Why competition? 250 years of learning and forgetting in Political Economy1
Opening remarks
The notion that promoting competition is a Good Thing has become a consistent theme in economic policymaking in recent decades, accompanied by an implication that “the more of it the better” should be a presumptive policy stance. In contrast, very many members of the public appear to find these propositions far from obvious, not least those who are owners of, or workers for, business enterprises.
A significant part of the divergence in views can be attributed to the tendency in policy circles to use the word ‘competition’ in abstract, context-free ways; whereas public perceptions are likely formed from much more grounded, particular experiences.
In general usage, the meaning of ‘competition’ is pretty much what the dictionaries say it is: a state or process of rivalry in which the relevant agents strive for something they can’t all have. With this meaning in mind, we can ask the question: Is rivalry always a Good Thing? And the answer is clearly no: it depends on the relevant context.
Two examples serve to illustrate the point. First, there can be rivalry/competition between two or more parties for control over a piece of territory, an intense form of which is a protracted war. Promoting more intense rivalry in this domain of human conduct is not a self-evidently Good Thing to Do, though it does attract some supporters. Indeed, in a Hobbesian conception of things Leviathan exists precisely in order to prevent/suppress forms of rivalry that can be expected to have ‘nasty and brutish’ consequences.
Second, we can consider the less extreme form of competition that is sibling rivalry. That too doesn’t necessarily work well for a wider social benefit, the story of Cain and Abel being a mythic illustration of the point.
In theory and practice, economists have attempted to narrow the meaning of the word ‘competition’ by means of qualifying adjectives that enable it to be deployed in more specific contexts. Thus, in the relevant literature we find references to ‘perfect’ competition, ‘monopolistic’ competition, ‘oligopolistic’ competition, ‘workable’ competition, ‘effective’ competition, and so on.
This kind of typology is useful in the lecture hall and seminar room, where the meaning of the qualifying adjectives can be examined, and it does at least recognise the significance of contextual differences when assessing different processes of rivalry and their effects.
On the other hand, it can all be very confusing in more general discussion and debate about public policies. What is ‘perfect’, in an ordinary language use of that word, about the economic notion of ‘perfect competition’? Where is the rivalry in situations where all suppliers are assumed to be price takers? Why is the word monopolistic (suggesting a single supplier) attached to a form of competition among large numbers of businesses? If oligopolistic competition is ‘competition among the few’, how few is few? What distinguishes workable competition from non-workable competition?
Whilst being a step in the right direction — toward recognising the significance of contextual differences when assessing rivalries and their effects – the schematic economic typology is but a small step. It comes nowhere close to capturing all the salient features that might be influential in determining how a process of rivalry will play out in a particular, given set of circumstances.
The different economic characterisations of competition do, however, have a common feature that is worth noting: they all implicitly presume a situation in which the relevant situations of interest are characterised by rivalry among suppliers of products and services to win the business of customers. Put a little more generally, the focus of attention is on exchange transactions, on the buying and selling of goods and services (and rights), which can encompass competition/rivalry among buyers, as well as among sellers.
The same focus of interest was a characteristic of the early developers of political economy, so common ground between past and present practitioners is to be found here. However, the fact that the relevant claims concerning the benefits of competition are restricted to a relatively narrow range of the socio-economic activities (engagement in exchange transactions) is not always well communicated by today’s makers of those claims.
This tends to lead to a poor rhetoric. Promoting rivalry is less obviously a Good Thing to Do than, say, promoting ‘health and safety’ or a ‘better environment’. In consequence, when there are trade-offs to be assessed at a general level of public policy, it can be relatively easy to underweight the priority that should be appropriately given to competition issues.
In addition to being clear about the nature what the rivalry is about – acquiring control of territory, winning the affections of parents, winning the business of customers, etc. – it is important also to pay attention to the means of attaining the desired ends (see the Cain and Abel story on fratricide as a means of competing).
