| ... the ideas of economists and political philosophers,
both when they are right and when they are wrong, are more powerful
than is commonly understood. ... Practical men, who believe themselves
to be quite exempt from any intellectual influences, are usually
the slaves of some defunct economist. Madmen in authority, who
hear voices in the air, are distilling their frenzy from some
academic scribbler of a few years back.
- John Maynard Keynes |
It is no secret that economists often disagree. Post-Keynesians
and monetarists, for example, hold opposing and mutually incompatible
theories of inflation. This chapter will examine the nature of
the persistence of disagreement in economics and take note of
some of the differences between the social sciences and the natural
sciences. Then it will conduct a brief tour through the history
of economics. This tour is sort of an aerial view providing an
opportunity to observe the flow of economic thought from the beginning
of capitalism to the present. Later chapters, particularly Chapter
6, will probe these streams of thought in more depth.
The work of scientists is to solve puzzles. This is true in the social sciences as well as in the natural sciences. Before looking at the particular problems of social scientists, it will be useful to examine the puzzle-solving methods of scientists in general.
What constitutes a puzzle? Usually it is something observable that does not seem to fit the set of interrelated theories that make up the basic precepts of the science. Scientists normally work within a paradigm - that is, an interwoven web of key theories and/or axioms at the core of their particular science. The prevailing paradigm indicates what phenomena are considered to be puzzles and establishes an accepted set of methods by which these puzzles are to be solved. In today's academic system, the apprentice scientist absorbs the paradigm in the first years of graduate school and then solves a puzzle in the process of completing a Ph.D. dissertation.
The usual method of scientific puzzle-solving is to carefully state a hypothesis and then to craft an experiment. Before conducting the experiment, the scientist will state what experimental outcomes would be consistent with the hypothesis and what experimental outcomes would indicate that the hypothesis is incorrect.
An agronomist, for example, might form a hypothesis about the effect of a particular chemical on plant growth. Since she is not claiming that this chemical is the only thing that affects plant growth, she needs to design an experiment in which all of the other things that might affect plant growth are held constant. Perhaps she will plant seed of exactly the same type on five different plots of land that have the same type of soil and will receive the same amount of water and sunlight. All five plots will be cultivated in exactly the same manner. But each will receive a different amount of the chemical that is being tested. The crop will be harvested and weighed. Any differences among the five plots can then be attributed to the chemical. If there is no difference, she must reject her original hypothesis.
Of course nothing is quite this simple. Some differences among the five plots may be due to the random variations that are part of all biological processes. So our agronomist will need to delve into statistics in order to calculate what magnitude of variation is large enough to be due to something other than random factors. Additionally, when our agronomist reports her results, she will describe her methods in sufficient detail that any other agronomist can reproduce the experiment - perhaps even on a larger scale to reduce the effects of random variation.
Sometimes all the interesting puzzles are solved and the field is downgraded from a science to a mere technology. But often there is an accumulation of anomalies: observed phenomena that cannot be explained within the precepts of the paradigm, or that cannot be explained without adding numerous and klutzy subtheories to the paradigm. Eventually, a scientist - often a younger scientist or someone trained in another field - will eliminate the anomalies by developing a new paradigm. This will lead to a period of struggle during which the adherents of the new theory attempt to win converts. If they are successful, a new paradigm is born. [1]
The classic example comes from astronomy. The paradigm of Ptolemaic astronomy was based on two central precepts: that the Earth was at the center of the planetary system and that all heavenly bodies traveled in perfect circles. Even during the second century when Ptolemy developed his system, astronomers could track the movements of the sun, moon and planets well enough to know that this would not work. The solution to the puzzle was to place each heavenly body on a circular path with its center on the circumference of another circular path. These were called epicycles. As observations built up over the centuries, astronomers were confronted with new puzzles which they solved by layering epicycles on top of epicycles. Finally, Polish astronomer Nicolaus Copernicus [1473-1543] put the sun at the center and German astronomer Johann Kepler [1571-1630] replaced the circles with ellipses and a new paradigm displaced the old.
