I do not need to say very much about Keith Hancock’s work in economics, as the Festschrift edited by Joe Isaac and Russell Lansbury provides both a comprehensive survey of his writings and a bibliography that is complete down to 2004. ^{Footnote 1} As someone with a joint honours degree in economics and history and a PhD supervised at the London School of Economics by the great Henry Phelps Brown, ^{Footnote 2} as a lifelong scholar of industrial relations and a former Deputy President of the Australian Industrial Relations Commission, Hancock embodies the multidisciplinarity and pluralism that once characterised Australian economics but has now succumbed to the increasing formalisation and Americanisation of our profession. ^{Footnote 3} Unless things change, rapidly and substantially, we shall not see his like again.

Economics is unique among the social sciences in having a single monolithic mainstream, which is either unaware of or actively hostile to alternative approaches. I offer three pieces of evidence in support of this claim, one of them anecdotal (but in my opinion quite convincing) and two documentary. The following is the anecdote: in 2007, I interviewed the eminent Austrian economist Kurt Rothschild, then 93 years old, and himself a long-term advocate of pluralism. ‘The present situation in economics’, he told me,

He might have been embellishing the precise numbers, but the principle is very clear, and it is confirmed by the documentary evidence.

First, there is the recent volume on the post-war history of the social sciences, edited by Reference Backhouse and FontaineRoger Backhouse and Philippe Fontaine (2010). ^{Footnote 4} At the core of this book are six chapters on the individual disciplines, written by international authorities in their respective fields. The editors contribute the chapter on economics, which turns out to be the only social science with a single, unified mainstream. There is no such thing in psychology, political science, sociology, social anthropology or human geography, and even in economics, it is a post-1945 phenomenon (see Reference Morgan and RutherfordMorgan and Rutherford, 1998). My second piece of documentary evidence comes from Oxford University, where in a recent issue of the alumni magazine, the head of politics, Stephen Whitefield, described his department in the following terms: ‘We are self-avowedly pluralist in our teaching and research with enough of us to operate on the zoo principle – two of everything’. ^{Footnote 5} I reported this statement to the readers of the Heterodox Economics Newsletter, a free electronic journal with many thousands of subscribers in many countries, and asked whether there was an economics department, anywhere in the world, that could claim to operate on the same principle. Unsurprisingly, I received not a single reply.

In this lecture, I shall present a case for pluralism in economics, derived from the complex and ceaselessly changing nature of the world in which we live. I use the indefinite article quite deliberately. Indeed, there would be something rather paradoxical about a claim to provide the case for pluralism. I am sure that this is only one of the many possible arguments that might be put forward, and that others would do it differently. ^{Footnote 6} Towards the end, I shall allude briefly to the work of Gillian Hewitson and Therese Jefferson, who stress gender issues, and Tim Thornton, who emphasises questions of ontology derived from the literature on the philosophy of science. For the moment, I shall just refer you to the 20 arguments for pluralism in economics that are listed in a recent book by Reference Garnett, Olsen and StarrRob Garnett et al. (2009).

I shall concentrate on only two arguments. First, economic reality is very complicated, so that the questions that economists ask are therefore inherently difficult, and it is unlikely that they have simple answers. Since no theory can consider all relevant factors in any particular economic context, this establishes a strong prima facie case for pluralism. Second, economic reality is fluid and subject to continuous change, so that the quest for a single, ‘general’ theory applicable to human behaviour in all societies, at all points in time, is a delusion. ^{Footnote 7} To illustrate the case for pluralism, I shall refer to the unfortunate consequences of denying it for the future of both economic history and the history of economic thought (HET). I shall also draw on the recent history of the global economy, the problems encountered in understanding it solely in terms of mainstream macroeconomics, and the dangerous policy implications that have been drawn from mainstream theory.

At this stage, I should probably explain precisely what I understand by mainstream economics: formal modelling of the behaviour of maximising individuals with unbounded rationality, operating in markets that have strong tendencies towards equilibrium – in Reference SolowRobert Solow’s (2008) three-word summary: ‘greed, rationality, and equilibrium’. A glance at the contents of any recent issue of the Economic Record will suffice to persuade you that this is not a straw man. Incidentally, I am not at all convinced by commentators such as Reference Colander, Holt and RosserDavid Colander et al. (2004) and Reference DavisJohn Davis (2008), who maintain that the mainstream is fragmenting, with a ‘cutting edge’ that no longer corresponds to the outmoded Solow definition; but to deal adequately with that would require another lecture (see Reference LeeLee, 2012).

