Economics

Know this topic well? Help others and get FREE products!

The Social Science Encyclopedia, Second Edition

In the draft of the introduction to his monumental study of the history of economic analysis, Joseph Schumpeter (1954) writes as follows:

Few contemporary economists view their subject as stretching back continuously to Graeco-Roman civilizations, and some would regard a close association with history or philosophy as unnecessary, if not positively undesirable. Yet Schumpeter’s account takes up nearly 200 pages before reaching Adam Smith’s (1776) The Wealth of Nations, which is where economics began according to a vulgar view.

There is no great contradiction here. A reading of Schumpeter shows that he, as much as any other commentator, recognizes that the study of economic questions underwent radical changes of method, and an accelerated pace of development, in the late eighteenth century. Yet contemporary historians of economic thought agree with Schumpeter in not attributing the beginning of modern economics to the European Enlightenment, still less to the Scottish Enlightenment and Adam Smith. On examination the mercantilist and liberal writers turn out to share as much in common as they have differences. As in other cases, a sharper contrast exists between modern and early thinkers together, and their pre-modern predecessors (on merchantilism, see Magnusson 1994).

What began, whenever it began, and accelerated in the late eighteenth century, was not the description, not even the analysis, of economic institutions and questions, all of which are ancient. It was a redefinition of the centre of focus for such studies. A part justification of the simple belief that it all began with The Wealth of Nations is that the way in which Smith defined the subject is, in broad outline at least, recognizably the same as the subject defines itself today.

First, the renaissance idea that humankind is part of nature is sovereign. This frees economic analysis to apply rationalist and reductionist methods in order to attack economic superstitions, such as the belief that true wealth consists in the possession of gold.

Second, economics is liberated from morality, but not in the direction of an amoral political economy of nation-state power, which the mercantilists and others had explored. Rather many simple ‘moral’ views are discredited, often by the device of showing that their intentions fail to translate into moral outcomes. A phrase frequently used by Adam Smith captures this change of philosophy perfectly. Mercantilist restrictions are unreasonable. This means, of course, that like all institutions they are subject to critical rational review, and that they fail that test.

Third, if economics is not about selecting policies to promote the power of the sovereign, then what is it about? A view characteristic of the liberal individualism of the seventeenth and eighteenth centuries, again inherited by the modern discipline, has it that it is about the individual (or the individual family) and about the proper field, conceived as a wide one, for the pursuit of self-interest. The understanding that self-interest may be a socially constructive force is what separates the economist from the one who derives his views on economic issues solely from common sense. Equally, the understanding that self-interest may need to be channelled or even constrained to achieve good social outcomes is essential to a sophisticated economic insight.

Finally, this last tension, between the field for liberal policy, even laissez-faire, and the field for intervention, makes economics essentially concerned with policy. True there exist economic questions which are inherently abstract; questions addressed to the formalized issues which considerations of preference, decision and equilibrium throw up. Yet policy is never far away. As an illustration of the point, consider that no one would ever have wasted time on determining what is the level of normal prices if normal price was not considered as, at least potentially, an input to economic policy.

Description and analysis sound like science, but what of policy? Can there be a science of policy? Is indeed economics itself a science? The question is often asked thoughtlessly, without the enquirer having a clear notion of what counts as a science. Yet it conceals serious issues. Obviously economics is not physics, however physics may be conceived. Biology, or even climatology seem to be closer cousins. Economics has models, data and, certainly, simplifying assumptions. But it also concerns itself with values, with what is good or bad in economic states. Science is usually conceived of as value free; not in the sense that subject matter is chosen at random, but with the meaning that it concerns itself with what is, not with what ought to be.

Economics has tried to wriggle around this dilemma for most of its modern history. If there can be no ought to be, how can there be policy? And without policy, will not economics reduce to a kind of descriptive anthropology of economic life? If, however, values enter into economic discourse—a higher gross national product is good, fluctuations are bad, etc.—how is economics to escape from a mire of relativism in which its values are shown to be just one of many possibilities?

