In analyzing production costs, economists draw a useful distinction between the long run and the short run. Up to this point our analysis of production and cost has dealt only with the latter, which is defined as a period short enough so that one or more factors of production are effectively fixed in quantity. In the case of manufacturing - the conventional example dealt with in microeconomics - the fixed factor is plant and equipment. The short run is the period in which the manufacturer has to decide how much output to produce, and at what prices, from the firm's existing plant. The long run is defined as a period long enough so that all factors of production become variable. In the manufacturing case, that means a period long enough so that the firm could plan to build one or more new plants, discard old ones, or even go into a new line of business.
What are the analogous definitions in the live performing arts? In the case of plays offered in the commercial theater, the short run is the life of a single production. Over that period the play's production costs, comprising all expenses that were committed before opening night, are a fixed cost. The producer's decisions are limited to deciding how long the play should run and at what ticket prices. The long run in the commercial theater is a period over which the producer can contemplate mounting additional plays or musicals, different in type, larger or smaller in scale, in the same or other venues.
We argued that it is unrealistic to think in terms of single productions in the not-for-profit sector, since most companies put on a season or repertory of productions each year. This logic applies equally well to theater, opera, ballet and modern dance, and symphony concerts. Accordingly, the short run for the nonprofit performing arts organization is best defined not as the run of a single production, but as the length of one season. The season is the planning unit. The individual productions are conceived not singly but as a package, complementary to one another. While it may sometimes be possible to change course in midseason, that is rarely done. Thus, the production costs for a given season are essentially fixed once the season begins, which gives "the season" its economic character of being "the short run." The long run is then a period longer than a single season. In the long run, management can contemplate putting on more or less elaborate productions, a longer or shorter season, a season comprising a larger or smaller number of individual productions, or it can move to a different venue or even go out of business.
One of the most interesting questions that can be asked about any production process is, How do costs per unit of output behave when the scale of production increases? By scale, economists mean the size of the producing enterprise, with size measured by physical output when plant and equipment are operated at designed capacity. If we plot scale on the horizontal axis and average unit cost on the vertical, we generate the firm's long-run average cost curve. So another way of putting the preceding question would be, What is the shape of the firm's long-run average cost curve?
The question is interesting in part because there is no a priori answer; each industry has to be investigated empirically. Three possible long-run cost patterns are denoted as follows: economies of scale, if unit cost falls as the scale of output increases; constant returns to scale, if unit cost is unchanged; diseconomies of scale, if unit cost rises. In most cases the outcome will be some combination of these tendencies. In manufacturing, for example, firms in most industries enjoy economies of scale up to some minimum efficient size, after which there is a broad range of output marked by constant returns to scale. At very large scales, diseconomies may set in. Evidence on this last point, however, is hard to come by. Indeed, in a competitive world we would not expect to find many firms that had expanded into a range where unit costs were increasing, since such behavior would be self-destructive.
Since the season is the planning unit for most nonprofit performing arts enterprises, length of season, as measured by number of performances - or, for symphony orchestras, the number of concerts - is the appropriate indicator of scale for studying the behavior of costs in the long run. (This is analogous to the use of plant size as the measure of scale in the manufacturing case.)
The existence of economies of scale in the live performing arts has been confirmed empirically a number of times. Baumol and Bowen, probably the earliest to do an empirical study, found that for most of the eleven symphony orchestras in their sample, cost per concert fell significantly as the number of concerts per year increased. In the typical case, unit cost did not decline over the entire range of outputs. Rather, it reached a minimum at a point that varied across orchestras at somewhere from 90 to 150 concerts per year and then leveled off.8
8. William J. Baumol and William G. Bowen, Performing Arts: The Economic Dilemma
(New York: Twentieth Century Fund, 1966), pp. 201-7, 479-81.
The authors speculated that the observed economies of scale probably arose from two sources. Most important would be the fact that up to some point an orchestra can play more concerts without investing in more rehearsal time. For example, if it sells subscriptions in three series (say, Thursday evenings, Friday evenings, and Sunday afternoons), it can perform the same music three times a week. If demand picks up to the point where a fourth series is justified, it can play the same music a fourth time without additional rehearsal expense. A second source of economies of scale (probably less important) is the fact that the administrative expense of running the orchestra need not increase with each increase in the number of concerts performed. Thus, "overhead" can be spread over more output, reducing the level of average fixed cost per concert as the season lengthens.
Savings analogous to both economies of scale should be available to producers of theatrical repertory, opera, ballet, or other kinds of dance. Hence, we would expect to find economies of scale operating in all kinds of live performance art. Steven Globerman and Sam H. Book, using a somewhat different methodology than Baumol and Bowen, studied a sample of Canadian symphony orchestras and theater companies. They confirmed the existence of economies of scale in orchestra performance up to a level of about 115 performances per year. For theater companies they found that economies of scale extended much farther: "minimum cost per performance ... was obtained at approximately 210 performances."9 They surmised that the greater economies of scale available to theater groups reflected higher fixed costs per production in theatrical activity, as compared with symphony concerts.
Finally, in 1985 Mark Lange and his coauthors, using a much larger data set than Baumol and Bowen and a different econometric technique, also confirmed the existence of economies of scale for symphony orchestras.10 They found that average cost per concert declined as output rose from one to 65 concerts per year, was constant over the wide range between 67 and 177 concerts, and rose sharply at higher outputs. The authors speculated that greater commitments to touring or special events, or other differences in the type or quality
9. Steven Globerman and Sam H. Book, "Statistical Cost Functions for P erforming Arts
Organizations," Southern Economic Journal 40, no. 4 (April 1974): 668-71, cited at 671.
10. Lange et al., "Cost Functions for Symphony Orchestras," pp. 71-85.
of output, might explain the higher unit costs encountered by orchestras giving the most concerts per year, but they were unable to confirm that statistically.
Was this article helpful?