Methodology and ideology

Does the realism and completeness of the assumption set matter to the validity of derived hypotheses?

We preface our analysis of The General Theory with comments on the roles played by methodology and ideology in Keynes's clash with classical theory

Methodology and ideology 163 and laissez-faire policy in The General Theory.

Our criticism of the accepted classical theory of economics has consisted not so much in finding logical flaws in its analysis as in pointing out that its tacit assumptions are seldom or never satisfied, with the result that it cannot solve the problems of the actual world.

(CW 7, p. 378)

In other words, Keynes insists, you cannot build a realistic theory of cap­italism based on crudely unrealistic assumptions. The realism and com­pleteness of the assumption set matters. Keynes's views on this issue are in sharp conflict with classical theory and with Milton Friedman's argu­ment that, in his favored methodology of positivism, the truth content of derived hypotheses is unrelated to the realism or completeness of the assumption set.¹ Friedman's positivism is the quasi-official methodology of both neoclassical and Modern Keynesian economic theory.

Classical theory has only one equilibrium state, one in which all markets clear. This equilibrium state is stable by assumption. If tempor­arily moved from equilibrium by some exogenous shock, market forces ensure a speedy return to a general equilibrium bliss point.

Classical theory had both historical and ideological roots. The histor­ical underpinning was the fact that the British economy grew rapidly and became economically and financially dominant in the global economy from the mid-eighteenth century through most of the nineteenth cen­tury, a period that Keynes often referred to as the "glorious nineteenth century." This understanding of capitalism hid from view the fact that Britain's high trend growth was caused by an inherently transitory set of conditions. These unique and transitory factors were thus not understood to be unique and transitory in the classical model used to explain Britain's impressive nineteenth-century long-term growth rate. They were impli­citly considered to be permanent conditions of modern laissez-faire capit­alism, at least in Britain.

This brings us to the ideological function of classical theory. It needed an assumption set capable of supporting in theory the ideological belief that unregulated capitalism was an ideal system - the Schumpeterian "pre- analytic vision" that lay behind the theory. The classical thesis that any significant attempt by the state to interfere with free markets, no matter how well intended, could not help but lead to inferior results was thus not based on careful observation of British economic and political history and an understanding of the transitory nature of its nineteenth-century growth path - it was an ideological presupposition built into its construction.

Keynes argued to the contrary that, in order to be useful, a theory must begin with a set of assumptions that realistically describes the historic­ally specific institutions and behaviors that constitute actually existing capitalism in any particular place and time. For Keynes, for example, nineteenth-century British capitalism and interwar British capitalism had different inherent tendencies that cannot be understood unless the assumption sets used to derive the theories adequately reflect the actual "facts" of the era. He argued that the assumptions used to create the "special case" of general equilibrium represented by the classical theory "happen not to be those of the economic society in which we actually live, with the result that its teaching is misleading and disastrous if we attempt to apply it to the facts of experience" (CW 7, p. 3).

Keynes stressed the ideological roots of classical theory and policy. At the end of chapter 3, he posed the question as to how it came to be that classical theory could dominate economics in Britain in the interwar era even though it had no convincing way to explain the sustained high unemployment of the era that was consistent with the facts and supported policies bound to reproduce this problem. In the quotation below, Ricardo represents the classical tradition.

Ricardo conquered England as completely as the Holy Inquisition conquered Spain...

The completeness of the Ricardian victory is something of a curiosity and a mystery. It must have been due to a complex of suitabilities in the doctrine to the environment into which it was projected. That it reached conclusions quite different from what the ordinary uninstructed person would expect, added, I suppose, to its intellectual prestige. That its teaching, translated into practice, was austere and unpalatable, lent it virtue. That it was adapted to carry a vast and consistent logical super­structure, gave it beauty. That it could explain much social injustice and apparent cruelty as an inevitable incident in the scheme of progress, and the attempt to change such things as likely on the whole to do more harm than good, commended it to authority.

(CW 7, pp. 32-33, emphasis added)

So, classical economists loved classical theory in part because of its aes­thetic qualities: "its vast and consistent logical superstructure gave it beauty." This elaborate superstructure also made it possible to pretend that economics was a real science, like physics or chemistry, rather than a soft social science, and it eventually gave to those best equipped mathem­atically to do advanced theory great prestige within the profession. Most importantly, it enabled conservative economists, who, in main, opposed progressive economic and social change, to claim that economic science supported their policies. And it gave the capitalist class, which, along with rentiers, controlled government economic policy or was "the dominant social force behind [political] authority," a pseudoscientific defense of their belief that the political process should reflect their worldview and ensure that nothing stood in the way of their untrammeled pursuit of wealth. Classical theory, he said, teaches that under laissez-faire capitalism, "all is for the best in the best of all possible worlds" (CW 7, p. 33).

What is general about The General Theory?

