Which aspects of economics are most relevant to history? – An essay Do economists make assumptions about human behaviour that are useful to historians? And do historians need to count and measure things? In 1962, Harry Court, a distinguished practioner ventured into a definition of economic history, noting that "economic choice form[ed] the centre of economic history", and citing a common definition of it as "that part of history which requires a knowledge of economics for its understanding ". Singularly, the 1960s were a period during which economics went through major refinements, becoming an increasingly technical and abstract subject whose various concepts and ideas are more or less relevant to the study of economic and social history. However, we have the strong belief that some underpinning concepts used by economists can be helpful to an historian, and may help the latter to refine his/her ideas about historical method more precisely. In such a context, human behaviour and motivation become key issues. In first instance, we shall briefly review some of the basic assumptions that use to be made by economists on human behaviour and individual rationality. We will then see how such assumptions are integral part of a system of economic common sense that may relieve historians from the possible bias of misleading 'fallacies' in interpreting history. Finally, we shall examine to what extent economics provide useful tools for talking about inconvenient objects of the past, including the goodness of statistics in history. 1. Economic assumptions about human behaviour and rationality A series of assumptions about human behaviour and motivation are to be found at the heart of the orthodox economic analysis, that largely reflect the philosophical ideas of early nineteenth century liberalism. The classic behavioural assumptions are that economic agents are individualistic (concerned essentially with their own welfare, without an over-riding need to contribute voluntarily to collective well-being), and that they are rational, in that they are able to calculate the costs and benefits of choices correctly, and will select the one which will maximise their welfare or 'utility'. Naturally, economists would accept that such assumption are weak (that is, they are not very prescriptive or limiting), but on the other hand, they are powerful (they can be applied to most cases, most of the time, and provide a good basis for explaining actions and motivations).

From these foundations, economists make broad assumptions about the circumstances in which people make effective choices as informed, autonomous decisions made under the conditions of a competitive market, that is, a non co-operative one a priori . However, asserting that people ought to be able to make decisions motivated by a free, rational choice based on perfect information does not necessarily mean that they always do so. That a bet does not come off does not always indicate it was a poor bet. Indeed, economists are adept at constructing models of the world which project what would be the case if their assumptions were correct. They can then identify where reality differs from theory, and investigate which aspects of reality prevent the predicted outcome from occuring. We know that very few choices are made under the perfect circumstances often assumed, because competition is not perfect, because women could not vote, or because large groups of people are dominated or oppressed, or held in ignorance by some form of informational asymmetry. However, such models are very useful in that they may explain part of the choice, or the direction in which cumulative choices were tending over time, or simply the reason why apparently ineffective or inappropriate choices were made. Game theory provides a good illustration of the economists' thought process, where players are always assumed to be individualistic and rationalistic, and where a wide range of situations have been identified that may produce outcomes which diverge from optimality, or where reality may yield some apparently irrational behaviours. Consequently, the theoretical framework is refined to provide a framework in which previously seemingly irrational choices become compatible with the reality (e.g. co-operation through a trigger strategy in a repeated prisoner's dilemma). Whatever the context, the economist normally supposes that people do things for reasons, not habitually or accidentally or mistakenly. Furthermore, the economist supposes that what people do spontaneously is probably good. The economist is no Dr Pangloss, believing that all is for the best in the best of possible worlds. Yet it must be admitted that he has Panglossian tendencies, to which he will occasionally surrender, usually for the sake of the argument. Some will complain about the use of economic theory that it supposes people to be reasonably close calculators. The objection seems a reasonable one: we see ourselves failing to make the best decision about which food to buy or whether to change jobs. Since most of us wander in a fog of