Mainstream economics is vulnerable to criticism on every count, as Reference Keynes and KeynesKeynes (1951) observed back in 1924 in his obituary of Alfred Marshall:

In a footnote, he invoked the authority of Max Planck, who ‘once remarked to me that in early life he had thought of studying economics, but had found it too difficult!’ Keynes did not take this literally, as Planck ‘could easily master the whole corpus of mathematical economics in a few days’. However, there was a deeper truth in his statement:

Similar views have been attributed to Noam Chomsky, for whom ‘the human condition seems not to be subject to the laws of nature’. Chomsky’s explanation is that the hard sciences keep to quite simple systems:

Returning to mainstream economics, the formal model works (at the micro-level, as we shall see, things are even more complex in macroeconomics) only so long as our three assumptions can be maintained. First, individual agents attempt to maximise a well-defined objective function subject to well-defined constraints. Second, they have unbounded rationality. Third, these markets display strong tendencies to equilibrium. For some problems, this is an appropriate set of assumptions. For the most part, however, they do not describe ‘the world in which we actually live’ (Reference KeynesKeynes, 1936: 3).

In the first place, the prevalence of fundamental uncertainty means that the relevant objective functions and (especially) the constraints cannot be known with certainty, or even with certainty-equivalence. To quote Reference KeynesKeynes (1937) once again,

Second, rationality is necessarily bounded, since the human brain cannot cope with all the information, or make all the decisions, that would be required to make unbounded rationality possible. Hence, agents use routines, habits, rules of thumb and conventions (Reference SimonSimon, 1957). ^{Footnote 8} Third, in a world characterised by circular and cumulative causation and, in consequence, by path-dependence, the very notion of ‘long-run equilibrium’ becomes unclear, as was recognised by the young Reference KaldorNicholas Kaldor (1934: 122–137). As Reference KaldorKaldor (1972) later realised, this problem also has important macroeconomic implications, to which I shall return shortly (pp. 1237–1255).

There is an additional problem, which casts further doubt on the relevance of mainstream economic theory: markets are necessarily incomplete. This is obvious in the (important) case of insurance, where the question once asked of Plato (‘who will guard the guardians?’) might be reformulated as ‘who will insure the insurance companies?’ Not, evidently, other insurance companies. The only way of avoiding a rather nasty infinite regress is to introduce the state as insurer of last resort; with hindsight, it seems that the bailout of American International Group (AIG) at the height of the Global Financial Crisis in late 2008 did more to prevent a major economic downturn than any other single act by the US government. Left to their own devices, ‘greed, rationality and equilibrium’ would not have sufficed to avert a global economic disaster.

In sum, the world as seen by mainstream economics is simple enough for one type of theory to do the job. The real world, however, is very much more complicated, as is recognised by the philosopher of science, Daniel Hausman, who acknowledges the frequent failure of mainstream economists to apply the scientific method to their subject, and the disappointing results that are often obtained when they do. Orthodox economists make totally unrealistic assumptions; they hedge their bets with unfalsifiable ceteris paribus clauses; they rarely test their basic theories, and ignore evidence that casts doubt upon them, and they waste their time developing useless, abstract mathematical models. To a very considerable extent, Hausman argues, these failings are due to the peculiar nature of the economist’s subject matter. He identifies

I shall have almost nothing to say, in the time available to me, about the first of these problems, but I shall elaborate on the second and third. Hausman, who is much more sympathetic to mainstream economics than I am, does not draw what seems to me to be the inevitable conclusion from his argument concerning the complexity of the economic universe: several (perhaps many) types of theories will be required in order to do it justice. This conclusion is reinforced by the non-reducibility of macroeconomics to microeconomics, which is denied by mainstream macroeconomists as a consequence of what I have termed, in a recent book, ‘the microfoundations delusion’ (Reference KingKing, 2012). By this, I mean the mistaken claim that all propositions in macroeconomics can be reduced to microeconomic propositions, that is to say, to statements about the behaviour of individuals. A very similar principle has been advocated for the life sciences by Richard Dawkins, who describes it as ‘hierarchical reductionism’. As an example,

There are two fundamental problems with this example of hierarchical reductionism. First, since cars have social, political, economic and cultural significance, it is not possible to infer all the properties of the car from a complete knowledge of its parts. To deny this involves a fallacy of composition. Second, changes in the social, political, economic and cultural context in which cars are driven frequently affect not just the car as a machine but also some or all of its parts: fins are added to boost sales and airbags to satisfy concerns about safety. Thus, causation runs downwards, from the larger to the smaller units, and not just upwards from the smaller to the larger, as Dawkins maintains.

