10: Monopoly

Natural monopolies

Traditionally, monopolies were viewed as being 'natural' in some sectors of the economy. This means that scale economies define some industries' production and cost structures up to very high output levels, and that the whole market might be supplied at least cost by a single firm.

Consider the situation depicted in Figure 10.1. The long-run ATC curve declines indefinitely. There is no output level where average costs begin to increase. Imagine now having several firms, each producing with a plant size corresponding to the short-run average cost curve , or alternatively a single larger firm using a plant size denoted by . The small firms in this case cannot compete with the larger firm because the larger firm has lower production costs and can undercut the smaller firms, and supply the complete market in the process. Such a scenario is termed a natural monopoly.

Figure 10.1 A 'natural' monopolist

When LR average costs continue to decline at very high output, one large firm may be able to supply the industry at a lower unit cost than several smaller firms. With a plant size corresponding to ATC₂, a single supplier can supply the whole market, whereas several smaller firms, each with plan size corresponding to ATC₁, cannot compete with the larger firm on account of differential unit costs.

Electricity distribution in some of Canada's provinces is in the hands of a single supplier – Hydro Quebec or Hydro One in Ontario, for example. These distributors are natural monopolies in the sense described above: Unit distribution costs decline with size. In contrast, electricity production is not 'naturally' a monopoly. Other suppliers were once thought of as 'natural' monopolies also, but are no longer. Bell Canada was considered to be a natural monopoly in the era of land lines: It would not make economic sense to run several sets of phone lines to every residence. But that was before the arrival of cell phones, broadband and satellites. Canada Post was also thought to be a natural monopoly, until the advent of FEDEX, UPS and other couriers proved otherwise. Invention can compete away a 'natural' monopoly.

In reality there are very few pure monopolies. Facebook, Microsoft, Amazon, Apple, Netflix and Google may be extraordinarily dominant in their markets, but they are not the only suppliers of the services or products that they offer. There exist other products that are similar.

National and Provincial Policy

Government policy can foster monopolies. Some governments are, or once were, proud to have a 'national carrier' in the airline industry – Air Canada in Canada or British Airways in the UK. The mail service was viewed as a symbol of nationhood in Canada and the US: Canada Post and the US Postal system are national emblems that have historic significance. They were vehicles for integrating the provinces or states at various points in the federal lives of these countries.

In the modern era, most of Canada's provinces have decided to create a provincial monopoly crown corporation for the sale of cannabis. But competition abounds in the form of an illegal market.

The down side of such nationalist policies is that they can be costly to the taxpayer. Industries that are not subject to competition can become fat and uncompetitive: Managers have insufficient incentives to curtail costs; unions realize the government is committed to sustain the monopoly and push for higher wages than under a more competitive structure, and innovation may be less likely to occur.

Maintaining barriers to entry

Monopolies can continue to survive if they are successful in preventing the entry of new firms and products. Patents and copyrights are one vehicle for preserving the sole-supplier role, and are certainly necessary to encourage firms to undertake the research and development (R&D) for new products.

Many corporations produce products that require a large up-front investment; this might be in the form of research and development, or the construction of costly production facilities. For example, Boeing or Airbus incurs billions of dollars in developing new aircraft; pharmaceuticals may have to invest a billion dollars to develop a new drug. However, once such an investment is complete, the cost of producing each unit of output may be relatively low. This is particularly true in pharmaceuticals. Such a phenomenon is displayed in Figure 10.2. In this case the average cost for a small number of units produced is high, but once the fixed cost is spread over an ever larger output, the average cost declines rapidly, and in the limit approaches the marginal cost. These production structures are common in today's global economy, and they give rise to markets characterized either by a single supplier or a small number of suppliers.

Figure 10.2 Fixed cost and constant marginal cost

With a fixed cost of producing the first unit of output equal to F and a constant marginal cost thereafter, the long-run average total cost, LATC, declines indefinitely and becomes asymptotic to the marginal cost curve.

This figure is useful in understanding the role of patents. Suppose that Pharma A spends one billion dollars in developing a new drug and has constant unit production costs thereafter, while Pharma B avoids research and development and simply imitates Pharma A's product. Clearly Pharma B would have a LATC equal to its LMC, and would be able to undercut the initial developer of the drug. Such an outcome would discourage investment in new products and the economy at large would suffer as a consequence. Economies would be worse off if protection is not provided to the developers of new products because, if such protection is not offered, potential developers will not have the incentive to incur the up-front investment required.

