Uncertainty, International Money, Employment and Theory

When discussing the implications of uncertainty on economic activity and its policy implications, the fundamental differences among various economic theories involve (a) the analyst’s conception of the external economic reality in which decision-makers operate, and (b) the ability of agents to understand that reality.

There are two different concepts of uncertainty in economics: the classical concept and the Keynes concept. The ability of economists to explain the importance of money, liquidity and the existence of persistent unemployment in a market economy depends on which concept of uncertainty the analyst uses.

‘In economics you cannot convict your opponent of error; you can only convince him of it. And, even if you are right, you cannot convince him if there is a defect in your own powers of persuasion and exposition or if his head is already so filled with contrary notions that he cannot catch the clues to your thought which you are trying to throw to him.’1

Much of the Post Keynesian literature of the last two decades, including some that I have contributed to, is devoted to demonstrating the analytical shortcomings of orthodox neoclassical theory. This negative literature has played an important role in making even orthodox economists aware of the severe limitations of neoclassical theory and its resultant policy pronouncements. Some neoclassical scholars have grudgingly admitted the value of this negative Post Keynesian literature, while stubbornly refusing to admit that Post Keynesian theory has made any positive contributions to ‘Theory’. For example, James Tobin has written:

A school of self-styled post-Keynesians regard any synthesis or reconciliation, in substance or language, of Keynes and neoclassical economics, as a betrayal of the revolution. They reject equilibrium analysis altogether, stress the historical, institutional, and evolutionary aspects of economic development, and emphasize the macroeconomic implications of the noncompetitive structure of modern economies.

Their valid points do not add up to a coherent theory

. (Tobin, 1985, p. 115, emphasis added)

The current international payments system does not serve the emerging global economy well. The Financial Times and The Economist, both previously strong advocates of the existing floating rate system, have acknowledged that the system is a failure and was sold to the public and the politicians under false advertising claims.1 Can we not do better?

This paper explains why, in an open economy context, Keynes-type demand management policies appear to have lost their ability to create jobs domestically. Instead, stimulatory policies tend to create an import deficit (as jobs are created abroad) and the fear of inflation domestically. The paper then suggests remedies that will resurrect Keynes-type demand management policies to stimulate expansion without experiencing balance of payment problems or domestic inflation.

Keynes’s (1936) General Theory of Employment Interest and Money1 is developed primarily in a closed economy context. Keynes did, however, introduce an open economy analysis when he noted that: (a)trade could modify the magnitude of the domestic employment multiplier (p. 120);(b)reductions in money wages would worsen the terms of trade and therefore reduce real income, while it could improve the balance of trade (p. 263); and(c)stimulating either domestic investment or foreign investment can increase domestic employment growth (p. 335)

In this journal’s ‘Policy Forum’, Eichengreen, Tobin and Wyplosz (1995, p. 164) (hereafter ETW) argue that volatility in foreign exchange markets due to speculation can have ‘real economic consequences devastating for particular sectors and whole economies’. To constrain ‘speculative behaviour … they (ETW) propose a global transaction tax … to discourage short-term round tripping’ (Greenaway, 1995, p. 160). At the same time that this ETW proposal appeared in print, the winter 1994–5 Mexican peso crisis spilled over into the dollar problem. In international financial markets, where image is often more important than reality, the dollar was dragged down by the peso while the German mark and Japanese yen appeared to be the only safe harbours for portfolio fund managers.

The world unemployment situation is very grim. In most industrialized countries unemployment remains stubbornly high. The average unemployment rate in the EU was 11.3 per cent in July 1996, and, with the exception of the United States, unemployment rates in OECD nations have reached or remained close to historical highs not seen since World War II. The 1997 Asian financial collapse implies higher unemployment rates for Europe, Japan and the United States — and perhaps even Latin America in 1998. Does this dire trend mean that with the globalization of the market system, there is little or no hope of restoring full employment in most economies around the world?