In economics, the fallacy of composition and the phenomenon of downward causation constitute additional, and very important, sources of complexity, which mean that the microeconomic problems of fundamental uncertainty and path-dependence take on an additional macroeconomic significance. In my recently published book, I identify several examples of the fallacy of composition in macroeconomics, including the well-known paradox of thrift, the paradox of liquidity (or deleveraging) and the Kaleckian paradox of costs (Reference KingKing, 2012: ch. 3). First, and most obvious, is the paradox of thrift: a decision by any individual to save a larger proportion of her income will lead to more saving by that individual, but (in the absence of increased investment) this will not be true of an increase in everyone’s savings propensity, which will simply reduce their incomes and leave the volume of aggregate saving unchanged. Second, a realistic monetary theory must allow for the paradox of liquidity. If any individual financial company wishes to increase its liquidity, it can always do so (at a price), but if all financial companies attempt to do so, the consequence will be a reduction in aggregate liquidity and (in the absence of government intervention) the real possibility that the whole system will collapse. The Global Financial Crisis that began in September 2008 provided a dramatic example of this principle. Third, the Kaleckian paradox of costs is rather similar: a wage rise is very bad news for any individual capitalist, but it may be good news for them all, taken together, if the consequent rise in consumption expenditure raises the level of economic activity and thereby increases aggregate profits. In practice, this may or may not be the outcome: it all depends on the values of the relevant parameters. The microfoundations delusion denies these paradoxes. It also poses serious difficulties for economic policy, as I shall suggest later.

Before moving to the second part of my argument, the principle of historical specificity, I will mention two of the important issues that the monolithic mainstream cannot satisfactorily deal with but a pluralist heterodox economics can. The first is the diversity of markets, as analysed by Reference PraschRobert Prasch (2008), an institutional theorist. Prasch develops a set of Weberian ideal types of markets, distinguishing markets for commodities – ‘non-status goods used in everyday consumption’ – from those for credit, assets and labour. He places special emphasis on the peculiar characteristics of labour markets, which are very different from other markets, not least in the long-term nature of many employment relationships (Reference PraschPrasch, 2008).

The second theme is the importance (and complexity) of power in economics, viewed much more broadly than it is in the narrow conception of ‘market power’ that exhausts the mainstream treatment of the subject. Here, Reference RothschildKurt Rothschild (2002) rightly insisted that

There was, he noted, an element of reflexivity here: the ideological preference of powerful wealthy interests for a neoclassical theory that did not look too deeply into the sources and ramifications of their own power reinforced the neglect of power by mainstream theorists, not least by channelling research funds their way. ‘Extremely formulated’, Reference RothschildRothschild (2002) concluded, ‘one could say that societal power promotes the study of models of powerless societies’ (p. 440).

These two issues came together in a recent dispute over the operation of the labour market that will have intrigued Keith Hancock. This was the controversy over the chronically low pay in the community sector that was argued out before Fair Work Australia in 2010. The mainstream economist who testified as an expert witness for the Australian Industry Group insisted that there was only one economic viewpoint – the orthodox viewpoint – on the question of the undervaluation of the labour of any occupational group. The heterodox economists who testified on behalf of the Australian Services Union argued that this made it impossible for them to deal adequately with social structures, social roles and power relations, leading them to deny the possibility that apparently gender-neutral market institutions might in fact be responsible for systematically undervaluing the commodities produced by women (Reference Austen, Jefferson and PrestonAusten et al., 2013; Reference Austen and JeffersonAusten and Jefferson, 2012). For once, there was a happy ending: Fair Work Australia disagreed with ‘the economic approach’ and awarded the community sector workforce a big pay rise.