While copyright and patent protection is legal, predatory pricing is an illegal form of entry barrier, and we explore it more fully in Chapter 14. An example would be where an existing firm that sells nationally may deliberately undercut the price of a small local entrant to the industry. Airlines with a national scope are frequently accused of posting low fares on flights in regional markets that a new carrier is trying to enter.

Political lobbying is another means of maintaining monopolistic power. For example, the Canadian Wheat Board had fought successfully for decades to prevent independent farmers from marketing wheat. This Board lost its monopoly status in August 2012, when the government of the day decided it was not beneficial to consumers or farmers in general. Numerous 'supply management' policies are in operation all across Canada. Agriculture is protected by production quotas. All maple syrup in Quebec must be marketed through a single monopoly supplier.

Critical networks also form a type of barrier, though not always a monopoly. Microsoft's Office package has an almost monopoly status in word processing and spreadsheet analysis for the reason that so many individuals and corporations use it. The fact that we know a business colleague will be able to edit our documents if written in Word, provides us with an incentive to use Word, even if we might prefer Wordperfect as a vehicle for composing documents. We develop the concept of strategic entry prevention further in Chapter 11.

Marginal revenue and marginal cost

Table 10.1 displays price and quantity values for a demand curve in columns 1 and 2. Column 3 contains the sales revenue generated at each output. It is the product of price and quantity. Since the price denotes the revenue per unit, it is sometimes referred to as average revenue. The total revenue (TR) reaches a maximum at $32, where 4 units of output are produced. A greater output necessitates a lower price on every unit sold, and in this case revenue falls if the fifth unit is brought to the market. Even though the fifth unit sells for a positive price, the price on the other 4 units is now lower and the net effect is to reduce total revenue. This pattern reflects what we examined in Chapter 4: As price is lowered from the highest possible value of $14 (where 1 unit is demanded) and the corresponding quantity increases, revenue rises, peaks, and ultimately falls as output increases. In Chapter 4 we explained that this maximum revenue point occurs where the price elasticity is unity (-1), at the midpoint of a linear demand curve.

Figure 10.3 Total revenue and marginal revenue

When the quantity sold increases total revenue/expenditure initially increases also. At a certain point, further sales require a price that not only increases quantity, but reduces revenue on units already being sold to such a degree that TR declines – where the demand elasticity equals –1 (the mid point of a linear demand curve). Here the midpoint occurs at Q=4. Where the TR is a maximum the MR=0.

Related to the total revenue function is the marginal revenue function. It is the addition to total revenue due to the sale of one more unit of the commodity.

The MR in this example is defined in the fourth column of Table 10.1. When the quantity sold increases from 1 unit to 2 units total revenue increases from $14 to $24. Therefore the marginal revenue associated with the second unit of output is $10. When a third unit is sold TR increases to $30 and therefore the MR of the third unit is $6. As output increases the MR declines and eventually becomes negative – at the point where the TR is a maximum: If TR begins to decline then the additional revenue is by definition negative.

The MR function is plotted in Figure 10.4. It becomes negative when output increases from 4 to 5 units.

Figure 10.4 Monopolist's profit maximizing output

It is optimal for the monopolist to increase output as long as MR exceeds MC. In this case MR>MC for units 1, 2 and 3. But for the fourth unit MC>MR and therefore the monopolist would reduce total profit by producing it. He should produce only 3 units of output.

The optimal output

This producer has a marginal cost structure given in the fifth column of the table, and this too is plotted in Figure 10.4. Our profit maximizing rule from Chapter 8 states that it is optimal to produce a greater output as long as the additional revenue exceeds the additional cost of production on the next unit of output. In perfectly competitive markets the additional revenue is given by the fixed price for the individual producer, whereas for the monopolist the additional revenue is the marginal revenue. Consequently as long as MR exceeds MC for the next unit a greater output is profitable, but once MC exceeds MR the production of additional units should cease.

From Table 10.1 and Figure 10.4 it is clear that the optimal output is at 3 units. The third unit itself yields a profit of 2$, the difference between MR ($6) and MC ($4). A fourth unit however would reduce profit by $3, because the MR ($2) is less than the MC ($5). What price should the producer charge? The price, as always, is given by the demand function. At a quantity sold of 3 units, the corresponding price is $10, yielding total revenue of $30.

Profit is the difference between total revenue and total cost. In Chapter 8 we computed total cost as the average cost times the number of units produced. It can also be computed as the sum of costs associated with each unit produced: The first unit costs $2, the second $3 and the third $4. The total cost of producing 3 units is the sum of these dollar values: . The profit-maximizing output therefore yields a profit of $21 ().

