economics of Lucas (1981a), which claimed the demand stimulus could increase employment and output only by tricking workers into accepting a lower real wage that they thought were getting. Unfortunately, Robert Lucas (1981b), in his review article about Tobin (1980a), did not engage with Tobin’s first lecture about disequilibrium dynamics, stability, and failure of self-adjustment, concentrating instead on protesting against the description in Tobin’s second lecture of Lucas’s New Classical approach as “Monetarism, Mark II,” merely Friedman’s natural rate of unemployment hypothesis and expectations-adjusted Phillips curve with rational expectations in place of adaptive expectations.

“Writing as J. M. Keynes,” Tobin (1997, 4) stated,

The central questions before economists of our generation are: “Does our market capitalist economy, left to itself, without government intervention, utilize fully its labor force and other productive resources? Does it systematically return, reasonably swiftly, to a full employment state whenever displaced from it?” The faith of the classical economists assures us “yes.” The answer of The General Theory is “no.” ... Fortunately, it appears that the remedies lie in government fiscal and monetary policies and leave intact the basic political, economic and social institutions of democracy and capitalism [contrary to the faith of the young Marxists who, to Keynes’s dismay, were prominent in the Cambridge Apostles in the 1930s].

Writing as himself, Tobin (1997, 27) concluded “Classical faith that demand-deficient economies will recover on their own failed theoretical and empirical challenge in Keynes’s day. It fails now again, more than half a century later.”

Microeconomic foundations for IS-LM

Tobin was present at the creation of Alvin Hansen’s one-good version of the IS-LM model of goods market and money market equilibrium that became the mainstay of American Keynesian teaching. Tobin, then a junior member of Harvard’s Society of Fellows, and Seymour Harris, as editor of the Economic Handbook Series, were the only people thanked in the preface to Hansen (1949, vi) for reading and commenting on the manuscript, and Hansen (1949, 168n), when citing Tobin (1947–48), declared “I have relied heavily upon his analysis.” Tobin (1947–48) had used the IS and LM curves (named for Investment, Saving, Liquidity Preference, and Money Supply), and the small system of simultaneous equations underlying them, to show that the preference of pioneer monetarist Clark Warburton (1945) for monetary policy rather than fiscal policy rested on an unstated assumption that the demand for money was insensitive to changes in the interest rate.

Post Keynesians rejected the IS-LM model as underplaying the importance of fundamental, uninsurable uncertainty (as distinct from insurable risk), and because Keynes would never have countenanced representing his theory by a system of simultaneous equations – although it turns out that a four-equation IS-LM model first appears in a lecture by Keynes in December 1933, attended by David Champernowne and Brian Reddaway, who later published the first models equivalent to IS-LM (Dimand 2007). Monetarists such as Milton Friedman also shunned the IS-LM diagram as being drawn for a given price level (e.g., the critiques of the “Yale school” by Brunner 1971 and Meltzer 1989), except when Friedman used it in Gordon (1974) in an attempt to communicate with his Keynesian critics – an instance later cited by some Post Keynesians as evidence that mainstream American Keynesian users of IS-LM were really classical rather than Keynesian. Tobin (1980a, Lecture I) responded to this monetarist objection to IS-LM by using IS-LM diagrams with the interest rate and price level on the axes to analyze situations of full employment, drawing the curves fo...

Table of contents

• Cover
• Title
• Introduction
• 1  An American Keynesian
• 2  Transforming the IS-LM Model Sector by Sector
• 3  Consumption, Rationing, and Tobit Estimation: Tobin as an Econometrician
• 4  Portfolio Balance, Money Demand, and Money Creation
• 5  Tobins q and the Theory of Investment
• 6  Money and Long-Run Economic Growth
• 7  To Improve the World: Limiting the Domain of Inequality
• 8  Taming Speculation: The Tobin Tax
• 9  Tobins Legacy and Modern Macroeconomics
• Notes
• Bibliography
• Index