My own understanding of the operations of the invisible hand, derived from the classicals, is different. The capitalist economic system generates powerful ordered patterns that transcend historical and regional particularities. The forces that shape these patterns are neither steely rails nor mere constellations of circumstance. They are, rather, moving limits whose gradients define what is easy and what is difficult at any moment of time. In this way they channel the temporal paths of key economic variables. Indeed, these shaping forces are themselves the results of certain immanent imperatives, such as “gain-seeking behavior,” that dominate the capitalist social form in all of its historical expressions. Agency and law coexist within a multidimensional structure of influences. But this structure is itself deeply hierarchical, with some forces (such as the profit motive) dominating others.
From this point of view, systemic patterns are generated in and through continual fluctuations: disorder is the operative mechanism of order. To attempt to theoretically separate order from disorder, or even to merely emphasize one over the other, is to lose sight of their intrinsic unity, and hence of the very factors that endow the system with its deep patterns. In this sense, order is not synonymous with optimality, nor is disorder synonymous with chaos. Order in-and-through-disorder is a brute force that tramples both expectations and preferences. This is precisely the source of the system’s vigor, whether or not one likes the outcome.
It is, of course, necessary to identify particular mechanisms through which order and disorder operate in given circumstances. The great virtue of the classical approach, in my opinion, is that it is able to derive a large variety of phenomena from a very small set of coherent operative principles that give rise to forces which make actual outcomes gravitate around their ever-moving centers of gravity. This is the system’s mode of turbulent regulation, whose characteristic expression is the pattern recurrence. The theoretical and empirical applications of these two notions are woven into the structure of my work.
Turbulent regulation and recurrence apply to the system’s various gravitational tendencies. Of these, the first set consists of those that channel the actual movements of commodity prices, profit rates, wage rates, interest rates, equity prices, and exchange rates. Equalizing tendencies driven by the restless search for monetary advantage reduce the very differentials that motivate them while at the same time giving rise to new differences. For example, equalization processes make individual wage and profit rates gravitate around the corresponding averages, which are themselves affected by the processes and by other factors. These are familiar notions in economics, but the point here is that there is a big difference between gravitation around an ever-moving balance point and equilibrium-as-a-state-of-rest. To study the properties of balance points, as the classicals do with natural prices or Marx does with balanced reproduction, is not to assume that these points exist as such. On the contrary, the relevant variables are generally away from this point and hit it only as they pass through from one side to the other. Among other things, this implies that one cannot assume that agents make their decisions as if they are in equilibrium.
The principle of turbulent regulation has its roots in the method of Smith, Ricardo, and Marx, for whom economic “laws” are dominant regulative principles that exert themselves in and through various countertendencies. The theory of real competition has similar roots in the economics canon, but also in the work of P. W. S. Andrews and Roy Harrod, two prominent members of the Oxford Economic Research Group. Elements can also be found in the business literature, most notably in the work of Michael Porter. A characteristic feature of this vision is that competitive firms necessarily engage in price-cutting and cost-cutting behavior, that technical and labor conditions vary across firms, and that only the firms with the best generally available conditions of production (best practice) have their profit rates equalized with those of similar firms in other industries. The resulting patterns closely resemble those found pointed out in business studies and in the literature on imperfect competition. Yet they represent the outcomes of price and profit rate equalizing competition, not “imperfect” competition. I have spent a great deal of time over the years developing and testing the theory of real competition.
The second set of gravitational tendencies arises from the system’s turbulent macrodynamics. This gives rise to its characteristic expansionary processes, with its waves of growth and slowdown, persistent unemployment, and periodic bouts of inflation. Once again, it is the profit motive that is the dominant factor in the regulation of production, investment, economic growth, employment, business cycles, and inflation. The emphasis on growth also has roots in the classical tradition, as well as in the works of Harrod and Joan Robinson. The latter two share an emphasis on growth as the normal state but disagree on its determinants. For a given technology, Harrod believes that growth is driven by exogenously given savings rates,² while Robinson (like Keynes) argues that it is the profitability of investment that drives growth,³ In the Pasinetti-Kaldor extension of Harrod, profitability adapts to growth, while in Robinson’s argument, it is the other way around. I take the classical-Keynesian-Robinsonian path here. But because the classical starting point accords a central role to production, the end result is characteristically different. Supply is neither the imperial force of neoclassical economics nor the ghostly presence of Keynesian and Kaleckian economics. Supply and demand are coequals here, and as always, profit is pulling the strings.