The sought-after end on a football pitch might be a trophy, but victory could be achieved by conduct that can be characterised as playing a beautiful game, or by conduct that amounts to cheating or thuggery. In the field of exchange transactions, sellers might win business by providing their customers better value for money than their competitors, or, depending on circumstances, by being more proficient in the practice of fraud. Again, ‘promoting
competition’ is too broad a slogan to command widespread assent.
Paying attention to the means of rivalry leads naturally on to questions concerning the ‘rules of competition’ and of issues to do with the enforcement of those rules. These things are part and parcel of the institutional architecture of a commercial society and are, or at least should be, the chief concerns of competition policy. Yet in today’s universities the teaching of microeconomics is heavily dominated by analysis that largely ignores the significance of the ‘rules of competition’ and the effectiveness of their enforcement. It abstracts from these features of the economic landscape. It pays little or no attention to them and, when used in the policy field, often ends up with false dichotomies, for example as between ‘markets’ and ‘regulation’, failing to notice that markets are themselves regulatory institutions: they regularise/govern transactional conduct.
Given these preliminaries, the question for today is ‘Why competition?’ Why and in what circumstances might it be thought desirable to establish, sustain or promote rivalry in transactional (buying and selling) processes? And it asked because I do not think it is a question that has been well answered on the basis of the ‘blackboard economics’ – an expression used by Ronald Coase, which dates him a little! – that still dominates in university teaching.
Models versus realities: the problem of idealised alternatives
Let me start with an autobiographical anecdote. Being driven around Krakow’s ‘new town’ (a product of the Soviet era) in April 1990, my host apologised for the pot-holes in the road and added that, with a competitive market economy in prospect, they would soon be gone.
The sentiment was obviously based on contemporary doctrines about the merits of such an economic system, but my first response was to invite him, on a future visit to the UK, to come to my Oxfordshire village to inspect the (extensive) pot-holes to be found there and hence see that a competitive market economy would not necessarily lead to their disappearance. My second, later response was to insert the following into the text of one of a series of lectures at the Jagiellonian University in Krakow which I had been invited to deliver: “The real world is not like that. Market processes are often messy: outcomes can be both inefficient and unfair. The soundest argument for markets … is simply that, very frequently, they are the least bad of the alternatives. To paraphrase Churchill’s remark on democracy, markets are the worst method of resource allocation, except for all the others that have been tried.”
This view of competitive market processes is, I think, much preferable to the starry-eyed idealisation of the effects of transactional competition that seem to motivate those economists and politicians who have spoken about the ‘magic of markets’. The tendency toward idealisation of competitive market processes seems to come either from the implications of economic models/theories heavily-laden with restrictive assumptions that are wildly out of joint with observable evidence, or, without those analytical formalities, with some handwaving reference to the invisible hand metaphor used by Adam Smith. In the latter case, it is perhaps ironic that it rests on a misreading of Smith’s own use of the metaphor, a mistake that, in their time, has been made by more than one Nobel Laureate.
Models of competitive processes are necessarily abstractions. They typically exclude consideration of large numbers of factors likely to be relevant for assessment of a given, realworld context and the high level of abstraction is usually motivated by a demand for analytic tractability. Assumption is piled on assumption, including assumptions about the mathematical forms of the relationships between those variables that are encompassed by the analysis, a recourse to linearity being a frequently observed illustration.
These remarks do not, however, imply that theorising/modelling is a valueless exercise. It is an exercise that can be useful as part of a heuristic process, i.e. as an aid to learning, to discovery, to understanding a context, or to problem solving. The point is simply that, by and of itself, it is not sufficient (and usually very far from being sufficient) to serve as a representation of realities upon which sound decisions can be made. Always take modelling results with a very large pinch of salt is the general advice.
Old masters on the relative merits of competition
‘Promoting competition’ is not a recommended general objective that is easily found in the work of past maters of political economy, a feature of which is an antipathy to the utopianism inherent in the use of idealised alternatives drawn from abstract analysis. Citation and examination of a number of their expressed views on the evaluation of competition can illustrate this. I start with Alfred Marshall, because he was a major reference point in much of my own, long-ago undergraduate training.
