Power and Economics

Abstract

Power and economics are not often put together as a topic. Economists—although they regularly deploy notions such as market power and bargaining power—do so unreflectively: they have little, and usually nothing, to say about the concept of power, about what power is, and how to study it. It is, it would seem, either uninteresting or difficult for economists, and in particular mainstream economics, to deal with this notion. There is little about it in the literature of economics; if you look for articles and books about power in economics, you will find very few. There are two interesting books, one by John Kenneth Galbraith and another by Kenneth Boulding, but they were maverick economists.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Power and Economics by Steven Lukes

Power and economics are not often put together as a topic. Economists—although they regularly deploy notions such as market power and bargaining power—do so unreflectively: they have little, and usually nothing, to say about the concept of power, about what power is, and how to study it. It is, it would seem, either uninteresting or difficult for economists, and in particular mainstream economics, to deal with this notion. There is little about it in the literature of economics; if you look for articles and books about power in economics, you will find very few. There are two interesting books, one by John Kenneth Galbraith¹ and another by Kenneth Boulding,² but they were maverick economists.

I was familiar with Galbraith’s ideas while working on my own book on power, Power: A Radical View, which was published in 1974. Working further on a revised and expanded edition published in 2005, I turned to his book on power (published in 1983) and realized how relevant it was to what I was trying to argue, and I now see its continuing relevance to discussing the topic of this volume. Galbraith quotes another economist, Melville Ulmer, who contributed to one of the very few books on power and economics, actually entitled Power and Economics¸ a little Penguin book published 40 years ago.³ Ulmer wrote, ‘Perhaps no subject in the entire range of the social sciences is more important and at the same time so seriously neglected as the role of power in economic life.’ And Galbraith himself says similar things in his book and offers, moreover, the beginnings of an explanation of why reflecting on power might be a problem for economists and, more widely, for anyone trying to understand the role of power in economic life. What Galbraith says, among other things, in The Anatomy of Power is:

Nothing is so important in the defence of the modern corporation, as the argument that power does not exist; that all power is surrendered to the impersonal play of the market; all decision is in response to the instruction of the market. Nothing is more serviceable than the resulting conditioning of the young to that belief.⁴

What caught my interest in Galbraith’s book is the idea of conditioning. Galbraith has three kinds of power. Indeed, it seems that everyone writing about power seems to want to have three kinds of it. (Why should power be triadic?) Galbraith’s three kinds of power are: condign power (relying on threats or negative sanctions), compensatory power (relying on inducements), and conditioned power. He says of conditioned power:

There is a successful expression of power when the individual submits to the purposes of others, not only willingly, but with a sense of attendant virtue. The supreme expression, of course, is when the person does not know that he or she is being controlled. This at the highest level is the achievement of conditioned power: belief makes submission not a conscious act, but a normal, natural manifestation of the approved behaviour.⁵

This thought was close to what I argued when I became involved in what came to be called ‘the power debate’ in the 1970s, which was generated within American political science but spread beyond it and is still alive. I welcome Jonathan Hearn’s contribution to our discussion, since his recent book Theorising Power ⁶ admirably pulls together this debate alongside various other discussions of power, both in the Anglo-American and the European Continental traditions, and makes some pretty sharp criticisms of what I had to say.

I am also a triadic power theorist: my claim was that power should be viewed in three dimensions. C. Wright Mills, in the 1950s, wrote his great book The Power Elite, and his idea that there was a power elite (operating in three domains: the military, the governmental, and the corporate) in the USA was subjected to a healthy dose of scrutiny by Robert Dahl, the great American democratic theorist and political scientist. Dahl asked a very good question: What can be shown empirically about this claim that there is a cohesive elite in power? His answer was: We have to submit it to the empirical test of observability, we have to observe behaviour.

Dahl then came up, in his classic study of New Haven politics, Who Governs?, with the clear and straightforward idea that power consists in prevailing in decision-making where you can observe conflict over ‘key issues’ between actors in situations where there is a clear conflict of interests, those interests being revealed by preferences that are in conflict. The task was to trace the distribution of power, and thus determine whether there was unified power exhibited by some group that could be identified as a ruling elite. Dahl’s conclusion was that what existed in New Haven was pluralism: there was no power elite in New Haven (and he would later maintain, across the nation), since different interests prevailed in decision-making over different issues.