The question was posed most sharply by Professor Lionel Robbins (1952), who asked by what right economists maintained that the abolition of the Corn Laws in nineteenth-century Britain was a good thing. While it was good for manufacturers, and possibly for their workers, it was surely bad for landlords. Who were economists to say that landlords losses counted for less than the gains of manufacturers?

Showing that history was on the side of the manufacturers would hardly settle the matter. History after all was on the side of Cortés and against the South American aborigines at one time, but few people today regard that as settling the value issues. A different tack was tried by Kaldor (1939), and a similar argument explored by Hicks (1969). An economic change could be regarded as an improvement, argued Kaldor, if it carried the potential of making everyone in the economy better off. This fascinating argument, and its problems, almost define what the modern field of welfare economics is about.

The assumption that individual tastes are sovereign was not contentious for economists, although others might question it. Preferred by all had to mean better. Making individual self-interest basic to welfare evaluation goes back to Pareto and beyond. Hicks (1969) is a striking challenge to this view. Later investigation showed that even if the sovereignty of individual preference is conceded, there remain serious conceptual problems for welfare evaluation. The most fundamental statement of these difficulties is the hugely influential work of Arrow (1963), which shows that a social welfare function derived from individual preferences, according to weakly constrained principles, implies the dictatorship of one individual over the ranking of at least three possibilities. Arrow’s axioms for the social welfare function have proved controversial (see Sen 1979) but there is no trick solution to a problem which just reformulates itself when the question is redefined.

Some economists have elected to throw away value neutrality and to investigate the implications for policy of specific value weights. Others have relied upon the important and useful fact that several important results in economic theory are value neutral in the sense that they hold true for all value weights. An example would be the very general result of public finance theory which says that lump sum taxation imposes less cost on those taxed (has a lower excess burden), for a given revenue collected, and regardless of how the burden of taxation is distributed, than any other form of taxation (see Atkinson and Stiglitz 1980: ch. 11).

This is not to say that one can easily manoeuvre around the problem that economic pronouncements are heavily value-laden. Lump-sum taxation completely severs the link between efficiency and distribution, and creates a Utopian world in which economic institutions function perfectly to take the economy to its utility possibility frontier, while leaving the policy maker to select a point on that frontier. In general, efficiency and distribution are more closely bound together when policy is designed with feasible taxes and other instruments to hand. Modern economics has made much of incentive compatibility, which means the design of institutions, including taxation, but much more as well, so that the individual agents involved will find it in their self-interest to do what the policy maker intends. On incentive compatibility and the associated concept of mechanism design, see Fudenberg and Tirole (1991: ch. 7).

Policy design as conceived by modern economic theory is hugely more complicated and sophisticated than its eighteenth-century ancestor, with its sometimes simple ‘get rid of irksome interference’ approach. The role of the state and how it should best operate are well modelled in the context of the design of systems of taxation and regulation. Outside those particular areas, policy design needs a good deal of improvement, a point to which I shall return.

A good measure of how economics has grown and changed over its recent centuries of history is provided by the story of GNP (gross national product) or its equivalents. Before the twentieth century it would be no exaggeration to say that there was no well-developed flow concept of economic well-being. The wealth of nations, or regions or peoples, were the object of comment and investigation. Naturally, wealth and income, as the writings of Cantillon (1697–1734), Petty (1623–87) and Quesnay (1694–1774) confirm, are closely related concepts (see Cantillon 1755; Petty 1927; Quesnay 1759). Yet they part company in the short run, as when a depression lowers incomes substantially while little affecting wealth. To record fluctuations in national income at all precisely demands a systematic collection of statistics, via a census of production, which had to wait until the mid-twentieth century to become reality.

Before that happened, the theory of national income as a concept and as a measure of economic welfare had been developed. Pigou (1955) is a good short exposition. The theoretical framework was implemented in modern GNP accounting which has graduated in half a century from an academic exercise to perhaps the most important objective and indicator of government policy.