indecision, the bright sunlight in which the rational man strides is hard to credit. However, the brighteness needed for rationality is easily exxagerated. In fact, a crude decision is rational if information to make a more subtle one is expensive (information asymmetries are large). The postulate of rationality in economics is not really a postulate in the Euclidean sense, an allegedly indubitable premises from which irrefutable conclusions can be drawn. It works merely as a working proposition. The objection that economics 'assumes' rationality is naïve. The facts can always speak to the assumption, pro or con. 2. Economic common sense and how it should overcome the bias of misleading 'fallacies' in the interpretation of history Now, as put by the great Swedish economist and historian Eli Heckscher: "...if economic theory is at all what it ought to be, its reasonings should apply to economic life as such, and consequently to that of all ages. No doubt much remains to be done in the field of economic theory; but an attempt might at least be made to utilize economic theory for the work of economic history ." [Heckscher, 1930] All the assumptions made by economists that we underline supra are part of a wider ensemble forming an economic "common sense" which represents a self-standing system of thinking and arguing, rather than a collection of settled conclusions about the world. And this is precisely where such a common sense might be useful to historians in interpreting the human past, and avoiding the trap of various biases that tend to yield historical tautologies to the detriment of consistent explanations of the past. However, a difference exists in the act of applying economic theory, which is different from the engineer applying his knowledge of the strength of materials to a particular bridge or a barrister applying her knowledge of a precedent to a particular case. Economic theory by itself has very little to say about the world. Economists sometimes say "theory tells us" such and such - for instance, that free trade is desirable. The expression is as foolish as the parallel expression, which sometimes escapes the lips of historians, "the facts tell us". Mute facts unarranged by human theories tell nothing; human theories unenlivened by facts tell less than nothing. The method of economics applied to history is not necessarily mathematical. It is not far removed from common sense, and consists in arguing by analogy, thought experiment, dialogue, arguments a fortiori, a contrario, a definitione .

The study of the economic and social past is redundant with interpretations that are based on assumptions made by historians, and likely to become "fallacies" where they do not turn out to be great truths. These fallacies tend to bias the way history is analysed and interpreted, and economics provide a useful framework to get rid about these deviations. • For example, historians are sometimes tempted to think, by opposition to the economists' view, that everything is for the worst in the worst of possible worlds, except in the wise world of the government. This is a legal or political fallacy at which economists will oppose the primary instinct of doubting about the ability of the present government to do better than the past ones. • Similarly, the secondary preoccupation of economists for distributional effects might help to overcome the geographical fallacy consisting in attributing the responsibility of a wider trend (say a fall in the world price of Cotton textiles) to a particular country (as it was the case when Indian competion was once thought to be the cause of British cotton industry decline). • Finally, it is sometimes supposed that institutions do not adjust to market conditions, because the former are seen to be rigid and powerful and the latter weaker. (e.g. saying that rising the output of the British cotton textile industry from 1815 to 1860 would spread fixed costs over a larger output and therefore reduce costs per unit would seem ugly to an economist as the number of institutions is not given, and the number of firms would grow with the total fixed costs). Of course, there is no easy conclusion to the economic investigation of a particular historical event, not least because issues themselves might be very complex. 3. Talking about inconvenient objects of the past: the tools of economics AND the goodness of statistics in history It may be argued that economic theory reduces humankind to a diagram or a mathematical equation. But there seems to be no point in ignoring the formal side of economics, because it can be of great assistance in trying to talk about inconvenient objects of the past. A set of tools is made available to historians by economists that prove useful in studying the past. Moreover, economic historians cannot get on without having to count, providing a justification of the goodness of statistics in history.

The geometry of budget lines, for example, provide a useful framework for representing historical events such as, for instance, the effect of free state schools on the choices faced by parents. The figure shows the combinations of education and other goods that a given family could consume in 1869. In the absence of free schooling, that family had to stay in the shaded area below their budget constraint curve: they could have more education only at the cost of giving up quite a lot of other goods. With the introduction of free schooling, the family can now reach the "free offer" point, taking the free offer of schooling and devoting all their income to other goods. Moreover, they can either accept the offer of free schooling or stay on the old budget line: they cannot do both. They are better off but paradoxically, consume less education after 1870. The above