This was behaviourism: the assumption that power is revealed by behaviour in decision-making situations. This approach was, in turn, criticised in a very influential article entitled ‘Two Faces of Power’ in the American Political Science Review. The political science profession widely discussed this article by Peter Bachrach and Morton Baratz, who claimed that power was not only revealed in decision-making, where you have observable conflict over issues in contention, but argued that there is also a second face, which is crucial to consider, that consists in what they (somewhat confusingly) called ‘nondecision-making’. Who decides which issues reach the political arena? Who decides what is decided upon and how? This they called ‘the mobilization of bias’—where some issues are organised into politics and others organised out. The whole issue of agenda control was thus raised by that very short and rather remarkable article, which they then elaborated in further articles and a book, Power and Poverty. Thus, a city or a country could be pluralistic in its decision-making but elitist in its nondecision-making.

I was impressed by this advance in thinking about power, but also dissatisfied. I decided to change the metaphor from faces of power to dimensions, because who could want two dimensions if you could have three? I was really trying to suggest that there is a way of looking beyond the two faces and seeing further and deeper into the phenomenon. So, I claimed that there is a third dimension of power, which has a clear relation to Galbraith’s idea of conditioning, though it goes beyond it. For me the strategy of locating power by focusing on conflicting preferences, as revealed by observable behaviour, seemed inadequate; I was concerned with the question common among Marxist theorists, especially those influenced by Antonio Gramsci and the idea of hegemony, that there is something important to say about power over thought, desires, beliefs, and thus preferences.

I claimed in the little 1974 book that power can also be seen to be at work in shaping beliefs and preferences that can adversely affect people’s interests. This should certainly be seen as itself an unduly narrow way of seeing the topic of power, because it focuses on power’s adverse effects on people’s interests and thus portrays power from a negative or pejorative perspective that assumes that people are better off when free from control by others. That limitation (which I later sought to correct) was path dependent: it resulted because of the way the debate until then was concerned with asymmetric, top-down power, or control, ways in which some are disadvantaged because others prevail. This perspective, as many went on to point out, neglected the many ways in which the power of some over others can be to the latter’s advantage. My concern, however, was with (different) ways in which people’s preferences can themselves be shaped by power.

This question bears directly on the relation between economics and the concept of power. How do interests relate to preferences? The question of how preferences are themselves shaped, rather than being assumed to be just exogenously given, seems to me an issue right at the centre of the discussion we need to have. Economists typically ignore or avoid it, or assume that the answer is straightforward. Hence the favoured expression ‘revealed preferences’. Thus, for instance, George Stigler and Gary Becker write, ‘It is neither necessary nor useful to attribute to advertising the function of changing tastes; advertising affects consumption not by changing tastes, but by changing price.’⁷

My claim, to repeat, was that power is not only to be seen in the first dimension, where you have observable conflict, where the most obvious case is visible and observable coercion—that is, cases where the question of who prevails is accessible to observation. It is not only, in the second phase, as claimed by Bachrach and Baratz, that power consists in suppressing or preventing from reaching the political arena grievances that are kept out of politics. It was always their unshakeable (still behaviourist) claim that these grievances—actual preferences of people whose access to the political decision-making process is blocked by observable mechanisms, either by neglect, but more often by deliberate agenda control, such as co-opting leaders—are, in turn, observable.

The grievances that are prevented from becoming effective political demands are themselves on this account observable. What interested me was the much more perplexing question of situations where people do not have grievances; where you can say that they have interests that do not become preferences, because of their beliefs: the shaping, in other words, of beliefs that can be seen to be adversely affecting people’s interests that are not revealed by their preferences.

As I proceeded with this it seemed to me there were further, more problematic questions that the idea of conditioned power, to use Galbraith’s phrase, raises. Does power thus understood always have to involve intention? Is it possible that power can be exercised by affecting people’s interests, thereby securing their compliance in ways that are not intended? Does the capacity that constitutes power—the power or capacity, in other words, to bring about compliance—always have to be intended? Bertrand Russell briskly defined power as ‘the production of intended effects’ and Max Weber, who wrote about power in a very pioneering and influential way, thought that power always had to be the expression of will. Is that so? I claimed, and still do, that power need not involve intention; that the most effective forms of power can be the result of others complying without the powerful even knowing it or intending it; for example, by merely following the dictates of roles and norms with unintended consequences. When managers in, say, corporations do their jobs and their companies accrue, or fail to accrue, profits, it need not be because the former intend the latter.