But economists, like other social scientists, study the social behavior of humans. Thus we face some problems which our colleagues in the natural sciences do not:
We cannot begin to do economics without some concept of the social psychology of humans. Even though our investigation is limited to those aspects of human nature that affect our material life, we are still entangled in a complex web. And, even if we could agree on what constitutes human nature, we could not agree on which aspects of human nature are most important for our investigations. Over the years, economists have stressed many different aspects of what they took to be human nature:
Rather than directly face the complexities of human behavior, economists have fabricated a much less complex abstraction that we call economic man. Our economic man enjoys consumption and dislikes work. His enjoyment of the consumption of any one thing, however, falls off the more of that thing he has already consumed. For example, the third steak in one week gives him less pleasure than the second steak but more pleasure than the fourth steak. On the other hand, his dislike of work increases with each additional hour that he works - the 45th hour of work in a week is more distasteful than the 44th. Since economic man is always and automatically comparing the benefits with the pains of every potential action, he will not spend a dollar on more hamburger if a dollar spent on ice cream gives him more pleasure. And he will cease working when the pleasure from consuming the goods and services he can purchase from the earnings from his last hour of work no longer outweigh the displeasure he obtains from working. When economic man is a business owner, his task is even simpler: to earn the maximum profits possible given the conditions that he faces - as long as the pleasure from increased purchasing power compensates for the pain of work.
One of the great advantages of economic man is that he fits into our theories and equations far more easily than his real world counterpart. His response to economic events is completely predictable. When the price of steak falls he purchases more of it than before. When his wages fall, he is not willing to work as much. We can fully capture his activities in a system of differential equations.
Our basic view of human nature will affect our approach to understanding the economy - after all, the economy is part and parcel of human society. Neoclassical economics, the currently dominant paradigm, is based on a vision of humans as rational calculators looking for every opportunity to increase our wealth and incomes. Moreover, according to this vision, we assume that others are just as materially motivated as we are - and this precludes many forms of cooperation that could make everyone better off!
Some recent studies have indicated that economists (remember, most economists today are deeply mired in the neoclassical paradigm) are somewhat out of the human mainstream with regard to greed and self-interest. One study found that economics professors give substantially less to charities than professors in other subjects, even though their incomes are on the average higher than the incomes of other professors.
Additionally, a number of experiments were undertaken with graduate students from various fields. The basic idea behind these experiments, which involved money, was that people who cooperated could get more money than people who didn't, but that this would only work if one had faith that others would cooperate as well. In one of the experiments, the subjects were each given an amount of money and asked to divide it into a private account and a public account. Each student could keep the money which went into the private account. The money put in the public account was multiplied by some amount greater than one (for example, 1.5) then equally distributed among all the subjects. Imagine that each of 20 subjects is given $60. If all of the subjects each put all of the money into the public account, each could then walk away with $90. But each subject had to divide the money without consulting with the others - so the decisions of each were at least partially based on what each thought the others would do. If one put all the money in the public account while all the others put it in their private accounts, the cooperator would end up with only $4.50 (1/20 of $90) while each of the others would get $64.50 ($60 plus 1/20 of $90). If one put it all in the private account while all the others put it all in the public account, the selfish one would walk away with $145.50 while each of the 19 cooperators would get only $85.50. The experimenters found that economics students put a significantly smaller portion (20% on average) of the money into the public account than did the students from other fields (50% on average). When asked why they divided the money the way they did, most of the non-economists indicated that they had wanted to be "fair." But most of the economists were either unconcerned with fairness or didn't seem to understand the meaning of fairness in this context. After reviewing several such experiments, The Economist concluded:
Students of [economics] are trained to regard self-interest as the force that decides economic choices. It is easy to imagine cases where cheating is advantageous. The economist's view is: others will see that the logic of the situation calls for cheating, so you had better cheat, too. This idea pervades the [economic] literature. But here's a disturbing thing: it may be having some effect. Nothing personal, but economists are an unpleasant lot. [4]
One obvious reason for the continuing coexistence of competing paradigms in the social sciences is that social scientists can rarely design the definitive experiment that will finally nail the coffin lid shut on yesterday's paradigm. Definitive experiments require at least two conditions that are missing from most social science research: an unambiguous system of measurement and the ability to control all of the relevant variables during the experiment.
We can start with measurement. Units of length, weight, temperature or energy can be defined once and for all. Even if one eccentric scientist insists on using pounds and ounces, others can convert the figures to kilograms by hitting a couple of calculator buttons. But economic measurement is quite different. If we 'measure' something in dollars in order to calculate a growth rate, we quickly bump up against the fact that today's dollar is not exactly the same thing as yesterday's dollar or tomorrow's dollar. The economist cannot simply go to the Smithsonian Institute and find a standard dollar kept in a vacuum under a bell jar. We have developed techniques to measure changes in the value of a dollar, but there are several to choose from and none are perfect. So a seemingly simple question such as 'Did average wages in the United States rise or fall from 1981 to 1990?' gets bogged down in an unresolvable argument over the accuracy of the Consumer Price Index. Similar difficulties apply to measuring such things as the amount of money in circulation and the level of unemployment. The Appendix to this chapter examines the difficulties involved in using money as a unit of measurement.