The second part of my case for pluralism is neatly summarised by Lance Taylor, who describes himself as a ‘structuralist’ macroeconomist: ‘The economy rests on changing social relations – all attempts to describe it in terms of timeless constructs like Newton’s laws, Einstein’s field equations, or Feynman’s diagrams are bound to fail’. ^{Footnote 9} I shall comment briefly on three examples of this principle of historical specificity; there will certainly be many more. One comes from the Marxian literature, and two from post-Keynesian economics. As Paul Sweezy has shown, Karl Marx not only insisted that the economic laws of capitalist society were quite different from those that applied to pre-capitalist and post-capitalist modes of production. He also distinguished two distinct stages of (early) capitalism, which he termed ‘manufacturing’ and ‘modern industry’. When manufacturing took over from traditional handicraft production, there was little change in the methods and instruments of production. It was the organisation of the labour process that was transformed, with the establishment of capitalist domination and a much greater degree of specialisation and division of labour. Productivity rose significantly, but in the context of an economic system that remained essentially conservative. Only when handwork was replaced by machinery, in the subsequent stage of modern industry, was the ‘constant revolutionising of the means of production’ described in the Communist Manifesto able to begin. This was extremely important since it was only in modern industry that economic growth accelerated markedly and the organic composition of capital began to display a powerful tendency to increase, placing strong downward pressure on the rate of profit and strong upward pressure on the reserve army of the unemployed, and also giving birth to the classical business cycle (the first instance of a downturn in which Marx saw in the crisis of 1825) (Reference Sweezy and EaglySweezy, 1968: 107–126).

My second example is Victoria Chick’s analysis of the evolution of the banking system and its implications for the theory of saving, investment and interest. Chick distinguishes five stages in the development of the banking system, culminating in the post-1970 phase of liability management. Each successive stage increased the independence of investment decisions from saving decisions that lies at the heart of Keynesian macroeconomics, with significant implications also for the analysis of inflation and interest rates (Reference Chick and ChickChick, 1992: 193–205). The third example comes from Chick’s teacher, Hyman Minsky, whose ‘financial instability hypothesis’ has often been cited by writers on the Global Financial Crisis. It has been much less frequently noticed that shortly before his death in 1996, Minsky himself had identified a new stage of financial innovation, which he termed ‘money manager capitalism’. The dominant players in the previous stage of US capitalism were investment bankers and their customers, typically large industrial corporations. Now, however, the ‘new institutional structure is dominated by mutual, pension, hedge, money market mutual funds, university endowments, sovereign wealth funds and other leveraged institutional investors’ (cited in Reference Nersisyan and KingNersisyan, 2012: 409). This process is often also described in the post-Keynesian literature as ‘financialisation’ (Reference Hein and StockhammerHein and Stockhammer, 2011; Reference WrayWray, 2009). It has reinforced the pressure on non- financial companies to focus on short-term profit maximisation to the exclusion of all other goals, and has further increased the fragility of the financial system and its potential instability. Moreover, Minsky himself recognised that his own analysis required constant revision in the face of relentless financial innovation.

You may, or may not, find these three pieces of analysis as enlightening as I do. However, I hope you will agree that something would be lost if economists were unwilling, as a matter of dogma, to acknowledge the principle of historical specificity. One important practical conclusion does seem to follow. As Keynes’s biographer Reference SkidelskyRobert Skidelsky (2009: 99) has argued, students of economics need to learn more history – and less mathematics. This is a live issue in the United Kingdom, where a recent conference supported by the Government Economic Service, the Bank of England and the Royal Economic Society heard calls for ‘a new eclecticism’ in the teaching of economics, with more emphasis being placed on economic history and the HET. ^{Footnote 10} Time will tell whether the ‘business as usual’ attitude of the great majority of mainstream economists really is changing, in Europe or in Australia. ^{Footnote 11}

As I said at the beginning, this is only one case for pluralism. There could be many more; I have not even mentioned Austrian economics, for example, and I could have said a lot more about behavioural economics, old and new (see Reference Jefferson and KingJefferson and King, 2010–2011).

If this lecture was being given by feminist economists like Gillian Hewitson or Therese Jefferson, they would have stressed two additional questions: the gender dimension to economic activity, which I have only touched on, and the methodological issues that are implicit in what I have been arguing, especially the questions of ontology stressed by thinkers in the post-structuralist and critical realist traditions (see Reference HewitsonHewitson, 1999; Reference JeffersonJefferson, 2006).