Does it necessarily even involve action? Why would we want to say that? Inaction can have significant consequences. Is it not the most effective form of power that others will comply with your interests without your having to lift a finger; without your having to exercise the power that you have, as when your status is enough to secure my compliance with your interests? As you can see, what I was up to was trying to expand and develop the concept of power, to broaden it in ways that end up by making it more and more problematic to study. Power has the remarkable feature of being at its most effective when it is least observable, by actors and by observers alike, which poses quite a neat paradox for the practice of social science.

It is rather obvious that this strategy of extending power’s conceptual reach poses a range of problems, not only for social scientists, but also for anyone seeking evidence for significant, and in this case troubling, claims about how the world works. The kinds of claim I was making have been severely criticised and, over the years, I have tried to meet these criticisms. I conclude by all too briefly addressing three, the second and third of which have been most recently and most trenchantly formulated by Jonathan Hearn.

First, there is an objection which is quite widespread in the literature of the power debate. It registers an aversion to the very idea that people can have interests they do not perceive and acknowledge. The critique, in short, assumes (as economists generally do) that our interests are revealed by our preferences: that revealed preferences express people’s interests. To think otherwise is to be presumptuous, even paternalistic. Who is the observer, let alone the social scientist, to attribute objective interests to subjects who do not recognise them? This is a tangled nest of issues, but here are three examples that should give such critics some trouble.

Consider, first, the repeal of the so-called death tax in the USA in 2001, and the fact that people widely believed that the death tax was against their interests. There is a fine book by Ian Shapiro and Michael Graetz⁸ which documents the way in which this framing of what the estate tax meant had the effect of getting many people to believe that the so-called death tax was going to affect them and was a shocking thing. That is an example of one way in which people could be induced through the framing of an issue to believe something—in this case, that the estate tax was against their interests—which it was not.

In that case, there were anti-tax conservatives actively involved in propagating this highly effective framing of the issue in a way that favoured the interests of the rich and powerful. But such manipulative (i.e., active and intentional) exercising of power is not necessary. Consider the whole issue of the subordination of women, as exemplified in traditional societies, caste societies, or many other patriarchal societies, but also in the beauty myth and the gendered preoccupation with body size and weight in our own culture,⁹ and in relationships where, despite sexual abuse and domestic violence, women can continue to view their abusers in a positive light. Are these not situations where three dimensional power is at work? A third example, which merits extensive discussion because of our very topic, is, indeed, hinted at by Galbraith himself in the citations above, in which he touches on how people can come to think about markets. Here, I can only point to what is called the ‘performativity thesis’¹⁰; the idea that more and more of us are increasingly thinking, in more and more spheres of life, like economists, taking the social world to be as mainstream or neoclassical economic theories represent it, thereby rendering us blinder than before to the less observable operations of power.

That is all by way of saying that I claim that the idea that people may have preferences that are underpinned by beliefs that work against their interests is not an implausible idea. Adducing evidence and reasons to support such a claim is, of course, no simple matter, but when people say, ‘How do you prove it?’ I am inclined to reply, ‘Are you really suggesting that this does not happen?’ I believe that Jonathan and I are not in disagreement on this point.

The second criticism of my view (and the first of two that Jonathan offers) is to raise the real difficulty of determining just when explaining outcomes in terms of power is appropriate and when it is not. We need a way, as he points out, of identifying the powerful and the mechanisms at work that distinguish the operations of power from general socialisation and the internalisation of cultural norms. Also, we must recognise the obvious truth that there are countless impersonal social processes where concatenations of individuals’ actions generate outcomes that may be intended by few or even no one, as when house buying choices by whites result in black ghettoes. (Hayek celebrates these, calling them ‘catallaxies’, with markets in mind, thus focusing exclusively on their mutually beneficial consequences).