Nor could we ordinarily undertake controlled experiments, even if there were no problems of measurement. Economists would like to measure the effect of changes in interest rates on investment spending. If only we could imitate the agronomist described earlier. After all, we have a similar problem. Our theory tells us that interest rates are one factor among many which affect investment spending. But, unlike most of the other factors, interest rates can be purposely manipulated. So economists - particularly economists working for a nation's central bank (which does the manipulating) - would dearly like to quantify the relationship between interest rates and investment spending. All we need is a ten year 'experiment' in which interest rates change while all of the other things that affect investment spending (business confidence, level of profits, growth of demand, etc.) remain constant. Of course we can't do that, so we use statistical procedures instead; but they are much less reliable and much more subject to varying interpretations.
Since we can't usually conduct experiments, economists have become quite adept in the use of statistical correlation procedures. When our theory tells us that there is a cause and effect relationship between two variables, we can subject it to statistical verification. For example, our theory suggests that an increase or decrease in wages will cause consumer spending to increase or decrease. This is not a particularly controversial hypothesis. But we should be able to do more than simply refute or accept the hypothesis in this case; we should be able to establish a quantifiable relationship between wages and consumer spending. That is, we would like to be able to say something like "Consumer spending increases by $0.80 for every $1.00 increase in wages."
Although wages may be the main thing that affects consumer spending, wages will not be the only thing. A drop in interest rates might increase spending on autos and major appliances. If the news media are full of articles about major layoffs, people may cut their spending because they fear layoff even though they are still earning normal wages. Or people might increase their spending because they believe taxes will be reduced in the near future. So our economist will have to construct a complex equation which expresses the hypothesis that consumer spending is affected by wages, interest rates, and a number of other things. Some of these things will be easy to quantify - interest rates, for example. For others, the economist will have to concoct a measurement of the unmeasurable - such as a numerical index of "consumer confidence." Each of these numbers will have to be available over a particular period of time. Perhaps there will be a figure for each of them every quarter for the last ten years. The numbers can be tossed into the computer and a statistics program will then crank out some processed numbers. One of the numbers will purport to indicate how much consumer spending changes whenever wages change. It might tell us that consumption will increase or decrease by $0.85 for every $1.00 increase or decrease in wages. Some other numbers will help the economist get a sense of how 'fuzzy' the first number is. They might warn: "50% OF THE CHANGE IN CONSUMER SPENDING IS STILL UNEXPLAINED," and/or "EXPECT THE $0.85 FIGURE TO BE OFF BY + OR - $0.35 20% OF THE TIME." In short, the results of statistical analysis of economic issues are usually fuzzy numbers which must be used with caution.
Another difficulty in relying on statistical correlation is that correlation cannot separate cause from effect. If event A and event B are strongly correlated, it is possible that A causes B; that B causes A; that we have circular causation [A causes B and B causes A]; that C, D, E, or even X [something we have not identified] causes both A and B; or even that A causes B under some conditions and that B causes A under other conditions.
The investigation of hypotheses about the causes of inflation provides an interesting example. Monetarists (a subsect of neoclassical economists) claim that inflation is caused by the amount of money in circulation growing faster than the economy grows. If, according to the monetarists, the economy grows by three percent per year and the money supply grows by eight percent per year, we will get an inflation rate of five percent per year. Of course any actual test of this hypothesis is beset by numerous measurement problems. What measure of the money supply do we use? What measure of inflation? Over what time frame is this supposed to occur? But even if we resolve these problems and even if we find a 'perfect' correlation (which we do not), all we could say is that the monetarist hypothesis is consistent with the data. The correlation only shows us that the money supply growth and the inflation rate go - very roughly - up and down together. But we still have to consider theories that claim something else causes inflation but that inflation in turn causes the money supply to increase faster. When a social scientist supports a hypothesis by establishing a statistical correlation between the alleged cause and effect, there is still a need for a coherent story explaining precisely how the cause brings about the effect.