They would also emphasise the need for plural methods to be used in empirical research, a procedure sometimes described as triangulation. As Reference Boumans and DavisMarcel Boumans and John Davis (2010) explain, this is

This shades into methodological pluralism:

Formal econometric techniques are not to be rejected, but they should not be regarded as the only legitimate means of conducting economic research. In his forthcoming La Trobe PhD dissertation, Tim Thornton summarises some of the powerful methodological arguments in favour of pluralism in economics that can be derived from the philosophy of science. ^{Footnote 12}

I believe these questions to be important, for several reasons. I shall mention three of them: hostility towards pluralism in economics upsets the neighbours, induces amnesia and encourages the infliction of serious self-harm. First, consider the neighbours, who have been caused some annoyance and distress by the ‘economics imperialism’ project. This rests on the claim, in Reference LazearEdward Lazear’s (2000) proud words, that

Moreover, there is much more in this vein. The Backhouse and Fontaine volume that I cited earlier provides ample evidence that the economics imperialism project has very largely failed, despite the fears expressed by radical critics like Reference Fine and MilonakisBen Fine and Dimitris Milonakis (2009). However, it must have had the effect of making interdisciplinary cooperation more difficult: one is more likely to be suspicious of visitors if one suspects them of having aggressive designs on one’s territory. Reference LazearLazear (2000) himself reports that economists are increasingly able to

Such pretensions have certainly caused a lot of wasted effort, with so-called Rational Choice Marxism as a prime example. Who knows what a subtle thinker such as Jon Elster might have produced if he had not wasted 20 years on a vain attempt to reduce Marxian political economy to neoclassical economics? (see Veneziani, in press). In addition, there is one very clear example of a near neighbour that has been seriously disturbed by the imperialist expansion of mainstream economics. This is economic history, where the triumph of ‘cliometrics’ has imposed a single, monolithic orthodoxy derived from mainstream economics on what was once a vibrant, multidisciplinary and (above all) pluralistic endeavour. ^{Footnote 14} Henry Phelps Brown would not have been impressed.

Second, the mainstream is subject to amnesia, as is demonstrated by the steep decline in recent decades in the status of the HET. Why would technically advanced practitioners of a rapidly progressing science take any great interest in ‘ancient errors made by dead white males’? Roger Backhouse has documented the decline in HET teaching in the United Kingdom since the introduction of that country’s Research Assessment Exercise (RAE) in 1989 and the pressure (direct and indirect) on individuals not to do HET research on the grounds that it was unlikely to ‘count’ for RAE purposes. ‘One respondent went so far as to say, “I have been told that my interest in HET is “unhelpful” to my department’s RAE effort” and that his teaching load had been increased and his research budget cut’ (Reference BackhouseBackhouse, 2002a: 86, Reference Backhouse and Weintraub2002b). The situation was likely to get worse, Backhouse suggested, not better, as existing HET specialists retired and were not replaced; 80% of those surveyed were over 40 years, the mode being 50–60 years (Reference BackhouseBackhouse, 2002a: 89). In sum,

Similar fears have been expressed in Australia, where they were strengthened by the abortive attempt of the Australian Research Council to eliminate HET research from the category of ‘Economics’ and to move it to ‘History and Philosophy of Science’ (Reference Kates and MillmowKates and Millmow, 2008).

Four years later, Backhouse regretted that his 2002 assessment ‘ended on a note that is more negative than I intended’, and suggested that ‘the flow of good PhDs in HET is not much different from what it ever was, and that they are mostly coming out of History departments, not economics, or perhaps philosophy’. ^{Footnote 15} This was just as well, since in the Economic Journal’s list of 156 dissertations passed by British universities in 2004–2005 and 2005–2006, classified by topic, there was not one in HET (Anon, 2006). I myself think it is very important that HET continues to be taught, and researched into, inside Economics departments. There is a dialectical reason for this: as that great subversive (Lord) Lionel Robbins once wrote, the study of the history of our discipline demonstrates the great diversity of approaches that have been taken to all the important issues that it has confronted. If HET was to be widely taught, that is to say, the case for pluralism would become very clear.