It is no surprise that the success of GNP accounting in measuring the level of activity for the economy has been followed by deep criticism of GNP as an objective of policy. More than one type of issue is involved and limitations of space prohibit a full discussion here. Yet two basic points are frequently encountered. They can be briefly labelled consumerism and the environment. In short, the consumerism argument says that GNP is biased towards a market notion of human welfare which overweighs capitalist lifestyle and consumption and underweighs non-market activities and lifestyle. Thus transport to commute to work, and health costs associated with commuting, enter into GNP, while the value of home-grown vegetables does not. Similar arguments become even more heated where the environment and finite resources are concerned. Modern industrial production based on the burning of fossil fuels boosts GNP. Yet the cost which is the reduction of the levels of those fuels remaining in the ground is not deducted from GNP. There are no easy answers to any of these issues; yet their existence reminds us of the change in focus of economic thought among professionals and non-professionals alike.

Probably the ideological side to economics explains why the subject has been prone to controversy and to discord between competing schools. In the Anglo-Saxon world, however, the predominant method has become the neo-classical approach. Most economists would agree with that view even when some would regret the fact. But what exactly is neo-classical economics, and what are the alternatives which have failed while the neo-classical variety has flourished?

In summary, neo-classical economics is built on the assumption that efficient markets arrive at an equilibrium in which the separate actions of rational maximizing agents are made consistent via market equilibrium (or market clearing) conditions. In the absence of externalities, it is then a theorem that the resulting allocation of resources is efficient in the sense of Pareto. This is the so-called first fundamental theorem of welfare economics. The second fundamental theorem states that any Pareto efficient allocation can be made a market equilibrium when suitable lump-sum transfers place each agent on the right budget constraint. These two results fully clarify the relationship between efficiency and market equilibrium for competitive economies.

Such a clear intellectual framework makes it easy to classify the main lines of attack for anti-neo-classical economics. These concern

Some of these arguments are old, some more or less modern. The economist’s assumption about the motivation of ‘economic man’ have always seemed absurd, and immoral as well, to many outsiders. Thorstein Veblen (1857–1929) articulated passionately the problems with the computational abilities implied by the assumption that economic agents maximize (Veblen 1904). Keynes, on one interpretation, is partly about the macroeconomic effects of sticky prices. The idea that imperfect information is crucial to market failure may have been mentioned in the nineteenth century, but until recent years there is no analysis of the idea worthy of the name. Imperfect competition, especially duopoly, was analysed by Antoine-Augustin Cournot (1801–77) amazingly ahead of his time in the mid-nineteenth century, and analysis continued ever since. The idea that markets may be missing, as opposed to just not functioning well, is certainly modern, and is tied up with the again modern concept that sources of information and transaction costs should be modelled explicitly.

Now we come to a powerful paradox. As models developed from the shortcomings of neo-classical theory listed above become increasingly refined and sophisticated, they frequently and typically emerge from the work of economists of the neo-classical stable. Indeed the neo-classical critique of neo-classical economics has become an industry in its own right. In this situation a truly anti-neo-classical economics has not been easy to construct. Akerlof (1984) is one of the most interesting attempts to do so, and the frequently neo-classical nature of approaches based on highly non-neo-classical assumptions is suggestive.

It must of course be easy to construct an entirely non-neo-classical economics if one is willing to ditch enough baggage. In particular, theories in which agents behave according to more or less arbitrary rules produce radically different conclusions. But is the theory that results economics? This debate tends to separate theory and empirical investigation. Ad-hoc behavioural assumptions are hard to motivate or to formalize. Yet empirical studies have sometimes forced economists to recognize them as fact. A notable case in point is choice under uncertainty There some evidence is solidly against the most powerful and appealing theoretical model. Yet the evidence seems to indicate what may be going on, and formalization of that has been attempted (see Machina 1987).

