back start next


[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [ 47 ] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75]


47

Fig. Vil IVIonopoly Output

Fig. Vlll monopoly Price & Profit

MR-MC-

1 \ N - -1 -\-1 \

! \ 1 \°

value of that second unit must be $12 to you. Otherwise, you wouldnt be willing to pay that much for it. Moreover, in the absence of externalides, nobody but the transactors themselves are affected by trade. This means that the marginal benefit (MB) to you of having one more unit to consume is idendcal to that of society. After all, the marginal benefit to everyone else is zero. To emphasize the point, we will write P = MB in the following analysis. Now consider cost. Recall from Secdon I that all costs in economics are opportunity costs. This means that the marginal cost (MC) of producing one more unit of output represents the most highly valued alternative use of the resources used in producing that added unit. Consequendy:

(1) at Outputs such as Q In Figure IX where MB = P > MC, added resources in this industry are higher valued than in the next best one. For example, if wc value another unit at $10 (MB = P = $10), and it costs only $2 to make another unit (MC = $2), we would want more resources to flow into this industry, where they have a relatively higher value.

(2) at outpuU such as Q* at which MC > MB = P, resources have a higher value in some alternative industry than in this one (MC > MB). Therefore, we would want resources to flow out of diis industry and into another where their value is higher.

(3) at the output Q, where MC = MB = R added resources in this industry are just as valued as in the next best alternadve; so output is opdmal given our assumption about externalides.

In summary, only case (3) is efficient. In the absence of externalities, the economically efficient output is the one such that price = marginal benefit =

marginal cost. Othervwse, society can always do better by increasing the output of some industry and decreasing that of another Note that efficiency does not necessarily imply jusdce or fairness in a moral or philosophic sense. To say that an allocadon of wealth isjust requires that we know what people deserve, something economists in an uncharacteristic bow to modesty profess not to know.

As we have seen in Secdon IV, compeddve firms produce tdthie pffilltTir whiclrP"=C;diei efui e-, they are efficient. Subsidies to compedtive industries, on the other hand, are inefficient. In Figure X we sketch the supply and demand curves of the toothpaste industry. The compeddve price is Pc and output is Qc- Now let government impose a fne cent per tube subsidy on toothpaste. The subsidy will not reduce the

Fig. IX The Efficient Output

1 \

1 >

V-.- 1 1

; \ D=MB=P



Fig. X The Inefficiency of Subsidies

S=MC

\ / s=s-5e

1 "" D=MB=P

amount of resources necessary to make toothpaste, so the marginal cost (MC) of making toothpaste to society will remain the original supply curve S. But the marginal cost (MC) of making toothpaste as seen by manufacturers will be S - 5 cenu along the new supply curve S. The new price is Ps and the output is Qs- But this is inefficient: at Qs, marginal cost exceeds the price. We know from the MC curve at Qj that another unit of some other good is more highly valued than the last unit of toothpaste. We would therefore want resources to flow out of toothpaste and into the other industry where their value is greater. Nonetheless, the subsidy keeps the resources tied up in toothpaste.

While subsidies represent the classical case of "too much," the classical case of "too litde" occurs in monopoly. In the monopoly equilibrium of Figure VIII, we saw that price exceeds MC; for that reason, we would want more resources to flow into the monopolized industry where their marginal benefit is higher. Neverdieless, the profit incentives of the monopolist preclude expansion beyond the output Qm-

The welfare loss under monopoly can be measured, and the simplest case is presented in Figure XI. Here we assume constant average total cost, so that MC = by the marginal/average rules. Under compeddon, equilibrium output occurs where the MC strikes the demand curve, or at Qc. The consumer surplus under these circumstances is the area - - . Under monopoly, equilibrium price is and output is Qm so that consumer surplus is reduced to A. Note that the monopolist takes as profit the area B, which would be captured as consumer surplus under compe-tidon. Thus, while consumers lose B, the monopolist gains it, so area is not lost to society. However, the

Fig. XI Monopoly Welfare Loss

\I \

\ \ -

triangle which is lost from the competitive consumer surplus is not gained by the producer: it is completelv lost 10 societ}. For this reason, area is called die wel-fai e or efficiency loss due to monopoly. Note that it is approximately shaped as a triangle and therefore has an area equal to half its base times its height.