In short, disagreements in social sciences persist far longer than in natural sciences. Certainly, all disagreements are not equal. Many occur when economists who subscribe to the same paradigm put emphasis on different factors. It should not be surprising that forecasts of economic growth will vary among different economists, even though all the forecasters agree on the same basic theories. The U.S. budget battle of 1995/96 is based on this sort of disagreement. Congressional Budget Office economists forecast an average of 2.5% per year growth over the next seven years; White House economists forecast an average of 2.7%.
More serious disagreements occur when economists working from different paradigms collide. A neoclassical economist will analyze proposed changes in the tax laws by hypothesizing how the incentive to invest will be affected; A post-Keynesian would start with a hypothesis on the effects of the new taxes on consumption spending.
The mainstream of economic thought in the capitalist economies starts with Adam Smith in 1776. Smith and his followers developed Classical Political Economy. By the 1870s, Classical Political Economy began to be replaced by Neoclassical Economics, based on a distinctly different paradigm. Parts of John Maynard Keynes' 1930s critique of Neoclassical Economics became incorporated into the Neoclassical/Keynesian Synthesis by the late 1940s. In the 1970s, the Keynesian part of the synthesis gradually began to be chipped away, leaving neoclassical economics as the dominant paradigm once again.
Early modern economics was a product of the early stages of industrial capitalism. Adam Smith, a Scottish philosopher, was writing The Wealth of Nations (published in 1776) even as his compatriot James Watt was developing the first practical steam engine. Certainly there were earlier writings on subjects we might now describe as economic. Aristotle's 'economic' writings shed some light on a society in which most work was done by slaves and work was considered to be beneath the dignity of a free man. Medieval theologian/scholars developed a theory of prices, but it focused on what prices 'should' be rather than on what prices actually were; we would now consider it more moral than scientific.
Nor could it be otherwise. When the market was at the periphery of society rather than at its center, material life had no visible existence independent from political, social or religious life. One did not need to develop a theory of labor markets to understand the relationship between serf and lord or between guild-master and apprentice. Access to what we now identify as the factors of production - land, labor and capital - was by birthright more than by competitive bidding. If time travel should someday become a tool of social science research we should send anthropologists and political scientists rather than economists back to investigate the material life of pre-market societies.
As the market began to penetrate more areas of social life in Western Europe - and as the traditions and structure of feudalism slowly lost their grip - there were a number of attempts to understand (or, often, to manipulate) the new set of social relationships embodied in market society. Foreign trade was a major object of interest, perhaps because the market linkages were more quickly developed there than in domestic trade. There was no shortage of thought on this issue, and much of it took the form of pamphlets or of advice to government ministers. These early economists, called mercantilists, were mostly concerned with helping their governments maintain trade surpluses which would bring a continual stream of gold and silver into the country. They dominated economic thinking in the seventeenth century, particularly in Britain and Holland.
Classical political economy began with Adam Smith's The Wealth of Nations in 1776 and dominated economic thought for nearly a century. Among the major works of classical political economy are:
While the classical economists came to many different conclusions about the nature of capitalism, their approaches were similar enough that they can be considered a distinct school of economics. Note that from Smith's time until well into the nineteenth century no one was actually educated as an economist. Political economy was considered a branch of social philosophy. Smith, Mill and Marx were all educated as philosophers. Malthus studied Mathematics then became an ordained minister. Ricardo was a successful stockbroker and a member of parliament.
The classical economists were influenced by other factors as well. Their philosophical training led them to search for a common factor which would explain all economic phenomena. The emerging physical sciences, particularly Isaac Newton's physics, suggested that everything could be explained by a few simple rules. Adam Smith found that prices were set by the counterpoising forces of self-interest and competition, just as Newton had found that the moon was held in place by the counterpoising forces of gravity and centrifugal force. Karl Marx set out to discover the "laws of motion" of capitalism. Classical political economists believed that the economy, just like the movement of the planets, was governed by natural laws. The task of the economists was to discover them.
Of course the major influence on their thinking was the time and place in which they lived. All were able to observe the early stages of the industrial revolution. It is no accident that most of the classical political economists were British. Britain had a 25 to 50 year head start over the rest of Europe on the industrial revolution. Marx, after absorbing philosophy in Germany and observing political turmoil in Belgium and France, spent most of his working life in London.