Third, and finally, the insistence that there is only one correct way to do economics is capable of causing real self-harm. This is being demonstrated on a daily basis in the Eurozone, as fiscal austerity leads to stagnation in the more fortunate member states and deep depression in the unlucky ones. This is the most serious practical consequence of the microfoundations delusion that I referred to earlier. It results from what in Germany is known as ‘Swabian housewife logic’, which can induce ‘suicide through fear of death’ (Reference SchulmeisterSchulmeister, 2012). The fallacy was beautifully expressed by Barack Obama, who

As Keynes is said to have remarked, it is better to be vaguely right than precisely wrong. On this issue, Obama is, indeed, precisely wrong. The paradox of thrift is a macroeconomic phenomenon, which is quite irrelevant to individual households, be they Swabian or Washingtonian. However, it does apply to households in aggregate, and also to governments, as revealed by the continuing tragedy of the PIIGSS ^{Footnote 16} in the Eurozone.

The continuing relevance of the paradox of thrift was demonstrated by a controlled experiment – pace Hausman – that began in 2007, before the onset of the Global Financial Crisis: it was called Latvia. Here drastic neoliberal macroeconomic surgery was performed at the behest of the International Monetary Fund (IMF). The patient survived, but it was a very near thing: gross domestic product (GDP) fell by 24.1% between late 2007 and late 2009, and 10% of the working population left the country (Reference Weisbrot and RayWeisbrot and Ray, 2011). This was suffering on a scale comparable to that of the Great Depression – actually more severe, by comparison with any European country. The largest peak-to-trough fall in real GDP in any European country between 1922 and 1935 was in Austria, at 22.5%. Only Canada (a 30.1% decline) and the United States (29.5%) fared worse (Reference MannMann, 2004: 50). Similar treatment is still being inflicted on the hapless European PIIGSS.

Before I conclude, I should perhaps emphasise that my case for pluralism does not entail the total rejection of mainstream economics, which can be seen, as it was by Reference RothschildKurt Rothschild (1999), as one approach to the subject that can be valid and useful in some circumstances. Let me offer one example; since it concerns tobacco control, it is not a trivial one. ^{Footnote 17} Proposals for reducing cigarette consumption include the elimination of all promotional material, including brand names, from cigarette packaging, on the grounds that this would reduce the attractiveness of the product as a social phenomenon, that is, as a source of group identity. This rests on a heterodox argument that goes back to the institutional economist Thorstein Veblen: consumption is conspicuous, and individual preferences are socially determined and interdependent. A second proposal is for a large increase in the tax on cigarettes, to bring their price in Australia up to the European Union (EU) average. This relies on the (mainstream) law of demand: the elasticity of demand for tobacco is known to be roughly −0.4, so that a 50% increase in cigarette prices would cut the quantity smoked by about 20% and save very many lives. This seems to be a very good idea, and it would be quite wrong to reject it purely on account of its origins in mainstream economics. In this case, mainstream economics provides part of the solution to an important social problem. However, it is only one part, and to suggest otherwise would be to impoverish the debate and to endanger public health. Much the same can be said, I suspect, about water pricing in Australia, and about the carbon tax as part of the remedy (if there still is one) for global warming (see, e.g. Reference StilwellStilwell, 2008).

In conclusion, I have to say that I am not at all confident that economics will become more open, tolerant and pluralistic any time soon. The experience of the 2011 Excellence in Research for Australia (ERA) exercise suggests that here in Australia, the reverse is much more likely. The way in which ERA was designed made serious damage to non-mainstream approaches to economics almost inevitable, as Harry Reference BlochBloch (2012), the former editor of the Economic Record, has demonstrated in some detail. This was most obvious in the journal rankings, now officially discarded but still in use as a matter of managerial fiat in many institutions, including my own; for this reason, I shall use the present tense in my discussion of them. There are no A* journals in heterodox economics, and of the two that are ranked A, only one (the Cambridge Journal of Economics) could be described as ecumenical. ^{Footnote 18} In the middle of the exercise, two non-mainstream journals were downgraded, for no obvious reason, from A to B in the case of Feminist Economics and from A to C in the case of the leading institutionalist forum, the Journal of Economic Issues. There could have been no clearer signal of mainstream hostility towards pluralism than this.

To return to Keith Hancock, I imagine that he would agree with much of what I have said tonight, and I strongly believe that his own distinguished career supports the case that I have made for pluralism in our discipline.