Alfred Marshall
“If competition is contrasted with energetic co-operation in unselfish work for the public good, then even the best forms of competition are relatively evil; while its harsher and meaner forms are hateful. … … But in the responsible conduct of affairs, it is worse than folly to ignore the imperfections which still cling to human nature.”
The point is clear: to presume the existence of universal, energetic cooperation for the public good is a fantasy, remote from the observational evidence on human nature and conduct. This view doesn’t entail denial of the existence of characteristics of human nature other than selfishness, examples of which include sympathy and benevolence. Indeed, more than a century earlier Adam Smith had set out and examined a range of these other characteristics in the Theory of Moral Sentiments, recognising their significance for human conduct. Rather, it implies only that it is folly to turn a blind eye to the ubiquity and prevalence of selfishness.
Turning a blind eye would imply the rejection of science, which demands the submission of theorising to observational evidence; and, from the period of the Scottish Enlightenment onwards, the major political economists considered themselves to be engaged in scientific endeavours.
Elsewhere, Marshall wrote as follows about the methodology of economics (the term that had increasingly replaced ‘political economy’ over the course of the 19th century, a process that itself reflected an unfortunate narrowing of the field of attention of the discipline): “… economic doctrine is not a body of concrete truth, but an engine of discovery of concrete truth.”
If ‘truth’ is to be assessed (as it is in a court of law) in terms of an account of things that corresponds with observations/evidence in particular (‘concrete’) situations/contexts, it cannot be found in theories or models, but only in their concordance with observables. The task of political economists is therefore to discover and develop analyses and understandings that fit the observational evidence. This requires a continual checking exercise, in the absence of which theories can easily transmute into ideologies, i.e. closed systems of beliefs that are resistant to reality-checks.
John Stuart Mill
“I do not pretend there are no inconveniences in competition, or that the moral objections urged against it by Socialist writers … are altogether groundless. But if competition has its evils, it prevents greater evils.”
Mill’s Principles of Political Economy was first published in 1848, the same year as the Communist Manifesto. Whilst the term ‘socialism’ has, like ‘capitalism’, subsequently become an ideograph (i.e. a vague expression that serves rhetorically as a rallying banner for the orientation or attitude of a set of partisans, without reference to anything very specific), the language indicates that he had in mind some rather more particular sets of views (expressed by a set of ‘Socialist writers’ of the time).
These writings were, for Mill, the chosen comparators for his assessment of the merits and demerits of competition. The comparators were themselves to be examined for their implications/effects: good intentions, irrespective of any ethical merit they may command at the level of the individual, carry no weight in evaluating alternative forms of governance for a complex socio-economic eco-system.
As for the defects/evils of the alternatives, Mill wrote that: “wherever competition is not, monopoly is”, and monopolies, whether publicly or privately created, were anathema to classical liberals. A flavour of their perceived economic defects is given by another passage, which draws heavily on Smith (see later):
“In addition to the tax levied for the profit, real or imaginary, of the monopolists, the consumer … pays an additional tax for their laziness and incapacity. When relieved from the immediate 6 stimulus of competition, producers and dealers grow indifferent to the dictates of their ultimate pecuniary interest; preferring to the most hopeful prospects, the present ease of adhering to routine. A person who is already thriving, seldom puts himself out of his way to commence even a lucrative improvement, unless urged by the additional motive of fear lest some rival should supplant him by getting possession of it before him.”
This (monopoly) was the greater evil to which the ‘Socialist writers’ were, explicitly or implicitly, inordinately attached.
Adam Smith
“Monopoly, besides, is a great enemy to good management, which can never be universally established but in consequence of that free and universal competition which forces everybody to have recourse to it for the sake of self-defence.”
Smith’s critique of monopoly is interestingly different from that of 20th century concerns focused on allocative efficiency, which tend to have an intense interest in the relativity between prices and costs of supply. It is much more focused on the effects of a structural monopoly on the motivation of suppliers, as is illustrated by a much more cited passage:
“It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own self-interest.”