I have focused so far on my life experiences as they have influenced my vision of economics. In what follows, I would like to illustrate some of the applications of my general approach.
At a methodological level, I have focused on the fact that gravitation of actual outcomes around their balance points (fundamentals) is generally mediated by expectations. In my earlier work, I focused on nonlinear dynamics as a means of formalizing such processes, inspired by earlier work by Kaldor and Goodwin. More recently, I realized that George Soros’s theory of reflectivity, which emerges from his considerable experience in the world of finance, provides a more general framework for the interactions of these three variables. Soros advances three general theses: expectations affect actual outcomes, actual outcomes can affect fundamentals, and expectations are in turn influenced by the discrepancies between outcomes and fundamentals. The end result is a process in which actual variables oscillate turbulently around their gravitational values. Expectations can induce extended disequilibrium cycles in which a boom eventually gives way to a bust (Soros 2009: 50–75, 105–106). Because expectations can affect fundamentals, the gravitational centers are path dependent. Hence, the future is not a stochastic reflection of the past, so that the overall system is nonergodic (Davidson 1991). The existence of extended disequilibrium processes invalidates the efficient market hypothesis, and the dependence of fundamentals on actual outcomes invalidates the notion of rational expectations. Last, it is important to recognize that although expectations can influence actual outcomes, they cannot simply create a reality that validates them. On the contrary, gravitational centers continue to act as regulators of actual outcomes, which is precisely why booms eventually give way to busts (Soros 2009: 40–44, 50–58, 75, 216–222). Such patterns are consistent with the empirical evidence and with classical ideas on turbulent equilibration, but they invalidate notions such as rational expectations and the efficient market hypothesis. By tracing the elements of Soros’s theory, I showed that it can be formalized in a simple and general manner that gives rise to testable propositions (Shaikh 2010).
Profit rate equalization is a central concept in all theories of competition. In the classical view of competition, profit rate equalization is conceived as a dynamic and turbulent process involving ceaseless fluctuations around a moving center of gravity. New conditions of production are constantly entering the battle of competition as older ones fall away. This perpetual fray gives rise to profit rates that generally differ across methods and firms. I have long argued that what is relevant to competition is the profit rate on the best-practice conditions of production, because their profitability is the relevant gauge for new investment. This led me to develop an approximation to the profit rate on recent investment in the form of an incremental rate of profit, defined as the ratio of the change in gross profits to the previous period’s gross investment (Shaikh 1998). In a subsequent paper I examined average and incremental rates of profit from 1970 to 1990 in eight manufacturing subsectors, each aggregated across eight major OECD countries; in subsectors of US manufacturing from 1979 to 1990; and in thirty US industries from 1987 to 2005. Average rates of profit were found to cluster around a common mean, but many remained persistently above or below that benchmark. By contrast, incremental rates of profit consistently moved back and forth across their common mean, as would be expected from the classical theory of the turbulent equalization of actual profit rates (Shaikh 2008). It should be said that an incremental rate of return, with its erratic path and boisterous interactions, is very different from the genteel marginal rate portrayed in standard theory.
I applied the same approach to the financial market. All theories of competition expect that rates of return are equalized between sectors, for instance between the corporate sector and the stock market. Orthodox economics builds the expectation of exact equalization into its theory of stock prices, through various versions of the discounted cash flow model. Yet this model performs so badly at an empirical level that economists such as the Nobel Laureate Robert Shiller have concluded that financial markets are driven largely by irrational expectations, fads, and fancies – not just in periods of bubbles, but in general. Shiller shows that the average rate of return in the equity market and the average profit rate in the corporate sector differ considerably in their levels, volatilities, and trends. This is the empirical foundation for his well-known thesis that equity markets exhibit “irrational exuberance” (Shiller 1989, 2001). I argued on theoretical grounds that competition serves to equalize returns on new investment, not average rates of return. This is significant, because industrial capital stocks are of varying efficiencies and vintages, whereas all equities in a given corporation are the same regardless of the date of their issue. My own calculations using Shiller’s data (which he has generously made publically available) showed that the incremental rates of return in the stock market and the corporate sector are extremely similar: both are highly turbulent, yet they have virtually the same mean and variance, and even move together most of the time except for specific and limited bubble periods, as in the 1990s. From a classical perspective, the combination of the incremental rate of return and Soros’s concept of reflexivity with its notion of a moving center of gravity provides a far better explanation of stock prices than does Shiller’s “irrational exuberance” (Shaikh 1998, 2010).