DISCRIMINATING MONOPOLY

So far weve considered only firms charging a single price for their product. But firms with down-\vard sloping demand curves can sometimes capture sdll bigger profits by charging different prices to different categories of buyers. The practice of doing so when production and selling costs are the same among buyer categories is termed price discrimination. We will consider only the simplest possible case, that of a consum average total cost monopolist. Although we will frame the analysis in terms of airline travel, it could potentially fit a number of other goods.

In Figure XII the hypothetical average total cost of providing airline travel to adults and children is consuni so that marginal cost is everywhere equal to average total cost (why?). Now recall that we normally generate the market demand curve by horizontally summing the demand curves of all relevant individuals, in this case, all adults and children. Suppose instead, we develop the demand curves of children and adults separately in the left and right hand panels. Then we would have two demand curves, D* and D respectively. (Note that D* does not necessarily mean that children pay their own airfare.) Since both these curves slope down, we also have the associated marginal revenue curves of adults and children, MR and MR* respectively. You will recall from our discussion in Section IV on Comparing Elasticities, that demand



Fig. Xll Discriminating Monopoly

Using Elasticity Formula (4) tat P" = (2 1/2) / (1 1/2) = 10/6, at P=(3/2) = 9/6.

....... o

\ >

t X.

V"i.....-\----

\ \»-.

\ MR

curve is less elasdc than demand curve D because the curves are parallel.

The monopolist maximizes profit by setting the

marginal cost equal to the marginal revenue obtained from the sale of child dckets and the marginal revenue from the sale of adult dckets, i.e., MR* = MC and MR = MC. Thus, Q<= child dckets will be sold and adult dckets. Childrens dckets will sell for the relatively low price of P* and adult tickets will be P. Using elasticity formula (4) of Section FV, we can see that the point elasdcity at P cexceeds that at P; at P* is about 10/6. d at P is 9/6. Hence, at the equilibrium prices, the more elasdc demanders, children, pay the lower price. This is only reasonable. The quantit) demanded by the elasdc group is relatively more sensitive to price increases than the inelastic one, so it makes sense to charge these more responsive demanders less. Intuitively, lumping elasdc and inelastic demanders together into one market demand curve would deprive the monopolist of the ability lo selectively raise price to those who tolerate it the most Consequently, his profits would be less.

Total profits will be the sum of the profits from selling to both groups. Those derived from selling to children are given by area 1 or (P*-ATC)Q. Those of selling to adults are area 2 or (P- )Q.

In picking the airline ticket example to illustrate price discrimination, we were careful to choose a service that cannot readily be resold. If you are starting to think as an economist, you ma) have already guessed the reason: Otherwise it would be possible for the low price buyers to profitably resell to the high priced buyers, perhaps through middlemen. Neglecting transactions costs, so long as there was any

difference between the two prices, opportunities for profitable resale exist. Clearly, this practice, termed arbitrage, threatens to destroy the whole price discrimination scheme. For example, imagine that prices differed. Then middlemen could buy at the low price, and profitably resell at the high price. Buying at the low price puU upward pressure on that price, while selling at the high price puts downward pressure on the high price. Hence arbitrage causes the prices to uldmately converge, thereby-destroymg-pricediserimi-nadon.

It is for this reason that goods sold under price discriminadon tend to be services that are difficult or impossible to resell. Examples are medical and dental care; dckets to movies, concerts, and plays; computer time; udlity charges and the like, In other cases the transactions costs may overpower arbitrage opportunities when prices differ Cement, for example, may be sold at different prices in different geographical areas, depending upon the regional elasticity of demand. Nonetheless, high transportation costs and the perishability of the product may preclude arbitrage and transhipment from low to high price regions. We can also turn the example around a bit. For example, if cement is not transhipped from low to high price purchasers, we suspect either: (1) the parties have perpetually bad itiformation regarding their opportunities, an unlikely situation as time goes on, (2) transactions costs are prohibitive, or (3) transportation costs wipe out any gains from arbitrage.

These conclusions stem from perfecdy general mental habits of economists. Throughout your studies of economics, continue to look.for opportunities for transactors to gain either added utility or added prof-



[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [ 47 ] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75]