If there is little truly anti-neo-classical economics but mainly neo-classical economics in need of more refinement and sophistication, where does that leave the Keynesian revolution? Keynes (1973) saw himself as correcting a (neo)classical economics, the root specifications of which he was to a great extent willing to accept. Specifically, he argued that his unemployment could be encountered in a competitive economy in some kind of equilibrium. How far he was correct is a controversial question.

A modern North American school of New Keynesian thinkers maintains that Keynesian economics can be made to work only by introducing the specific modelling of market imperfections, information asymmetries and missing markets (see Blanchard and Fischer 1989). Certainly the treatment of expectations differs notably between Keynes and his modern successors in macroeconomic theory. Keynes argued for the subjectivity and irrationality of expectations concerning the future. Some modern theorists, by contrast, have gone to the opposite extreme by studying rational expectations—expectations as efficient and perfect as it is possible to be in the circumstances (Begg 1982). Although taken to its logical limit, this approach soon becomes mathematically elaborate and even indeterminate, there lies at the bottom of the debate some fundamental questions concerning Keynes’s economics which have asserted themselves from the start of the Keynesian revolution.

With the flexible output prices assumed by Keynes, the theory asserts that an increase in effective demand will generate an increased output from the supply side by raising output prices relative to wage costs, that is, by cutting the real wage. How will this do when workers become aware of the fall in their living standards even when employed? There are various answers, none 100 per cent satisfactory. Keynes’s discussion suggests that price inflation co-ordinates real wage cuts for all workers which they cannot achieve acting on their own, but will willingly accept when generalized because imposed by price inflation. New classical macroeconomists argued that output could not be increased by a boost to effective demand, as workers willing to accept a cut in their real wage could have taken it anyway. Hence only inflation surprises would increase output from the supply side. With rational expectations the scope for policy using inflation surprises would be severely limited. These arguments contained a germ of truth, as with the conclusion that rapid and expected inflation is soon discounted and has few real effects. Experience has shown that Keynesian effects continue to exist but that their scope is shorter run than the earlier Keynesian theorists assumed. Unemployment, however, has proved to be a growing problem for industrial countries, leading some to conclude that structural factors and dual labour markets may now be more important than the demand side.

The intricacies of rational expectations analysis provides an example of the growing use, and eventual dominance, of mathematics in economic theorizing. Economics in the 1990s is an applied mathematical subject to an extent unimaginable a century ago. This is not because the mathematically literate have conspired to squeeze out their less advantaged colleagues. Rather, mathematics is a natural language for economic reasoning. Obviously mathematics, including the differential calculus, lends itself to maximization. Yet for all the arguments ever since Veblen, maximization is seldom vital; it often serves to provide a definiteness which could in principle be provided by alternative means. More important, however, is the power of mathematics in dealing with systems in which many influences make themselves felt simultaneously—a typical situation for social science models. Solving simultaneous equations really requires mathematics, and anyone who thinks that they can do it intuitively is practising self-deception.

Another great merit of mathematical reasoning is that it facilitates connections between theoretical and empirical work. For all the hostility of Keynes himself to mathematical economics and econometrics, the Keynesian revolution served to bring theoretical and applied economics closer together than had ever previously been the case. The development of national income accounting is a leading instance of this point. Yet economics and econometrics, though always second cousins, have never married. Rather, they have developed sufficiently separately to retain distinctive methods and styles. Hence the Econometric Society at its regular conferences in different parts of the world divides its programmes into economics and econometrics. Some resent this but no one could claim it to be an absurd division.

Growing specialization within economics does not affect only econometrics and is a result of the huge growth of the discipline and the proliferation of journals. Economists now tend to be development economists, labour economists, game theorists or specialists in other fields. The generalist hardly exists. Also, as a general category, economist is now more like architect than man of letters. It defines a profession. And clearly the number of people who can be said to earn a living by the practice of economics has increased massively since the Second World War. The total cannot easily be known but 150,000 worldwide would certainly not be an absurd guess.