To this, my response is to point to the link between locating power and attributing responsibility. In other words, when we want to find where power lies, our purpose is always to find out what is going on, discover the mechanisms, and identify actors, whether individual or collective, who could have acted otherwise, and who made a difference. Positing that they could have acted otherwise involves a counterfactual claim that there were other feasible possibilities, allied with the claim that there is a mechanism that could be identified as a casual process. There is a connection between power and responsibility that is inescapable. Cultures and impersonal social processes are enacted by human agents: cultural norms are promulgated, promoted, policed,¹¹ and enforced, and impersonal social processes can be encouraged and facilitated. Sometimes, cultural and structural explanations of troubling outcomes are proffered instead of explanations in terms of politics and power, thereby deflecting out attention from the role of the powerful.

Finally, this last observation relates to the second criticism Jonathan makes: that, alongside writers like Pierre Bourdieu and Michel Foucault, I think of power in a critical way, so that, in my earlier work, I did assume it to take the form of domination and to be harmful to people’s interests; that it is what he calls ‘cryptic’; that there is something nefarious at work here; that, in short, studying and analysing power is a work of debunking and unmasking. As indicated above, I take this criticism seriously: power, viewed three dimensionally, can indeed be beneficial and empowering, and in diverse ways. This clearly suggests a direction in which the ‘power debate’ can be carried forward. Debunking illusions and unmasking hidden mechanisms cannot be all there is to critical social enquiry.

Power and Economics by Jonathan Hearn

Steven has invited me to put the emphasis on our disagreements. They are, of course, friendly disagreements, all within a shared tradition of thinking critically about society. I will try to respond to what Steven has said and do my best to relate it to the conference theme of power and economics. But, obviously, we are focused on the question of his third dimension of power, and disagreements about that. To some degree, we will need to leave the rest of the contributors to draw the wider connections.

Pretty clearly, the first and second dimensions of power are very relevant to economics. Many power struggles in the economic arena are quite manifest and observable contests, and many others happen behind the scenes, away from public view, but nonetheless ultimately observable. By somewhat crude analogy to formal politics, we might say some power struggles happen openly in the marketplace, and others ‘around’ the marketplace, in acts that constitute it in the first place. But we are preoccupied at the moment with the ‘third dimension’. So, let me, following Steven, first try to restate what I understand to be the third dimension’s defining features, then address whether power necessarily involves intention and action, and, finally, turn to the other areas of criticism Steven has raised, concluding with some remarks on the task of criticism.

The third dimension involves ‘shaping beliefs and preferences that can adversely affect people’s interests’ according to Steven. Moreover, in its third dimension power operates by preventing grievances from arising, through ‘shaped’ preferences that fail to correspond to real interests. The main thing I would note here is that third dimensional power is fundamentally conceived as a critique of harm. While Steven allows that power need not be negative, and can even be beneficial, in the case of the third dimension, injury to people’s true interests is the basic evidence that power is at work. This kind of power tends to correspond with what is commonly meant by ‘domination’. However, it is worth noting that the defining feature of domination in this definition is found in the dominated, in their state of being harmed, of having a discrepancy between their interests and preferences; it is not found in the dominant, who are left relatively undefined.

Steven argues that power, especially in the third dimension, need not involve ‘intention’, or ‘action’. The nub of the argument is that power is most effective when its operations are least observable. In other words, when all intention and action appears to come from those being dominated, acting on their apparent preferences. But this raises two puzzles. First, are the powerful in this case merely ‘lucky’, the dupes of good fortune? Or, is their capacity to be the passive recipients of the beneficial intentions and actions of others (less powerful), nonetheless due to actions and intentions that make this situation more likely? Actions and intentions are not discrete isolated events. Tending to treat them as such was one of the weaknesses of the behaviourist paradigm. Instead, they come in complex patterns. Steven’s example of Shapiro and Graetz’s study of the repeal of the California ‘death tax’ is a case in point. Voters voted their preference for repeal, against their own interests, and in the interests of the very wealthy. This was not simply a fortunate accident for the powerful. As he points out, they had a hand in orchestrating the voters’ intentions and actions. Steven suggests that the subordination of women through such things as the ‘beauty myth’, or the naturalisation of the market in the popular imagination, provide even better examples of third dimensional power. However, the very fact that he raises these as problems, which we, his audience readily recognise, implies that they are not as ubiquitous and unquestioned as we might assume. Beliefs can be both prevalent and under question at the same time. Here, I would note that the recent focus on the performativity of markets in economic sociology is both an explanation of how markets become naturalised (in theorising markets as objects, economists help bring them into being), and a questioning of that naturalisation.