Supply and demand

When illustrating market behaviour it is convenient to describe behaviour by simple linear supply and demand functions that are continuous, rather than the 'step' functions used in the preceding example. As explained in Chapter 5, in using continuous curves to represent a market we implicitly assume that a unit of output can be broken into subunits. In the example above we assumed that sales always involve one whole unit of the product being sold. In fact many goods can be sold in fractional units: Gasoline can be sold in fractions of a litre; fruits and vegetables can be sold in fractions of a kilogram, and so forth. Table 10.2 below furnishes the data for our analysis.

The first two columns define the demand curve. Total revenue is the product of price and quantity and given in column 3. The cost data are given in column 4, and profit – the difference between total revenue and total cost is in the final column. Profit is maximized where the difference between revenue and cost is greatest; in this case where the output is 8 units. At lower or higher outputs profit is less. Figure 10.5 contains the curves defining total revenue (TR), total cost (TC) and profit. These functions can be obtained by mapping all of the revenue-quantity combinations, the cost-quantity combinations, and the profit-quantity combinations as a series of points, and joining these points to form the smooth functions displayed. The vertical axis is measured in dollars, the horizontal axis in units of output. Graphically, profit is maximized where the dollar difference between TR and TC is greatest; that is at the output where the vertical distance between the two curves is greatest. This difference, which is also defined by the profit curve, occurs at a value of 8 units, corresponding to the outcome in Table 10.2.

Figure 10.5 Total revenue, total cost & profit

At any quantity less than this output, profit would rise with additional output. This is because, from a less-than-optimal output, the additional revenue from increased sales exceeds the increased cost associated with producing those units: Stated differently, the marginal revenue would exceed the marginal cost. Conversely, outputs greater than the optimum result in a MR less than the associated MC. Accordingly, since outputs where MR>MC are too low, and outputs where MR<MC are too high, the optimum must be where the MR=MC. Hence, the equality between MR and MC is implied in this diagram at the output where the difference between TR and TC is greatest.

Note finally that total revenue is maximized where the TR curve reaches a peak. In this example that occurs at a value of 12 units of output. This is to be anticipated, as we learned in Chapter 4, because the midpoint of the demand schedule in Table 10.2 occurs at that value.

Figure 10.6 Market demand, the MR curve, and the monopolist's AC and MC curves

Figure 10.6 displays the demand curve for the market, the MR curve, and the monopolist's MC and AC curves. Consider first the marginal revenue curve. In contrast to the previous example, where only whole or integer units could be sold, in this example units can be sold in fractional amounts, and the MR curve must reflect this. To determine the position of the MR curve, note that with a straight-line demand curve total revenue is a maximum at the midpoint of the demand curve. Any increase in output results in reduced revenue: Stated differently, the marginal revenue becomes negative at that output. Up to that output the MR is positive, as illustrated in Figure 10.3. Accordingly, the MR curve must intersect the quantity axis midway between zero and the horizontal-axis intercept of the demand curve. Geometrically, since the MR intersects the quantity axis half way to the horizontal intercept of the demand curve, it must have a slope that is twice the slope of the demand curve.

By observing the data in columns 1 and 2 of the table, the demand curve intercepts are , and from above discussion the MR curve has intercepts . The AC is obtained by dividing TC by output in Table 10.2, and the MC can be also calculated as the change in total cost divided by the change in output from Table 10.2. The result of these calculations is displayed in Figure 10.6.

The profit maximizing output is 8 units, where MC=MR. The price at which 8 units can be sold is read from the demand curve1, or the first column in Table 10.2. It is $8. And, as expected, this price-quantity combination maximizes profit. Table 10.2 indicates that profit is maximized at $48, at q=8.

Demand elasticity and marginal revenue

We have shown above that the MR curve cuts the horizontal axis at a quantity where the elasticity of demand is unity. We know from Chapter 4 that demand is elastic at points on the demand curve above this unit-elastic point. Furthermore, since the intersection of MR and MC must be at a positive dollar value (MC cannot be negative), then it must be the case that the profit maximizing price for a monopolist always lies on the elastic segment of the demand curve.

A general graphical representation

In Figure 10.7 we generalize the graphical representation of the monopoly profit maximizing output by allowing the MC and ATC curves to be nonlinear. The optimal output is at , where MR=MC, and the price sustains that output. With the average cost known, profit per unit is AB, and therefore total profit is this margin multiplied by the number of units sold, .