In spite of the differences among such a large group of economists, we can identify the major elements of the classical political economy paradigm:
By the 1870s the field of study that had been known as political economy and had drawn the attention of philosophers was becoming mere economics and attracting the attention of engineers and mathematicians instead. We can observe the change by listing the titles of some of the major works of the early neoclassical economists:
Neoclassical economists shifted the emphasis from production to exchange. Their starting point was to capture the essence of economic man in a set of axioms. They then generated hypotheses based on these axioms. They strove to recast economics as if it were a branch of mathematics, similar to geometry. These hypotheses were presumably subject to statistical tests, but the reliability of these tests are often in doubt because of the softness of the data. Some of the more fundamentalist among the neoclassical economists argued that it did not really matter if economic man was a realistic abstraction or not, as long as the hypotheses derived from it survived statistical tests. [7]
Neoclassical economics became the dominant paradigm for a combination of scientific, esthetic and ideological reasons. On the scientific side, neoclassical economics clearly handled what we now call microeconomic subjects - the pricing of individual items - better than classical economics had. On the esthetic side, the smooth curves and sets of differential equations could actually be quite graceful. All right, I admit that this is an acquired taste, but so are James Joyce and good champagne.
The speed with which neoclassical economics displaced classical political economy was also due to its amicable treatment of capitalism. Classical political economists wrote about landlords, workers and capitalists contesting for a larger share of the pie. In neoclassical economics there are only factors of production. Land, labor and capital each receives a reward commensurate with its productivity. Karl Marx had worked from within the classical paradigm and turned political economy into a critique of capitalism itself. No one would be able do anything similar with neoclassical economics, which had shifted the stage from society to the individual.
As with classical political economy, we can identify the major elements of the neoclassical paradigm:
The Great Depression was a turning point in economic theory as well as in economic policy. The neoclassical economists could not offer solutions for something that, according to their theory, could not even exist.
In the 1930s Cambridge University economist John Maynard Keynes built a new economic model that could explain such events as depressions by exploring a linkage that had been ignored by mainstream economists - the linkage between overall spending and income. Whenever any form of spending fell, Keynes claimed, income also fell. This fall in income would bring down consumer spending, reducing income even more. The economy could get stuck in a slump. Consumers couldn't break the economy out of the slump because they lacked the income to increase their spending. Business would not increase output (and thus employ people) because the public did not have enough spending power to buy their products. And businesses would certainly not spend on new plant and equipment (investment spending) when they already had far more production capacity than they could use. The solution was for government to increase its spending (on roads, for example) and to cover the cost of the increased spending through borrowing rather than raising taxes. Newly employed workers would spend their incomes on consumer goods, businesses would increase their output to meet the new demand. Eventually business would be spending on new plant and equipment and the extra government spending would no longer be needed.
But Keynes' new paradigm did not replace neoclassical economics. Rather it found a place beside it. Henceforth the neoclassical theories of value, price and income distribution would be called microeconomics and the study of fluctuations in total output would be called macroeconomics. Both could be put into the same textbook and called "the neoclassical/Keynesian synthesis."
The synthesis eventually failed on two counts. First, economists as scientists were uneasy about needing two different sets of first principles - even two different definitions of demand. Keynesian economics was seen by many as a somewhat unsatisfactory set of 'epicycles' needed to explain the uncomfortable facts of recession and depression. Second, Keynesianism as a set of practical recipes for guiding a modern economy lost some of its relevance by the mid-1970s when the advanced industrial economies were subjected to stagflation. Keynesians had no ready explanation of the simultaneous occurrence of unemployment and inflation; moreover, they had no policy prescription which could alleviate one without making the other worse. Many of the Keynesian elements of the synthetic paradigm gradually slipped away leaving the intact neoclassical paradigm as the riverbed of today's economics mainstream.
The mainstream is not the only stream. At least three dissenting paradigms have survived from their beginnings to the present. Marxist Political Economy is deeply rooted in Classical Political Economy. It started with works by Karl Marx from the 1840s until Marx's death in 1883 and continues to the present. The Marxist paradigm will be skipped over in this chapter because it is quite similar to the classical paradigm. Marx's main economic theories will be covered in Chapter 11. Institutionalism started with a dispute over methodology among (mostly) German economists in the 1880s, and was relaunched in the United States by Thorstein Veblen around 1900. Institutionalism continues today in the work of John Kenneth Galbraith, Robert Heilbroner and others. Post-Keynesian Economics began when several members of Keynes' inner circle, led by Joan Robinson, became convinced that the Neoclassical/Keynesian synthesis was ignoring the central message of Keynesian economics. The Post-Keynesians have since built a strong base in some British and American universities.
