Monopolistic competition and unions

Gerard Jackson
BrookesNews.Com
Monday 3 March 2008

I have had numerous emails from lefties gloating over Labor's promise to repeal Howard's labor market reforms. (In any case, it's not Rudd they should be thanking but the stunning arrogance and incompetence of the Melbourne-based H. R. Nicholls Society). These ill-informed critics dismissed labour market reforms along the lines of something like: "Typical economic rationalist thinking", "economic rationalism is rightwing ideology", "free labour markets exploit labour", "the economic case for labor market reform is a fraud", etc. Now these accusations have been around for years without a word of protest from 'free market advocates' like the HRNS. Even when the ACTU attacked neoclassical economics they remained silent.

The ACTU based its case on indeterminacy, meaning that there is no determinate price for labour services. In an effort to defend destructive union wage-setting Grant Belchamber (ACTU's chief research officer), Hristodoulidis and Andrew Watson (both ACTU staffers) used the concept of imperfect competition to support their argument. I've come across this approach before. For example, or as the argument goes, it doesn’t matter if Australia has thousands of employers because the economy is dominated by oligopolies while the remaining businesses are price takers.

Obviously the term oligopoly means few sellers, implying domination of the economy by a handful of huge firms who then use the market power to extract monopoly profits from exploited consumers and employees. But this approach makes no reference to the problems that the idea of oligopoly presents to economics.

Using perfect competition to define the competitive process logically leads us to breaking up car firms, power stations, steel mills, etc., into a large numbers of high-cost, low output inefficient producers. Ironically, this very point was made by Edward Chamberlin, one of the founders of the idea of “monopolistic”, who also repudiated the concept of perfect competition. This is something that invariably embarrasses adherents of the perfect competition model. The literature on oligopoly is considerable and quite a bit is difficult for the layman to follow. Unfortunately many critics’ knowledge of the subject usually comes from an introductory text. This misleads too many of these critics into thinking they have learnt all there is to know on the subject.

Monopolistic competition and oligopoly are economic dead-ends that have wasted an enormous amount of intellectual talent. They emerged in the 1930s as an attempt to explain the apparent rigidity of certain prices. It was decided that these prices were actually administered by a few dominant firms whose size allowed them to defy market forces. Therefore, the problem, some thought, was to try and explain the process by which this was achieved.

This gave rise to the theory of the kinked demand curve, which was developed simultaneously by Paul Sweezy in America and by Hall and Hitch in Britain. To keep it brief as well as simple, the theory states that firms will quickly follow price reductions but not price increases. It is this behaviour that produces the kinked demand curve. In plain English, any point above the kink is price sensitive, what economists call elastic. This means raising prices above this point will not only reduce sales but also revenue.

Complicating the issue is that the assumption of a kink also means there is a discontinuity in the marginal revenue curve, i.e., the additional revenue from every unit sold. It is also assumed that the marginal cost curve, the additional cost of each unit of output, intersects the discontinuous section of the MR curve. (I’m sorry to have to go into this, but this is what one usually finds in economic textbooks and that is where many critics of deregulation get their anti-market ideas).

What this boils down to is that a recession can result in prices remaining unchanged so long as marginal costs are within the discontinuous range of the marginal revenue curve. A situation that would not arise where there were a large number of competitors or very low barriers of entry. Now the possibility that a firm can keep its price cuts a secret or might be a price leader is also considered. In either case, the kink disappears. The first possibility can be discounted while the second possibility does not mean, in my opinion, that the kink really disappears, unless one also assumes that the price leader can determine the shape of the demand curve. This is because there is no theoretical reason why a demand curve cannot have a number of so-called kinks in it.

It would only mean that there are a range of elasticities along the curve. All the kink really means is that the market clearing price has been reached. The so-called price leader is in fact the leading price searcher who can afford to take the risk of losses if he exceeds the market price. If he makes economic profits the others will follow and it will only indicate that they were operating below the kink.

After all, an economic profit is a signal that demand exceeds supply and that more must be produced. Furthermore, any firm’s demand curve is a set of expectations of how many units it can sell at certain prices. As it consists of expectations there no reason why the firm cannot build into its demand curve expected reactions of other firms. I fail to see why this should not then make the situation a competitive one, thus tending to make demand sensitive to price. The important thing, however, is not posted prices (sometimes called administered prices) but transaction prices: the prices at which the goods are actually sold. If it can be shown that transaction prices are flexible then the whole question is transformed into a curiosity of economic thought.