How one interprets financial market activity and chooses a policy stance regarding the regulation of such markets depends on the underlying economic theory that one explicitly, or implicitly, utilizes to explain the role of financial markets in an entrepreneurial economy. There are two major alternative theories of financial markets: (1) the Classical efficient market theory (hereafter EMT) and (2) Keynes’s liquidity preference theory (hereafter LPT). Each theory produces a different set of policy prescriptions.

Economists have probably spilled more printer’s ink over the topic of money than any other. Moreover, while monetary theory impinges on almost every other conceivable branch of economic analysis, confusion over the meaning and nature of money continues to plague not only the economics profession but also central bankers and other policy-makers. The resulting semantic obfuscation results in needless debate over the proper monetary policy to produce a prosperous economy. Accordingly, the purpose of this article is to develop a precise definition of money to help economists provide a scientific analysis of the role of money. Only after that is achieved can we develop a scientific approach to the question of monetary policy.

Professor Hahn, a distinguished theorist, has recently written: ‘The most serious challenge that the existence of money poses to the theorist is this: the best developed model of the economy cannot find room for it’ (Hahn, 1981, p. 1).

Since Aristotle’s time, as Schumpeter (1954, p. 288) noted, there have been two views regarding the nature of money in a money-using market-oriented entrepreneurial economy: the Chartalist view and the Monetarist (or Metallist) view. The latter assumes that the institution of money evolved from barter relationships where producible goods always traded for other producible goods. Ultimately, in this Monetarist view one commodity (normally a metal) became the embodiment of the other goods traded for any specific good. Thus, behind the Monetarist conceptualization of money is the idea that money merely represents the trading of one specific producible commodity for all other producible commodities. From this there develops the idea that the use of such a (producible) commodity-based money is more efficient than direct barter for it avoids what Clower called ‘the double coincidence of wants’.

Alfred Eichner died in February 1987. His greatest legacy to the economics profession therefore must be his students rather than his ideas. Because he died before his full potential was developed, it is my view that the concepts and approaches to economics that he was struggling with had not yet been fully fleshed out. I know that Alfred still had an open mind on many issues and that he was still willing to learn and eagerly absorbing new material. Accordingly almost all that he wrote was still in the formative stage. And nowhere is this more true than in his provisional text entitled The Macrodynamics of Advanced Market Economies (1987).

In his first news conference after the [1992] election (on 12 November), President-elect Clinton announced that his highest priority will be to enact an economic recovery bill to stimulate job growth in the short run and deficit reduction in the long run. His short-term stimulus plan involves (a) an investment tax credit, (b) accelerating expenditures for highway construction, and (c) additional legislation to expand spending on infrastructure.

In March 1992, 1 was one of 100 economists (including 6 Nobel Prize winners) who sent an open letter to President Bush and members of Congress urging immediate action to solve our short-term and longer-term depressing unemployment problem. The 100 economists recommended (a) spending an additional $50 billion per year for the next five years for infrastructure, (b) an investment tax credit targeted only for additional investment projects undertaken, and (c) not to worry about federal deficit reduction for at least three to four years when the economy will be growing sufficiently so that the National Debt as a proportion of Gross Domestic Product will fall. Taxes should not be increased nor government spending cut merely to lower the current government deficit in the next two to three years.

It was reported in a recent New York Times article that a sheetmetal worker at the TWA facility in Kansas City who lost his job during the 1991 recession was asked if he was happy that the Clinton Administration had delivered on its election promise to create jobs. (Over three million new jobs have been created since January 1993.) The man’s response was ‘Yeah! My wife and I have four of them’. The Times then noted that in 1993 this man’s family income was less than what he had earned when he worked for TWA despite the fact that he now worked full-time both as a school bus driver and for McDonald’s, while his wife restocked shelves in Toys ‘R Us as well as working in the same McDonald’s.