There are two important ways in which The General Theory is general. The first is that the book presents a very simple abstract model of the deter­mination of income and employment and then uses that model to create five more specific or applied models of different aspects or dimensions of economic activity in a capitalist economy. Mainstream Keynesians typ­ically assert that there is only one model in The General Theory - a short­term stable equilibrium model such as Hicks's IS/LM model - but this is not true. In Chapter 19 of this book, we show that there are five different applied models of Keynes's "general theory" in The General Theory.

The second way in which Keynes works from the abstract or general theoretical level to the concrete or applied level is through the specifica­tion of the behavioral relations or functions that constitute the applied models of the economy. In Keynes's methodology, the equations in the applied models must adequately reflect the concrete institutional, behav­ioral, and empirical "facts" of the historically specific form of capitalism under investigation. These change over time. If these differences are sig­nificant enough, they change the variables in and parameters of the behav­ioral equations of the applied models of the economy. This alters the static and/or dynamic properties of the model. The applied models therefore must faithfully reflect these differences in order to adequately represent the behavior of the specific national capitalisms under investigation. This makes it possible for the theorist to deduce effective policy regimes for these particular economies. The insistence that applied theory must be built on a realistic assumption set that reflects the important "facts" of the historically specific economy under investigation is a core component of Keynes's anti-positivist methodology.

We consider first the issue of moving from the abstract level to the applied level. Everyone who has taken a basic course in "Keynesian" macro theory knows that Keynes created a short-run model that demonstrated that equilibrium income and employment are strongly influenced by the strength of AD or aggregate spending (AS) and that high-unemployment equilibrium is a potential state of modern capitalist economies because AD can remain below the level of income consistent with full employment.³

The value of equilibrium output and income depend, in all of these applications of his general model, on the relationship between AD (or total spending) and AS (or the value of output and income). In the simplest abstract model, AD is the sum of capital investment and household con­sumption spending. In The General Theory, capital spending depends on the expected profit rate on investment, referred to as the "marginal efficiency of capital" (or mec), and the long-term interest rate (r). These variables are functions of expectations of future profit flows and future bond prices, respectively. (Endogenously generated expectations play a crucial role in all of Keynes's models.) If the expected profit rate exceeds the interest rate, investment projects should be undertaken, and vice versa. Household consumption spending is assumed to be a positive function of income. The "marginal propensity to consume" out of income (or mpc) measures the sensitivity of consumption spending to changes in income.⁴

Model 1 provides an example of both the concrete application of the "general" theory to Keynes's applied model of secular stagnation and of his insistence that the assumption set must incorporate the "facts" of spe­cific time and place. Chapters 13 and 14 of this book discuss Keynes's ana­lysis of long-run stagnation in both The General Theory and in an important lecture on long-term stagnation in the year following its publication. Model 1 is based on the same abstract-level variables used in his general model, but in The General Theory, he analyses and explains their behavior over the entire interwar period based on his understanding of the concrete histor­ical facts of the period. We presented his explanation of why he predicted post-WWI stagnation in his 1919 book The Economic Consequences of the Peace in Chapter 2 of this book, an explanation he augmented and updated in The General Theory.

Based on the institutional and behavioral "facts" of the interwar period through the mid-1930s, Keynes argued that both capital invest­ment spending and household consumption spending were likely to continue to remain too low to sustain full employment over the foresee­able future. Investment would continue to be constrained by a declining actual and expected rate profit rate (mec) and by a long-term interest rate that could not fall by enough to stimulate investment in the face of a declining mec. In his projection of the character of the post-WWI economy in The Economic Consequences of the Peace and in his explanation of secular stagnation between WWI and The General Theory, Keynes explained why the rate of profit on capital goods in Britain was higher in the "glorious" nineteenth century than it was in the 1920s and 1930s. The high profit rate and growth rate of the British economy in the nine­teenth century were caused by unique historical conditions that could not possibly be sustained forever, but classical theory was based on the implicit assumption that they were permanent conditions of modern capitalism.

His explanation of the decline of the rate of profit between the two periods stressed key changes in the historical "facts" from one era to the next. Key factors that helped drive nineteenth-century growth but were missing in the interwar period included: system-transforming technical change (such as the building of the national railroad system); the brutal creation, exploitation, and impoverishment of Britain's working class, especially prior to the mid-nineteenth century; imperialism with Britain at the center of the world's trading and financial systems; successful wars; rapid population growth; and a globally dominant British cotton industry dependent on slave labor and military force.⁵ As noted in Chapter 2, the British historian Eric Hobsbawm summed up this situation as follows:

There was a moment in the world's history when Britain can be described, if we are not too pedantic, as its only workshop, its only massive importer and exporter, its only carrier, its only imperialist, almost its only foreign investor; and for that reason its only naval power and the only one that had a genuine world policy.

(Hobsbawm 1969, p. 13)

By the interwar period, all of these conditions had vanished.

Keynes also argued that the long-term rate of interest could not decline by as much as the mec in the interwar years for institutional and behav­ioral reasons that he discussed in various places in The General Theory.⁶ It is important to understand that, for Keynes, the interest rate at the abstract level of the model represents the conditions of financial markets at the con­crete institutional level. International and many national financial markets changed substantially between the two eras, a fact to which we shortly return. Consumption spending would continue to be held down in Britain and the USA in the current era, he said, by a high degree of income and wealth inequality that caused the share of income spent by households on consumption goods (or mpc) to be low. Keynes thus argued that both the level of investment and the value of the investment "multiplier" were too low to generate a long-run rate of growth of AD sufficient to eliminate high unemployment.