That raises an obvious question: how successful is economics in the late twentieth century? In a crude ecological sense, economics is undoubtedly successful. It is reproducing itself with an awesome efficiency. However, on the test of its power as a problem-solving method, the success of economics must be regarded as more questionable. Indeed since the mid-1970s the subject has spent much of the time writing itself out of policy relevance. This is particularly marked with macroeconomics, where policy ineffectiveness theorems are central to a dominant mode of thought. It is true that the subject has always recognized that policy may be unable to improve on the market. With Adam Smith, however, that view owed more to scepticism about the usefulness of policy interference than to a faith in the perfection of unregulated markets as such. Recently, however, a faith in the perfection of markets, sometimes, but not always, qualified by the recognition of imperfect information and similar problems has attracted many and enthusiastic adherents.

The sometimes unfortunate effects of ill-considered deregulation exercises in the west may be blamed on that intellectual fashion. Of hugely greater importance in this context, however, has been the general failure of economics to provide useful modelling and advice for the former Soviet economies of eastern Europe following the fall of communism. That reform, which has required the creation of markets more than their liberalization, found economists more or less empty-handed. The result has too often been a flood of inconsistent advice derived from partial models based on western economies. It is too soon to write the history of the reform of former Soviet economies and economic thought but it must be hoped that some deep lessons will have been learned from an experience which underlines the fact that economics, for all its modern reach and power, remains a subject embedded in particular historical experiences and presumptions.

Christopher Bliss

Nuffield College

Akerlof, G.A. (1984) An Economic Theorist’s Book of Tales, Cambridge, UK.

Arrow, K.J. (1963) Social Choice and Individual Values, 2nd edn, Chichester.

Atkinson, A.B. and Stiglitz, J.E. (1980) Lectures on Public Economics, Maidenhead.

Begg, D.K.H. (1982) The Rational Expectations Revolution in Macroeconomics, Baltimore, MD.

Blanchard, O.J. and Fischer, S. (1989) Lectures on Macroeconomics, Cambridge, MA.

Cantillon, R. (1755) Essai sur la nature du commerce en general, Paris. (English edn, ed. H.Higgs, London, 1931.)

Cournot, A. (1927), Researches into the Mathematical Principles of the Theory of Wealth, (English translation by Nathaniel T. Bacon), New York.

Fudenberg, D. and Tirole, J. (1991) Game Theory, Cambridge, MA.

Hicks, J.R. (1969) ‘Preface and a manifesto’, in K.J.Arrow and T.Scitovsky (eds) Readings in Welfare Economics, London.

Corrected text from J.R.Hicks (1959) Essays in World Economics , Oxford.

Kaldor, N. (1939) ‘Welfare propositions of economics and interpersonal comparisons of utility’, Economic Journal 49.

Keynes, J.M. (1973) The General Theory, Royal Economic Society edn, London.

Machina, M.J. (1987), ‘Choice under uncertainty: problems solved and unsolved’, Journal of Economic Perspectives 1.

Magnusson, L. (1994) Merchantilism: The Shaping of an Economic Language, London.

Petty, W. (1927) Papers, ed. H.Lansdowne, London.

Pigou, A.C. (1955) Income Revisited, London.

Quesnay, F. (1759) Tableau economique, 3rd edn, Paris. (English edn, London, 1972.)

Robbins, L. (1952) The Theory of Economic Policy in English Classical Political Economy, London.

Schumpeter, J.A. (1954) History of Economic Analysis, New York.

Sen, A.K. (1979) Collective Choice and Social Welfare, Amsterdam.

Smith, A. (1981 [1776]) An Inquiry into the Nature and Causes of the Wealth of Nations, ed. R.Campbell and A.Skinner, Indianapolis, IN.

Veblen, T. (1904) The Theory of Business Enterprise, New York.

This is the complete article, containing 3,951 words (approx. 13 pages at 300 words per page).

View More Summaries on Economics