Whilst managers of both monopolistic and competitive entities may share an interest in increasing the financial returns they make, the difference is that lack of challenge in a protected monopolistic position affords managers a great deal more discretion in their decision making. The trade-off between, say, the ease of familiar routine and prospective financial returns is much less sharp. If, say, the immediate economic environment changes, the monopolist can choose not to respond in any significant way: it can continue to operate on a business-as-usual
basis. It will not then achieve as high a financial return as would be potentially feasible in the new conditions, but, barring a major change in circumstances, it will not face threats that could be said to be existential or near existential. It will still be fit enough for survival.
The same is not true in the competitive case. Financial returns are heavily geared to a business’s performance relative to its rivals and hence will depend on how other businesses respond and adapt to changes in the economic environment. The motivation/incentive to be constantly vigilant in the face of changes in market conditions, and to be more responsive to them, is therefore substantially stronger. Adaptation to change can become close to a necessity for survival, rather than a sustainable choice of lifestyle. There is constant pressure for performance improvement.
The structure of the reasoning anticipates Darwinian theorising that came to the fore a century later, but what is ‘selected out’ is of a rather different nature – a fact that was not taken fully on board by post-Darwin ‘social Darwinists’. In the social science case, it is behaviours (whether of individuals or organisations); in biological evolution it is genetic structures. And behaviours are much more mutable and transmissible (copied by others) than genetic structures.
Friedrich Hayek
“Competition is a procedure for discovering facts which, if the procedure did not exist, would remain unknown or at least would not be used.”
In effect, Hayek generalises the classical political economy view of competition as a dynamic, evolutionary process, going a step further in linking it specifically to the discovery and use of new information of economic value. The process motivates/incentivises both discovery and, to the extent that it is of economic value, the usage of the newly discovered information. And that characteristic is highly conducive to (as Smith would put it) to ‘economic progress’ or (as it now might be put) productivity growth.
A corollary of this view is that it is not possible to forecast the outcomes of a competitive process with any great precision: they will be unpredictable. ‘Hopes’ might be attached to any prospective venture, but not confident expectations about outcomes, nor even confident prior probabilities of outcomes (an issue that Keynes struggled with in his first work, A Treatise on Probability). We can be confident that new information and (potentially) new uses of information will be discovered, but, definitionally, are in a fog as to what discoveries yet to be made will be made, and hence about the outcomes.
The point should induce a high level of scepticism about large swathes of modern analyses of competition based upon static theorising, which are rich in predictions of what the outcomes of policies to change competitive conditions will be.
The better news is that it remains possible to be more confident in developing propositions about the characteristics and tendencies of processes and their structures, and Hayek does just that in his comparisons of competitive markets with central planning procedures, as did his predecessors Smith (competition vs monopoly) and Mill (competition vs socialist theories).
My own reservations about Hayek are chiefly to do with the absence of a body of work that connects high-level principles to the ‘concrete’ realities of economic life, an exercise that preoccupied the minds of Smith and Marshall. So, let me draw one small connecting line on a thinly populated canvas by way of a second anecdote.
In the early 1990s I sat round a table with senior executives of a privatised utility company, previously state-owned and protected by legislation that afforded it a statutory monopoly. In the new UK context created by privatization and the establishment of sectoral regulators, I asked if there was any internal information available on the costs of its retailing activities.
The answer was a straight no, and that was unsurprising. Discovery of such information would have been a costly activity and, up until the relevant time, there had never been any perceived, significant financial benefit from it. The potential use-value of it being low, so why allocate significant resources to the exercise.