By 1900 there were essentially two economic paradigms. The neoclassical paradigm was dominant but continually challenged by the Marxists. Yet the two competing paradigms did have a few things in common. Both assumed that humans were rational beings who would act in their own economic interest - although this was part of 'human nature' in neoclassical economics but was behavior peculiar toparticular economic systems in Marxist economics. Both were deterministic. Marx had identified an inevitable progression from feudalism to capitalism and from capitalism to socialism. Marxism was an extension of the natural law approach of classical political economy. Neoclassical economics claimed to be an axiomatic science. In neoclassical theory, consumers with a given set of preferences (tastes) interacted with the technologies of production to fully determine the distribution of income and the rate of economic growth. And the foundations of both paradigms were based on the idea that the economic side of life was more fundamental than other aspects of social life.
The Institutionalists rebelled against these deterministic theories. Thorstein Veblen, in The Theory of the Leisure Class [1899], offered a picture of consumer behavior in which the consumer was anything but rational. Veblen's consumer used goods and services as a form of ceremonial display: men wore white starched shirts to show that they didn't do any useful work; women cultivated long fingernails so everyone they met would know they could afford to have maids do their dishes.
This is but one example. Present-day Institutionalists such as John Kenneth Galbraith and Robert Heilbroner refuse to separate the economy from the rest of society. Institutionalists claim that the market system is itself a cultural creation rather than something inherent in the nature of humankind as the neoclassicals would have it. We see even the modern economy as a social system in continual evolution. Unlike Marxists, we do not see the outcome of this evolution as an inevitable movement toward a more perfect society. We are more inclined to agree with Veblen that the future of society is what we make of it, for good or for bad:
History records more frequent and more spectacular instances of the triumph of imbecile institutions over life and culture than of peoples who have by force of instinctive insight saved themselves alive out of a desperately precarious institutional situation. [8]
Our preferred method of research is to study an economic event or institution historically, to examine its unfolding over time. Contrast this with the neoclassical methodology of describing an economic issue in mathematical formulas at one point in time. Institutionalists reject the idea that the natural sciences represent a model for economics to follow. The movement of electrons and the division of cells are governed by natural laws. Humans are governed by human laws, and human laws are rooted in human social institutions, which - unlike natural laws - are subject to change.
It is somewhat more difficult to characterize an institutionalist paradigm than the classical or neoclassical paradigms. Institutionalists often define ourselves by identifying the elements of neoclassical economics that we reject. Certainly Institutionalists have little use for 'economic man.' Nonetheless, the institutionalist paradigm includes:
The Institutionalists claimed that neoclassical economics was built on false premises, undertaken with overly formal and rigid methodology, and led to results that bore little relevance to the conditions of any actual economy. The Keynesians claimed that neoclassical economics was correct only under conditions of full employment and built a new theory that could explain situations of less than full employment yet still dovetail with the dominant neoclassical paradigm. But a new movement, called Post-Keynesianism in the United States and Neo-Ricardianism in Great Britain, started in the 1950s. The Post-Keynesians initiated a direct attack on the very logic of neoclassical economics rather than simply note the irrelevance of neoclassical economics and then ignore it as the Institutionalists had done.
The neat mathematical models of the neoclassicals, according to the post-Keynesians, assume that all economic actors - firms, workers, consumers - know what the outcomes will be of all of the actions they might undertake (such as investing in a new steel mill). If they do not know the actual outcomes, they at least know the probability of any particular outcome, just as you and I know the probability of drawing an ace from a shuffled deck of cards. The major contribution of Keynes, and the one which never entered the neoclassical/Keynesian synthesis, was to recognize that all expectations of the future are uncertain and unknowable; even the probabilities of particular outcomes are not known. Investing - or undertaking any economic action in the real world - is more like trying to draw an ace from a 52-card deck that might contain anywhere from 0 to 52 aces. So the post-Keynesians set out to build a new economics based on uncertainty.
The post-Keynesian paradigm contains the following elements:
Social sciences are fundamentally different from natural sciences, yet economists and other social scientists often emulate natural scientists. Unfortunately, most of the measurements we use are far less accurate than measurements used by natural scientists. And we can rarely carry out actual experiments. Statistical analysis, while quite useful, is a poor substitute for actual experimentation. Like natural scientists, we sometimes resolve an accumulation of anomalies by creating a new paradigm. Unlike natural scientists, we are unable to completely discard the old paradigms and thus we continue to have competing schools of thought.