Stigler and Amherst studied so-called administered prices over a two year period. The work was published in America’s National Bureau of Economic Research in 1970 under the title The Behaviour of Industrial Prices. To put it mildly, the results of this study completely undermined the administered-price thesis. And it was this thesis that oligopoly theory was developed to explain. I have found that those who parrot oligopoly assume that a small number of large producers mean ‘excess’ profits. (I can never get them to define excess profit).

This is just another variation of the industry-concentration argument. In 1951 Professor Joe Bain published a study which purported to show that there was indeed a correlation of above normal profits with industry concentration. The study quickly became a gospel for those suffering an oligopoly fetish, or shared monopoly as some preferred to call it.

However, Professor Yale Brozen attempted to find evidence that would support what had now become, what Galbraith would caustically call, conventional wisdom. His findings were published in 1970 and found that the results of the studies supporting the concentration thesis were in error. They had failed to take time into sufficient account. Another blow had been dealt to the alleged importance of ‘oligopolies’ and their so-called ability to fix prices and inflate their profits.

Barriers to entry are frequently cited as a form of market failure. But these alleged barriers do not seem to have profited so-called oligopolies. Why? Because the barrier to entry is basically the same for all business activities — scarcity of capital goods. If these were not scarce anyone could literally enter any line of business without ever suffering losses. In the absence of coercion there is always freedom of entry.

Those who think otherwise are confusing freedom of entry with economic abundance. This fact does not stop some from arguing that people cannot compete with our economic giants. Several points have to be made here: Every giant firm started as a baby. Hewlett Packard, for example, started in a garage. (Computer companies are famous for their modest beginnings). The same argument was also used regarding the canal companies and then the railway magnates.

How many are now prepared to invest in either of these areas? In addition, economies of scale are also not the barrier to entry that many claim any more than they are source of monopoly power or a symptom of market failure. If companies do not compete against companies enjoying economies of scale they are admitting that they cannot provide consumers with a superior alternative.

Why should this be condemned as some kind of market failure? We just cannot know the future. Anyway, what makes people think it is harder to compete today than it was in the past? Just imagine a mediaeval peasant trying to accumulate enough capital to build a mill or a tavern. Impossible. But today, total capital and wealth has advanced to the extent that now makes it easier for individuals to launch businesses.

Think how much easier it would be without excessive regulations and heavy taxation. That these raise barriers to increased competition and higher wage wages is never mentioned by critics of the market process. Also forgotten is that ‘economic giants’ are not owned by one person or even one family — they are owned by millions of shareholders. The pooled savings of these shareholders make these companies possible in a free market. This is not the place to get involved with the structure of cost curves. But enough work has been done to completely undermine the concepts of “monopolistic” competition and “oligopoly” and render union reliance on them totally irrelevant to labour market reform.

Further comments on monopolistic competition:

We can thank the model of perfect competition for the emergence of 'monopolistic competition. In reaction to the unrealistic assumptions of perfect competition Joan Robinson and Edward Chamberlin produced — independently of each other — the model of monopolistic competition in the belief that it was more representative of the real world. It was not. They made the fundamental error of assuming that because the demand curve was downward sloping this must restrict result in higher costs and lower output. But all demand curves are downward sloping. Only in the fantastic world of perfect competition can the demand curve be horizontal.

Then there is the neoclassical error of making the cost curve tangent to the demand curve. In addition, they thought the marginal approach could be used to determine the monopoly price. Chamberlin at least quickly realized that the intersection of the marginal cost curve and the marginal revenue curve can tell us nothing about a monopoly price. In any case the whole approach is wrong to begin with. It is a gross error to start the analysis with marginal curves, particularly smooth ones. Mises was among those economists who saw the error:

The planning entrepreneur is always faced with the question: To what extent will the anticipated prices of the products exceed the anticipated costs? If the entrepreneur is still free with regard to the project in question, because he has not yet made any inconvertible investments for its realization, it is average costs that count for him. But if he has already a vested interest in the line of business concerned, he sees things from the angle of additional costs to be expended. He who already owns a not fully utilized production aggregate does not take into account average cost of production but marginal cost. Without regard to the amount already expended for inconvertible investments he is merely interested in the question whether or not the proceeds from the sale of an additional quantity of products will exceed the additional cost incurred by their production. (Ludwig von Mises, Human Action, Henry Regenery Company, 1963, p. 343).