The State’s monopoly power
It is a feature of political economy that it views the state as an integral part of its field of study, not as something akin to a deus ex machina, as it is often conceived in large parts of textbook/’blackboard’ economics. So viewed, not only is it a social entity to which very considerable powers of decision are afforded across more or less the whole field of economic activity, but it is also a spawning ground for the granting of monopoly rights to others. Smith and Mill were acutely aware of the latter propensity, living in times when the conduct of the East India Company was a major subject of contention. Their critiques of the limitations of monopoly therefore extend to the conduct of government itself and, because of the magnitude and scope of the powers involved, they extend to it in spades.
In respect of government, Mill wrote that:
“All the facilities which a government enjoys of access to information; all the means which it possesses of remunerating, and therefore of commanding, the best available talent in the market—are not an equivalent for the one great disadvantage of an inferior interest in the result.”
And that:
“It must be remembered, besides, that even if a government were superior in intelligence and knowledge to any single individual in the nation, it must be inferior to all the individuals of the nation taken together.”
The first citation here is a straight read across of the classical critique of monopoly, irrespective of its public or private nature (and the two natures were entangled in institutions like the East India Company). It concerns incentives/motivation. The second is a slightly different point, and it can also be made for large, vertically organised business hierarchies. It concerns limited information processing capacity, which tends to become a more acute factor the larger is the scope of the reach of the relevant entity.
There is also a third point that might be added when considering the state’s monopolistic powers: the problem of arbitrariness, which much concerned Smith (see his ‘canons of taxation’, which place a high weight on ‘certainty’). Being only weakly constrained by rivals, except perhaps in times of existential war, government has considerable discretion in how it conducts itself. Consequentially, there is significant scope for arbitrariness and instability in decision making.
Thus, even at end of a period of unusually elevated strategic clarity in government, John Vickers and I expressed the view that “The spectre of policy instability still lurks.” (in ‘The British Electricity Experiment’, Economic Policy, 1991). The point is that this arbitrariness/instability can have deeply damaging effects on the performance of the economy as a whole: a rule of law system loses effectiveness when laws and regulations are themselves constantly adjusted in unpredictable ways.
End comments
For rhetorical purposes, the political economists cited tended to set out their reasoning in terms of a general, simple binary comparison between competition and monopoly, or, in Hayek’s case, between competition and central planning (which entails monopolisation across a wide economic domain). Their underlying points, however, survive the obvious potential rejoinder that things are, in practice, not so binary.
The chief interest lies in the level of pressure there is on suppliers to be motivated to constantly strive to produce greater value for their customers, and ‘degree of pressure’ is a continuous variable. Rivalry intensifies that pressure and motivates/incentivises businesses to be constantly on the look-out for opportunities to get ahead of their competitors, for example by innovation, and, a fortiori, on the look-out for threats to be avoided, so as to avert the financial losses that would result if the business fell behind in the customer-value stakes. It is, as Hayek said, a discovery process and it drives the economic ‘improvement’ or ‘progress’ that all the forementioned scholars thought desirable. Where the pressure is weak, the effort put into discovery activities can be expected to be lower, muted by the comforts of familiarity and routine.
Put in more political terms, competition provides checks and balances on the accumulation of economic power to levels at which the harmful features of what the classical writers referred to as ‘monopoly’ become prominent. This ‘checks and balances gestalt’ connects naturally with widely held political views and renders it straightforward to communicate why competition/rivalry is such an important issue in transactional processes. It is not focused on trying to argue that competition is a Good Thing, but rather that highly asymmetric distributions of economic power (and/or, where they are unavoidable or difficult to flatten, abuses of that power) are Bad Things.
This gestalt/schema is now something of a lost legacy: today’s competition policies are grounded in narrower economic and legal thinking – a domain for specialist experts, who can find it a struggle to provide any widely understandable and acceptable reason to afford it a more prominent role in public policymaking. The loss is, I think, highly consequential.
1 A developed text version of a presentation given to a conference on the Role of Competition in Public Policy,
jointly organised by the Centre for Competition Policy (University of East Anglia) and the Regulatory Policy
Institute, in Westminster on 8 March 2010, two months ahead of General Election and in circumstances in
which, thanks in significant part to the 2008 financial crash, the merits of competition policy were being called
into question.