Total profit is therefore

Note that the monopolist may not always make a profit. Losses could result in Figure 10.7 if average costs were to rise so that the ATC were everywhere above the demand curve, or if the demand curve shifted down to being everywhere below the ATC curve. In the longer term the monopolist would have to either reduce costs or perhaps stimulate demand through advertising if she wanted to continue in operation.

Figure 10.7 The monopoly equilibrium

The profit maximizing output is Q_E, where MC=MR. This output can be sold at a price P_E. The cost per unit of Q_E is read from the ATC curve, and equals B. Per unit profit is therefore AB and total profit is P_EABC_E.

Perfect competition versus monopoly

The area ABF can also be considered as the efficiency loss associated with having a monopoly rather than a perfectly competitive market structure. In perfect competition the supply curve is horizontal. This is achieved by having firms enter and exit when more or less must be produced. Accordingly, if the perfectly competitive industry's supply curve approximates the monopolist's long-run marginal cost curve2, we can say that if the monopoly were turned into a competitive industry, output would increase from to . The deadweight loss is one measure of the superiority of the perfectly competitive structure over the monopoly structure.

Note that this critique of monopoly is not initially focused upon profit. While monopoly profits are what frequently irk the public, we have focused upon resource allocation inefficiencies. But in a real sense the two are related: Monopoly inefficiencies arise through output being restricted, and it is this output reduction – achieved by maintaining a higher than competitive price – that gives rise to those profits. Nonetheless, there is more than just a shift in purchasing power from the buyer to the seller. Deadweight losses arise because output is at a level lower than the point where the MC equals the value placed on the good; thus the economy is sacrificing the possibility of creating additional surplus.

Given that monopoly has this undesirable inefficiency, what measures should be taken, if any, to counter the inefficiency? We will see what Canada's Competition Act has to say in Chapter 14 and also examine what other measures are available to control monopolies.

Cartel instability

Some cartels are unstable in the long run. In the first instance, the degree of instability depends on the authority that the governing body of the cartel can exercise over its members, and upon the degree of information it has on the operations of its members. If a cartel is simply an arrangement among producers to limit output, each individual member of the cartel has an incentive to increase its output, because the monopoly price that the cartel attempts to sustain exceeds the cost of producing a marginal unit of output. In Figure 10.14 each firm has a MC of output equal to $F when the group collectively produces the output . Yet any firm that brings output to market, beyond its agreed production limit, at the price will make a profit of AF on that additional output – provided the other members of the cartel agree to restrict their output. Since each firm faces the same incentive to increase output, it is difficult to restrain all members from doing so.

Individual members are more likely to abide by the cartel rules if the organization can sanction them for breaking the supply-restriction agreement. Alternatively, if the actions of individual members are not observable by the organization, then the incentive to break ranks may be too strong for the cartel to sustain its monopoly power.

We will see in Chapter 14 that Canada's Competition Act forbids the formation of cartels, as it forbids many other anti-competitive practices. At the same time, our governments frequently are the driving force in the formation of domestic cartels.

In the second instance, cartels may be undermined eventually by the emergence of new products and new technologies. OPEC has lost much of its power in the modern era because of technological developments in oil recovery. Canada's 'tar sands' yield oil, as a result of technological developments that enabled producers to separate the oil from the earth it is mixed with. Fracking technologies are another means of extracting oil that is discovered in small pockets and encased in rock. The supply coming from these new technologies has limited the ability of the old OPEC cartel to increase prices through supply restriction.

Rent seeking

Citizens are frequently appalled when they read of lobbying activities in their nation's capital. Every capital city in the world has an army of lobbyists, seeking to influence legislators and regulators. Such individuals are in the business of rent seeking, whose goal is to direct profit to particular groups, and protect that profit from the forces of competition. In Virginia and Kentucky we find that state taxes on cigarettes are the lowest in the US – because the tobacco leaf is grown in these states, and the tobacco industry makes major contributions to the campaigns of some political representatives.

Rent-seeking carries a resource cost: Imagine that we could outlaw the lobbying business and put these lobbyists to work producing goods and services in the economy instead. Their purpose is to maintain as much quasi-monopoly power in the hands of their clients as possible, and to ensure that the fruits of this effort go to those same clients. If this practice could be curtailed then the time and resources involved could be redirected to other productive ends.

Industries in which rent seeking is most prevalent tend to be those in which the potential for economic profits is greatest – monopolies or near-monopolies. These, therefore, are the industries that allocate resources to the preservation of their protected status. We do not observe laundromat owners or shoe-repair businesses lobbying in Ottawa.