The second puzzle is this: if we define the third dimension of power in terms of harm done to interests, but we cannot connect this harm to intentions, it becomes very difficult to marshal a critique. Steven may intend a kind of ‘consequentialism’ here, where we are held responsible for the consequences of our actions, regardless of intent. But this certainly flies in the face of conventional moral and legal thought, in which responsibility is normally tied to intent. It would seem fruitless to try to build a position of critique disconnected from such conventions. There are of course in-between cases. In charges of ‘manslaughter’, we hold people responsible not for their intention to kill, but for their negligence in allowing death to happen. Thus, in the economic sphere we might want to distinguish the kind of harm done through deliberately mis-selling financial products or rigging LIBOR, from that done by failing to comprehend the implications of credit-default swaps and their liabilities. And, as Peter Morriss¹² has argued, we may want to allow a role for the critique of society, as a system with negative effects, which does not necessarily involve a critique of agents. It may be that highly liberalised forms of capitalism, in which capitals are insufficiently balanced by countervailing powers (to invoke J.K. Galbraith again), generate degrees of inequality that are systemically harmful to everyone’s interests. Some might attribute the current Greek debt crisis to the irresponsible behaviour of Greek citizens and politicians, but others might see it as the outcome of an irrational European monetary and credit system as a whole.

Let me turn now to Steven’s final three points. First, indeed, I agree with Steven that we must allow the possibility of claiming to understand people’s interests better than themselves. As long as the capacity to make such claims is reciprocal, not the privilege of some group (academics? social scientists?), and claims are open to debate, I would see this premise as not just acceptable, but necessary for a free society. Respecting people’s autonomy does not require placing their beliefs about their preferences beyond dispute. Here, the distinction between ‘interests’ and ‘preferences’ is crucial, and a challenge to any economic theory that would not want to make this distinction. Clearly, Steven’s conception of ‘interests’ is rooted in a notion of universal human goods (particularly autonomy and self-determination), while ‘preferences’ for him suggests a more contingent formation of wants and desires.

Second, yes; I think that we need to define the powerful in terms of their advantages, not just the subordinate in terms of the harm done to them. Without specifying the powerful, it is difficult to distinguish the ‘shaping of preferences’ from basic socialisation. Simply being a member of society is itself in some sense limiting—we are bound by shared norms. But we do not want to treat all instances of this kind of constraint as ‘harm’, as being subject to power. Steven’s response is that we must connect power and responsibility. But how do we do this if we have demoted ‘intention and action’ in our definition of power? Yes, we can posit counterfactual scenarios in which the powerful could have acted otherwise. But if this is to be more than the banal observation that life and history can take many different paths, the ‘could have’ needs to become a ‘should have’, which implies some intent and foresight¹³ on the part of the actor.

Finally, I have indeed argued that Steven’s concept of the third dimension is akin to others like Bourdieu’s ‘habitus’ and Foucault’s notion of ‘subjectivity’ as a construct of power relations. All of these place special emphasis on the hidden operations of power, what I have called ‘cryptic domination’.¹⁴ But the problem here is not one of being critical per se. I think there is a proper role for social critique, and for the critique of power. My point is that it is often suggested that these forms of cryptic domination take us to the heart of what must be critiqued, and I disagree. My point is that there are plenty of more readily discernible operations of power that can be more fruitfully critiqued, whose legitimacy can be more effectively challenged. If we become preoccupied with the question, ‘Why do people not see that they are being dominated?’, we focus our attention on the sad state of the dominated, to the neglect of the dominant, and how domination is done. I think more can be achieved through the more mundane task of clearly ascribing power to certain actors, defining what might make that power legitimate, and questioning whether it is legitimate. I worry about inviting too many bright young minds to puzzle over the mysteries of invisible power, to the neglect of more discernible and tractable problems of power. The critique of power must work in the light, or at least the half-light, of claims and counter-claims about legitimacy. To define power as hidden in its very nature is actually to put it beyond critique. Even Galbraith’s idea of ‘conditioned power’ suggests that while it is usually taken for granted, as in Steven’s opening quote from Galbraith about the unquestioned power of the market, the social mechanisms that produce and reproduce it—schooling, media, advertising, moral authorities, and so on—are normally well-known, and susceptible to scrutiny. Critique as such begins not with the question ‘Why don’t we see it?’, but with ‘Why do we, and why should we, accept it?’