Another strand of my work involves the explanation of growth. Along Harrod’s warranted path, long-term growth is driven by the savings rate (thrift). But in Solow’s influential growth model, thrift has no effect on the long-run rate of growth, which is instead driven largely by exogenous technical change. The endogenous growth theory of Frankel and Romer therefore makes technical change internal to accumulation in a very particular manner, precisely in order to reinstate thrift as the driver of capital growth. I observe that there is a striking discrepancy between micro- and macroeconomic reasoning in all three approaches. All sides agree on the notion that individual investment (capital expansion) is driven by its expected profitability, and yet all sides conclude that in the long run aggregate capital expansion is driven by something totally different. I argue that this break is neither theoretically necessary nor empirically plausible. The key to reconciling the microeconomic understanding and the macroeconomic results lies in recognizing that the business savings (retained earnings) are crucially linked to business investment: both are internal to any given firm, so they cannot be taken as independent. In a path-dependent nonergodic world, it is not possible to link the two through the usual calculus of all-seeing optimization. I show that that it is mathematically sufficient if the business savings rate responds (in any degree) to the gap between total savings and investment. This makes the overall savings rate endogenous, and then it is possible to reconcile profit-driven growth as in Keynes and post-Keynesian economics with roughly normal levels of long-run capacity utilization as in Harrod (Shaikh 2009). The further implications for the analysis of multiplier effects, particularly for those arising from deficit spending by the state, is addressed in my aforementioned book.
I have also developed an approach to inflation that derives from the classical link between growth and profitability. Marx, Leontief, and von Neumann established that the profit rate provides the upper limit to the sustainable growth rate of the economy even when there are no input (including labor) constraints. From this perspective, I argue that the ratio of the growth rate to the profit rate provides an index of the degree of utilization of an economy’s growth potential. In contrast to conventional measures of unemployment or capacity utilization, my measure of the utilization of growth potential works quite well in explaining actual episodes of inflation many countries, including the infamous “stagflation” of the 1970s and 1980s in the developed world (Shaikh 1999; 2016, Ch. 10).
Finally, and perhaps most controversially, I have long argued that the theory of comparative costs is fundamentally incorrect on both theoretical and empirical grounds. The theory of international trade is actually a subset of the general theory of competition. In a capitalist world, free international trade is conducted by businesses. Domestic exporters sell to foreign importers, who in turn sell to their residents, while domestic importers buy from foreign exporters and sell to us. At each step in the chain, it is profit that motivates the business decision. Comparative cost theory rests on the proposition that a trade surplus will drive up the real price of the country’s currency, which in turn will reduce the surplus, until at some point both the balance of trade and the balance of payments are automatically reduced to zero. A trade deficit would have the opposite initial effect, leading to the same conclusion. In Ricardo’s original derivation, the nominal exchange rate is fixed, so imbalances generate money inflows and outflows that raise or lower national price levels, thereby moving the real exchange rate in the opposite direction – until the trade balance is zero and the terms of trade lie between comparative cost limits. In the case of flexible exchange rates, the money flows move the nominal exchange rate to the same ultimate point. In either case, it is the real exchange rate that adjusts automatically. Both Marx and Harrod make a compelling counterargument: money inflows increase liquidity and lower interest rates, while money outflows have the opposite effects. Neither of these substantially alters the trade balance. Instead, they induce short-term capital flows, which bring overall payments into balance by covering the persistent trade deficits (Harrod 1957: 90–96, 112–116, 130–138). My own extension has been to show both theoretically and empirically that international terms of trade are, in the end, relative prices regulated by relative real costs. Thus, international competition operates in much the same way as national competition, rewarding cost advantages and punishing cost disadvantages (Shaikh 2007; Shaikh and Antonopoulos 2012).