Modern economics started with the Industrial Revolution in Great Britain. From 1776 (Adam Smith) to 1867 (Karl Marx) most economists worked from a distinct paradigm which we now call classical political economy. Much of this paradigm lives on today in the somewhat modified form of Marxist political economy. The mainstream of Western economic thinking has been based on the neoclassical paradigm from about the 1870s to the present. Neoclassical economics suffered a major setback during the Great Depression, but quickly absorbed some of the elements of Keynesian economics into the Neoclassical/Keynesian Synthesis. But the effect was rather like a neoclassical snake swallowing a Keynesian gopher: the Keynesian part was quite visible at first, but the compatible components were eventually digested, the incompatible pieces were eventually egested, and very little of the gopher is left in today's neoclassical mainstream.
There are, however, three dissenting paradigms. Marxist political economy has inherited the classical paradigm. Institutionalism rejects the natural science pretensions of both neoclassical economics and Marxist economics. Post-Keynesian economics is carrying forward the elements of Keynesian economics, particularly the role of uncertainty in economic decision-making, that never became part of the neoclassical/Keynesian synthesis. The major differences between the post-Keynesians and the Institutionalists are found in their methodologies and in their stance toward neoclassical economics. Institutionalists claim that neoclassical economics is irrelevant and generally ignore it, while the post-Keynesians have launched a direct attack on the logic of neoclassical economics.
1. The role of paradigms in scientific work is covered in Thomas S. Kuhn, The Structure of Scientific Revolutions, Second Edition, Chicago, 1970. This concept is fruitfully applied to economics in Benjamin Ward, What's Wrong with Economics, Basic Books, 1972.
2. Wealth of Nations, Book I, Chapter II.
3. The General Theory, Chapter 12, Section VII.
4. The Economist, "How do you mean, "fair"?" May 29, 1993.
5. Karl Marx holds a dual position in the history of economic thought. He is both the last classical political economist and the first Marxist.
6. When elaborated on by Ricardo and (later) Marx, this became the Labor Theory of Value. Theories of value will be examined in Chapter 6.
7. One of the clearest explanations of this method is found in Milton Friedman's essay, "The Methodology of Positive Economics," in Essays in Positive Economics, Chicago, 1953, University of Chicago Press.
8. The Instinct of Workmanship, 1914.
1. Why are actual experiments usually stronger methods of evaluating hypotheses than statistical methods?
2. Why does economics retain competing paradigms longer than physics, astronomy or biology?
3. Should economics try to imitate the approach taken in physics and astronomy? Might the biological sciences be a better model for economists to follow? How about law?
Neoclassical/Keynesian Synthesis
Economic Theory - Economics Explained, Chapter 2, "Three Great Economists."
Brockway, George P., The End of Economic Man.
Third Edition, W. W. Norton, 1995. A lively and perceptive institutionalist
critique of mainstream economics. Non-technical.
Colander, David C., Why Aren't Economists as Useful as Garbagemen?
Essays on the State of Economics, M.E. Sharpe, 1991. Read
this before you decide to major in economics.
Heilbroner, Robert, The Worldly Philosophers, Simon &
Schuster, 1980. Lively coverage of the life, times and ideas of
the major economists. Highly recommended.
Kuttner, Robert, "The Poverty of Economics," The
Atlantic Monthly, February 1985. Kuttner, now a regular economic
columnist for Business Week, accurately summed up the state
of economics in 1985. Unfortunately, very little has changed in
the intervening years and his essay is still current.
Landreth, Harry and Colander, David C. History of Economic
Thought, Third Edition, Houghton Mifflin, 1994. After you've
read The Worldly Philosophers and taken a few economics
courses Colander and Landreth can provide that deeper knowledge
of the history of economic thought that you will be craving. A
thorough history of economic thought for the serious student.
Many of the major works of early economists
are available electronically: http://socserv2.socsci.mcmaster.ca:80/~econ/ugcm/3ll3/index.html
Institutionalists publish
the Journal of Economic Issues. An index of articles from
1982 to the present, by issue and by author, can be found at:
http://www.uis.edu/~tpowell/jei.htm
The Post-Keynesian home page at http://csf.colorado.edu/econ/
is cross-linked to many post-Keynesian sources and forums.