When Economic Giants Roamed the Earth, 8/13

Keynes and Samuelson were two of them

When Archibald and Lipsey try to draw for Patinkin a picture of what a “classical” monetary theorist believed in, they are pretty much in the position of a man who, looking for a jackass, must say to himself, “If I were a jackass, where would I go?”

Mine is the great advantage of having once been a jackass.”

—Paul Samuelson, What Classical and Neoclassical monetary theory really was

These are my all-time favorite sentences ever written by an economist.

Nowadays there must be many students of economics who know very little about the way that economic thinking evolved in the decades following World War II. That period may still cast a shadow over economics as it exists today. I cannot claim to be an expert in this history of economic thought, but I can echo Samuelson: I was once a jackass.

Think of three schools of thought that were contending in mid-century: Marxist; Austrian; and Neoclassical. In this schema, Keynes and especially Samuelson are in the Neoclassical camp. Call this Samuelsonism.

For the Marxists, the market is socially constructed to enable the capitalist class to exploit workers. Today, whenever you hear people bad-mouth capitalism and extol socialism, you are hearing echoes of this Marxist view. I think it reflects overly simplistic “moral dyad” thinking, in which you think of one actor (the capitalist) as having all agency and no feelings, while the other actor (the worker) has all feelings and no agency.

For Austrians, the market is a process that leads to an emergent order that evolves to generate improvements in living standards. Government is an inferior process that often threatens this order.

For Samuelsonists, the market is useful, but it has to be corrected using technocratic management informed by economics. The economist comes up with analysis that enables government to tune the market to optimize social welfare.

Mathematical optimization was at the center of Samuelsonism. The Defense Department helped to fund the growth of the MIT economics department to develop methods like linear programming. When I was a grad student at MIT in the late 1970s, we still had to read Dorfman, Samuelson, and Solow’s 1958 textbook.

An incredible amount of academic economic writing from the 1940s through the 1980s wrestled with three projects that were key elements of Samuelsonism:

1. Macroeconomic modeling, in order to apply Keynesian economic policy to eliminate recessions;

2. The production function, which tries to describe economic activity as machinery taking in inputs and turning out outputs;

3. A social welfare function, which the technocrat can use to judge whether a given policy option makes us better off or worse off

These projects were positioned in opposition to the Marxists. Keynesian macroeconomics was billed as the answer to the Marx’s prediction that capitalism would collapse in a series of crises. Marxists claimed that the Great Depression was vindication. Samuelsonism said that capitalism need not collapse, because Keynes had the answer.

The production function challenged the Marxist view of labor exploitation by capitalists. The production function says that output is a function of capital and labor. If you know the production function, you can use mathematics to derive both the average product and the marginal product of labor. Under certain assumptions, the marginal product of labor equals the average product of labor equals the wage rate. This makes the wage rate “fair” in some sense. Even more important, the wage rate increases as the economy grows. Capitalism is not a process that drives workers into ever-worse misery.

The social welfare function provided an alternative to having public policy be dictated by class interests. Instead, it provided an objective way to measure whether a policy made society as a whole better off.

Discredited but Valid?

Each of these projects was subjected to a critique that was regarded as devastating and ultimately unanswerable: the Lucas critique of macro; the Cambridge Capital Controversy over the production function; and Arrow’s Impossibility Theorem concerning a social welfare function.

But mainstream economics proceeded as if these critiques did not matter. Technocratic economics proceeds as if macroeconomic models, the production function, and a social welfare function are all available in the background. Economists find these concepts, discredited as they are, too useful to give up.

I am willing to brush aside the famous post-war critiques of Samuelsonism. I have different critiques, which are more Austrian in spirit.

My problems with macroeconomic modeling have been spelled out elsewhere. At the most technical level, see Macroeconometrics: The Science of Hubris.

My problem with the production function is much deeper. I do not view economic activity as machines transforming inputs into outputs. I view economic activity as specialization and trade. To take an example from David Friedman, suppose that we ship grain to Asia, and the ships come back loaded with cars. It is misleading to describe that as a production function in which grain is an input and cars are the output.

Patterns of specialization and trade are constantly being reshaped. There is no fixed relationship between inputs and outputs.

Capital, labor, and output are highly heterogeneous. In my view, the attempt to describe an aggregate production function is futile.

The problem with the social welfare function is the futile search for objective measures of trade-offs. I recommend Cost and Choice, by James Buchanan. He writes,

A century has elapsed since the subjective-value revolution in economic theory, but the subjective theory of value has not been fully reconciled with the classically derived objective theory. As the notes on the development of the concept of opportunity cost indicate, economists have not drawn carefully the distinction between a predictive or scientific theory and a logical theory of economic interaction. As subsequent chapters will demonstrate, this methodological confusion is the source of pervasive error in applied economics.

Economists understand that cost is in the eye of the chooser. But they nonetheless think of cost in objective terms, as if it were embedded in the good being chosen. Buchanan thinks hard about this problem, and I believe that his conclusions are more devastating to the Samuelson social welfare optimization project than even the Arrow theorem.

Economists nowadays are not engaged in the sort of big debates over Samuelsonism that were still salient when I was in graduate school at Samuelson’s MIT in the late 1970s. Instead, for the most part 21st-century economists examine very narrow technocratic questions.

One result of this is that economists no longer are actively debating market vs. government. The giants no longer roam the earth.

Today, pretty much everyone who writes on economic policy starts from the assumption that markets are misguided and government regulation is better. Sometimes, economists point out specific problems with non-economists’ regulatory proposals. But more fundamental, Austrian-inflected critiques go unappreciated.