My work has generally focused on understanding and explaining fundamental patterns in the developed world. This is not due to a lack of interest in economic policy or in economic development. On the former front, I worked for several years with Wynne Godley, Dimitri Papadimitriou and Gennaro Zezza on the macroeconomic model of the Levy Institute of Bard College, helping put out a biannual macroeconomic report on the patterns and prospects of the US economy. On the latter front, I have always believed that an analysis of the developed world is an essential foundation for an adequate understanding of economic policy and economic development. Finally, I have always believed that economics must be a moral science. In the midst of a global great depression, the International Labor Organization reports that income inequality has actually worsened, that in 2013 there were more than 900 million working people in the world living below the US$2 poverty line, that there were 1.52 billion workers in vulnerable employment, and that young people are nearly three times as likely to be unemployed as are adults. Moral and ethical differences affect the goals to which we subscribe, and theoretical differences affect the prescriptions we offer. One important task is to make these differences explicit and to confront their implications. There is no such thing as a value-free or socially neutral economics.
Becker, Gary S. 1962. “Irrational Behavior and Economic Theory.” Journal of Political Economy, 70, pp. 1‐13.
Davidson, Paul. 1991. “Is Probability Theory Relevant for Uncertainty? A Post Keynesian Perspective.” The Journal of Economic Perspectives, 5(1), pp. 129‐143.
Harrod, Roy F. 1957. International Economics. Chicago: University of Chicago Press. Laughlin, Robert B. 2005. A Different Universe: Reinventing Physics from the Bottom Down. New York:Basic Books.
Sardoni, Claudio. 1987. Marx and Keynes on Economic Recession: The Theory of Unemployment and Effective Demand. New York: New York University Press.
Shaikh, Anwar. 1974. “Laws of Production and Laws of Algebra:The Humbug Production Function.” The Review of Economics and Statistics, LXI(1), pp. 115‐120.
____. 1998. “The Stock Market and the Corporate Sector: A Profit‐Based Approach,” InFetschrift for Geoffrey Harcourt, ed. M. Sawyer, P. Arestis and G. Palma, 389‐404. London: Routledge & Kegan Paul.
____. 1999. “Explaining Inflation and Unemployment: An Alternative to Neoliberal Economic Theory ” In Contemporary Economic Theory, ed. A. Vachlou, 89‐105. London: MacMillan.
____. 2007. “Globalization and the myth of free trade,” In Globalization and the Myth of Free Trade, ed. A. Shaikh, 50‐68. London: Routledge.
____. 2008. “Competition and Industrial Rates of Return,” In Issues in Economic Development and Globalisation, Festschrift in Honor of Ajit Singh, ed. P. Arestis and J. Eatwell, 167‐194. Houndmills, Basingstoke, Hampshire: Palgrave MacMillan.
____. 2009. “Economic Policy in a Growth Context: A Classical Synthesis of Keynes and Harrod.” Metroeconomica, 60(3), pp. 455‐494.
____. 2010. “Reflexivity, Path‐Dependence and Disequilibrium Dynamics.” Journal of Post Keynesian Economics (forthcoming), (Fall).
Shaikh, Anwar and Rania Antonopoulos. 2012. “Explaining Long‐Term Exchange Rate Behavior in the United States and Japan ” In Alternative Theories of Competition: Challenges to the Orthodoxy, ed. J. Moudud, C. Bina and P. L. Mason. Abingdon: Routledge.
Shiller, Robert J. 1989. “Comovements in Stock Prices and Comovements in Dividends.” The Journal of Finance, 44(3), pp. 719‐729.
____. 2001. Irrational exuberance. Princeton, NJ [u.a.: Princeton Univ. Press.
Soros, George. 2009. The Crash of 2008 and What It Means. New York: Public Affairs.
2 Differentiating saving rates by income class (wages and profits, for instance, as in Kaldor and Pasinetti) allows changes in the distribution of income to modify the aggregate savings rate. Even so, it is the assumed fixity of class savings rates that leads to the result that the distribution of income (the profit/wage ratio) must adapt to make the actual growth rate conform to the natural rate of growth.
3 Keynes also notes that it is profitability, not demand, which drives production itself. “An entrepreneur is interested, not in the amount of the product, but in the amount of money which will fall to his share. He will increase his output if by so doing he expects to increase his money profit, even though this profit represents a smaller quantity of product than before” (Sardoni 1987: 75).