The purpose of this paper is quite modest. The theoretical argument starts by accepting Keynes’s fundamental axiom that money is never neutral in an entrepreneurial economy.1 It is then possible to trace out why, if all prices are flexible, falling prices and especially an asset price deflation can have a devastating impact on the financial system and permanently lower employment and real economic growth. (If, on the other hand, money was neutral, then falling prices would be the process through which the economic system either maintains or restores full employment prosperity.)

I tell my students that the best way to evaluate any logical model is to consider the model builder as if she is a magician. Model-builders rarely make logical errors in moving from axioms to conclusions, any more than professional prestidigitators drop the deck while performing a card trick. Model-builders excel at creating the illusion of pulling relevant policy conclusion rabbits out of their black hat model. The more surprising the rabbits pulled from the hat, the greater the audience enjoyment and applause.

Keynes was primarily a monetary theorist. The words money, currency or monetary appear in the titles of most of his major volumes in economics. Post Keynesian monetary theory evolves from Keynes’s revolutionary approach to analyzing a money-using economy.

New Keynesians Joseph Stiglitz (1989) and Lawrence Summers (Summers and Summers 1989), following the lead of Old Keynesian James Tobin (1974), have argued that an ad valorem tax on financial market transactions is socially desirable in that it will reduce the observed volatility in our ‘super-efficient financial markets’. All of these Keynesians claim that Keynes initiated the recommendation for a universal financial transactions tax as a socially desirable policy.

Politicians of both major parties argue that they want to provide a legislative economic programme to promote a ‘civil society’ that emphasizes the importance of family and community values. Unfortunately, today’s political decision-makers are trapped by an economic policy rhetoric that focuses so heavily on market values and individual self-interest that it derogates the very family and community values that politicians want to promote.

I welcome Rod Cross’s comments suggesting that the New Keynesian conceptualization of ‘hysteresis’ is ‘amenable’ and ‘relevant to the Post Keynesian understanding of economic processes’, where economic events occur in a historical context with an uncertain future. More than one-half century after Keynes’ criticism of ‘Professor Tinbergen’s Method’ (1937) of analysing time series econometrically under the assumption of stationarity, it is gratifying to note that some macroeconometricians are trying to develop an analysis that deals with changing macrofactors that both affect, and are affected by, agents’ behaviour.

The principle of a ‘truth in labelling’ law that protects consumers1 from false and misleading claims is often violated by economics textbooks. Under the truth in labelling law, a minimum quantity of beef is required in a patty before society permits anyone to sell it as a hamburger. Similarly, some minimum quantity of Keynes’s logical analysis should be an essential ingredient in any theory being sold as Keynesian, especially in textbooks to yet uneducated consumers. Paraphrasing a famous slogan of the 1988 Democratic presidential primary, ‘Where’s the Keynesian beef in New Keynesian economics?’.