He concluded that the facts of the interwar years made it necessary to replace the laissez-faire capitalism of nineteenth-century Britain with Liberal Socialism in order to achieve sustained full employment in Britain.

Consider a second example. As explained in chapter 22 of The General Theory and Chapter 18 of this book, Keynes explained why business cycles of this era in the USA were unusually volatile. Model 3 incorporates the endogenously created dynamic interaction of the real sector and financial markets in driving booms far beyond sustainability while creating finan­cial commitments that cannot be fulfilled when the booming economy eventually hits a "crisis." Minsky called this a condition of "financial fra­gility" in which the shift from boom to bust in the context of overstressed balance sheets in both real and financial sectors sharply exacerbates the rate and depth of the collapse.

Keynes's theory of financial investors and financial markets in model 3 (and in model 5) is fundamentally different from the theories of agents and agent choice of classical theory and Mainstream Keynesian theory because they are based on a realistic set of assumptions about institutions and agent behavior that lead to different specifications of the equations in the applied models. Keynes built his theory of agent choice on the core assumption that future states of the economy, including future financial asset prices, are unknowable or fundamentally uncertain in the present moment. This means that Keynes's financial market investors cannot be the fully informed, optimizing mathematicians of classical and Modern Keynesian theory. His agents have to construct predictions of future finan­cial asset prices through behavioral and psychological conventions or heuristics and determine how much "confidence" they have in the truth content of their forecasts before they can decide on an investment strategy. Since Keynes argued that because both asset-price expectations and con­fidence in those expectations are endogenous and pro-cyclical, financial asset prices are inherently subject to bouts of instability. But he also insisted that the degree of instability in any institutionally specific financial market depended heavily on the "facts" that characterized that market. In times of light market regulation, heavily debt-financed asset purchases, fragile balance sheets, and exceptionally "liquid" markets (in which transactions can be made quickly and cheaply), financial markets can create and repro­duce spectacular speculative booms and busts - as in the late 1920s and early 1930s in the USA. Keynes observed that "When the [real] capital development of a country becomes a by-product of activities of a casino, the job is likely to be ill-done" (CW 7, p. 159).

Keynes stressed that the kind of "insane" financial gambling in the USA in this period was not an inherent characteristic of all national capitalisms in all eras. He stated that one of the reasons why the London stock market did not suffer from the financial instability or mania observed in the USA in this period was because "compared to Wall Street" it was "inaccessible and very expensive" to buy and sell securities. It was, in other words, much more illiquid (CW 7, p. 159).⁷ "The jobber's 'turn' [or market-makers fee], the high brokerage charges, and transfer [or turnover] tax... suffi­ciently diminish the liquidity of the market. to rule out a large propor­tion of the transactions characteristic of Wall Street" (CW 7, pp. 159-160). The different institutional and behavioral characteristics of British and US financial markets in the interwar period led Keynes to use different specifications of applied models of stock and bond market performance in Britain and the USA in this period. The differences between the two financial markets help explain why Britain did not experience a tsunami of bank failures as in the USA. Indeed, it never experienced a financial crisis in the 1930s.

Neither the classical theory nor the Mainstream Keynesian theory of financial markets follows Keynes's methodology in this regard. Both offer general or very abstract theories in which financial markets are always well behaved. They offer a theory of the bond market in which agents know the probability distributions of the future interest rates; in which interest rates always fall when AD receives a negative shock; in which "insane" financial market gambling casinos cannot exist and therefore financial instability and financial crises cannot exist; and in which there is one model of interest rate determination that applies to all economies that are sufficiently capitalist. That is why classical theory was incapable of predicting or explaining the US financial market chaos of the late 1920s and early 1930s and why modern mainstream financial market theory did not predict and cannot explain ex post the global financial crisis that began in 2007. Keynes's applied theory of financial markets was designed to explain their behavior in both relatively stable times and in periods of great market instability.

In his writings on Britain's economic problems in the interwar period, Keynes reported the results of his extensive research on the concrete his­torical, institutional, and empirical analyses - including at the firm and industry levels - needed to understand why the economy performed so badly in the interwar years. Mainstream macroeconomists rarely do this because it does not fit their deductive positivist methodology. Their goal is to demonstrate either that capitalism has a unique and stable equilib­rium at full employment (true of monetarism, New Classical theory, real business cycle theory, and dynamic stochastic general equilibrium theory)⁸ or that a capitalist economy always has the basic institutions, practices, and incentives to achieve and sustain full employment as long as "Keynesian" countercyclical macro policy keeps the economy on its efficient long-term growth path (Mainstream Keynesian theory). To accomplish their object­ives, mainstream economists do not need to delve too deeply into the specifics of the sub-macro levels of the economy.⁹