figure adds conviction to the verbal argument. It is literaly a figure of speech. Also, looking at the "very poor" point, figure shows that very poor people were induced to take more education, not less. Also shows that state subsidies to private education had unambiguously induced more education to be purchased (dashed line represents a more favourable trade off). What the economic theory says therefore is that free education replaced a system that had made buying additional education more attractive (plus free offer no really free -> tax so new point probably below free offer ON the budget constraint). How it was financed entails further economic discussion. We can see that the economist's budget line is a powerful aid to the historian's imagination. Such is the theory of the rent, predicting that the renters of something to be used commercially will pay what they can earn from it. Such a reasoning has been used by an American historian, Roger Ransom, to evaluate the social returns from public transport investment, based on a case study of the Ohio canal. He reasoned that lower costs of transport would have raised the rent that farmers were willing to pay for land close to a new canal. By estimating the prices of land before and afte the canals, Ransom could estimate the elusive social returns. The desirability of something can therefore be measured from the benefits it procures, and such a rationale has been used in many other historical studies on slavery, land tenure...etc. Another typical economic tool used by historian is the "supply and demand" curves (such as in the IS-LM model), together with the notions of partial and general equilibria. For example, a supply curve falling in the cotton textile industry will explain a rise in prices for a constant or growing demand, drawing attention to factors likely to be the cause of such a shift (foreign competition, lower incomes, cheapening of transport costs...etc.). Supply and demand at its most concrete provides a narrative framework for the history of a particular industry, and economic

historians are natural experts in supply and demand, and have entered such histories with enthusiasm. In the meantime, the theory allows for useful simplifications as one cannot deal with everything at one (this is called bounded rationality in economics!). ---------But economic history is equally a counting subject. Sir John Clapham, a British contemporary of Heckscher trained by Alfred Marshall in the economics of the day remarked that "every economic historian should...have acquired what might be called the statistical sense, the habit of askig in relation to any institution, policy, group or movement the questions: how large? how long? how often? how representative? " [1930] The goodness of statistics in history is best shown by an example: [?] BOSWELL: Sir Alexander Dick tells me he remembers having a thousand people in a year to dine at his house... JOHNSON: That, Sir, is about three a day. BOSWELL: How your statement lessens the idea. JOHNSON: That, Sir, is the good of counting. It brings everything to a certainty, which before floated in the mind indefinitely. • Statistics make their numbers in aid of a Science of Behaviour. The economic historian finds it easy, for instance, to resist the historian's urge to collect every single scrap of evidence, because he is trained to sample, and knows that statistical theory shows that, oddly, the proportion does not matter to the accuracy of a sample. [Conventionally, a sample as low as 30 is said to be 'large' for purposes of estimating an average. So long as it is taken at random - entirely without system or bias - accuracy of sample varies inversely with the square root of the absolute number in the sample, a crucial formula for both the economist and the economic historian (standard error; x%, y% of the time). An application of the idea of average and standard deviation can explain, for example, the benefits of diversification in land holdings or industrial structure at a particular period of history.] • The most imposing quantitative tool of the economic and social historian goes by the peculiar name of regression. It amounts to no more than the drawing of a good straight line through a scatter of points. Regression analysis would ask, for example, in the view of the way American immigration and the wage actually behaved historically, the

best guess for the value of (a) and especially (b) if American immigration were defined as: immigration = a + b(wages) It would allow the historian to identify and measure the importance of factors that influenced immigration in the US around, say, the 1880s. Econometric modelisation allows historians to squeeze information out of what the past has left. It is a fitting of economic theory to the past using the most sophisticated methods for extracting the controlled effect of each of many variables. • Finally, the importance of simulation is especially plain in economic history. Regression is profligate with observations that the economic historian is often hard put to acquire in greater in quantity. Simulation allow for the replacement of data by carefully chosen assumptions in trying to fit the curves. The simplest example of simulation in economic history is cost/benefit analysis, where all the costs and benefits of undertaking a particular project are carefully listed, guided with a light hand by economic theory. What is being simulated could be, for example, the decision facing the cotton textile manufacturer contemplating the installation of ring spindles in 1900 in Britain,the economic historian supplying the list of possible costs and benefits with their estimated magnitude, based on engineering specifications, construction expenses, profit figures, training manuals and other bits of data. CONCLUSION.: shortcomings of economics to making sense of history? Perhaps the most important lesson that (economic) historians can learn from economics is that social sciences in general inevitably make some behaviourist assumptions about their human subjects. Historians may certainly dispute the assumptions that orthodox economists make, but if they do, they will need to replace them with something else. The use of economic theory in [historical economics], then, are methodologically broad. The theory is the refined common sense descended from Adam Smith, and spoken in many dialects. The economists' way of thinking has been broadened by using it to talk about large and inconvenient objects in the past. [Historical economics] is theoretical, but its theory is merely a way of conversing, a rich and useful way.