Tom Asimakopulos was a great Kaleckian scholar. His knowledge of the nuances and innuendoes of Kalecki’s approach to macroeconomics was impressive and usually definitive. In his last book, Keynes’s General Theory and Accumulation (1991), completed only months before his untimely and tragic death, Asimakopulos records his lectures to final-year honours economics students and to graduate students’ (p. xv). The first six chapters of this volume present Asimakopulos’s interpretation of Keynes’s General Theory. For this reader, these pages displayed a masterful Kaleckian interpretation of what should be the theory of employment equilibrium analysis. Chapters 7 and 8 discuss the theories of accumulation developed by Roy Harrod and Joan Robinson. Chapter 9 provides a 4½ page summary of the conclusions of this work. I shall limit my discussion to the short-period analysis of the first six chapters. In these chapters Asimakopulos makes the following points: 1.Keynes’s General Theory was a ‘significant departure from his earlier theoretical writings’ (p. 11);2.The concept of money is linked with the ubiquitous uncertainty of future economic events (p. 8);3.Involuntary unemployment can be discussed via a five-equation model where the net marginal product of labour curve is the ‘demand curve for labour’ (pp. 32–5) and Keynes’ aggregate supply function can be derived from these relations (pp. 53–7);4.Keynes’s aggregate supply and demand analysis is ‘less soundly based than Kalecki’s’ (p. 79) and Keynes relied on Richard Kahn to correct the many analytical errors committed by Keynes (p. 12);5.Keynes’s analysis of investment is found wanting for it ‘involves more monetary and financial elements than are found in the formal model’ (p. 79);6.Keynes’s formal model ‘is displayed’ in a set of equations that depict a situation of short-period equilibrium’ (p. 101). This equation-set combines the labour market equations with a traditional consumption function, an investment function, a demand for money function and an exogenous money supply function (p. 102). This ‘formal model’ bears an amazing resemblance to the neoclassical synthesis models of Hicks and Modigliani. What is so surprising in all this is that Asimakopulos did not explicitly indicate that the formalization that he presented is ‘Bastard Keynesianism’ to use Joan Robinson’s splendiferous phrase, rather than the economics of Keynes.7.Tn working out his [Keynes’s] model, the money wage rate was taken as exogenous’ (p. 29). Keynes had perceptive observations regarding the effects of money wage rate changes but these ‘observations were not part of his formal model … There is no suitable general functional form relating changes in money-wage rates to changes in the level of employment’ (p. 30).8.Keynes failed to understand the multiplier as a process over time. This leads to ‘inconsistencies’ in Keynes’s handling of the saving-investment relationship and the role of finance (p. 113).

‘In economics you cannot convict your opponent of error — you can only convince him of it. And, even if you are right, you cannot convince him … if his head is already so filled with contrary notions that he cannot catch the clues to your thought which you trying to throw to him’. These words were written by Keynes in an early draft of his General Theory. They correctly describe the position of Post Keynesians in responding to the criticisms that Mongiovi has developed from his unquestioning faith in classical theory. Mongiovi’s head is already so filled with Sraffian-neoRicardian classical notions that he has failed to grasp any of clues embedded in the explicit writings of Keynes in The General Theory or in other Post Keynesian writings including those listed in Mongiovi’s references.

Sixty years ago, an economic classic was published: John Maynard Keynes’s book The General Theory of Employment, Interest and Money (1936). A classic can be defined as something everybody cites and nobody reads. In the spirit of this definition, it appears that Keynes’s book is a classic, at least as far as New Keynesians are concerned.

Geoff Harcourt has been an amazingly prolific economist. His interest is not theory for theory’s sake — but theory as a means of understanding the economic world in which we live. Some of his greatest contributions have been in synthesizing debates in economic theory and offering the fair-minded reader a vision of not only how to resolve the controversy, but also to use the resolution to gain insights about the real world.

On p. ix of the Preface to the printed German language edition of Keynes’s The General Theory of Employment Interest and Money, published by Duncker and Humblot in 1936, the following sentences appear: ‘This is one of the reasons which justify my calling my theory a General [emphasis in the original] Theory. Since it is based on fewer restrictive assumptions [‘weniger enge Voraussetzungen stutz’] than the orthodox theory, it is also more easily adopted to a large area of different circumstances’.1

The Post Keynesian (hereafter PK) theory of employment is derived from Keynes’s chapter 3 (1936a) ‘The Principle of Effective Demand’ where labour-hire demand is derived from an effective demand point determined in the product markets.1 Tobin (1992, p. 392) states this chapter is ‘the most important innovation of the General Theory’. The effective demand principle means that there is an ‘employment function’ (Keynes, 1936a, chapter 20) relating alternate labour-hire decisions to alternate points of effective demand. There is no aggregate demand for labour schedule with the real wage as the independent variable (Davidson, 1983).

Prior to 1958, most economists and lay-persons believed that there was a direct relationship between price level changes and output growth, so that the price level generally increased when demand for products was growing faster than the economy could supply them. Inflation was often characterized as being caused by too much money chasing too few goods. In 1958, however, the price level in the United States rose significantly while at the same time employment and gross domestic output fell. These facts did not match the conventional wisdom regarding inflation.