The effects of monopolization on newspaper advertising rates.
It is almost universal assumed that monopolization in an industry leads to higher prices and lower output. The legislature, consumers and introductory textbooks point out the evils of monopolization and the virtues of competition. Economists have known that in theory monopolization can lead to lower prices because of economies of scale, but evidence supporting the benefits of monopolization is virtually never presented. This paper examines the daily metropolitan newspaper industry. Newspaper firms face substantial fixed costs and low marginal costs which creates the potential for a profit seeking monopolists to charge lower prices.
II. A Look at the Impact of Economies
of Scale on Average Costs
Newspapers are faced with high first issue costs; those costs necessary to produce the first copy of the paper which do not fluctuate with circulation. These costs include creating the content, news gathering, purchasing and editing costs, costs of typesetting, and printing plate preparation. James Rosse, Professor of Economics at Stanford University, estimates that first copy costs may account for as much as 40% of total costs for a "small circulation" (pp. 16).
Anumerical example can show the sharp increase in average costs which may be experienced under a duopoly. Let's assume A is the only firm producing a newspaper in a town and the circulation is 300,000 units of production. A's total costs are comprised of first copy costs and the variable costs associated with 300,000 units of production. Now suppose Firm B enters the market and both firms produce half the original circulation and charge the same price per reader for advertising as the monopolist had. Further, assume that all advertisers want to advertise in both papers to reach the complete original audience. Each firm will have to incur the first copy costs plus the variable costs associated with 150,000 units. The total industry's first copy costs will double, while variable costs and total revenue will stay the same. If first copy costs accounted for 20% of the monopoly newspaper's costs, industry costs have now arisen 20%, while industry production has remained unchanged. Average costs are much lower for the monopoly firm (first copy costs/300,000 + variable costs) than the duopoly firms (first copy costs/150,000 + variable costs).
The following table works through a simple numerical example to compare the firm and industry costs under the monopoly and duopoly cases. It will be assumed that fixed costs equal 20% of total costs for the monopoly firm and circulation is 300,000.
Today almost all metropolitan areas in the U.S. have either one or two major dailies; i.e. in each of these cities either a geographical monopoly or duopoly exists. The question is how the cost functions and the pricing decisions differ between duopoly and monopoly firms.
III. Previous Studies
This paper provides estimates of the effects of competition on advertising rates. Using the July 12, 1990, Newspaper Rates and Data, all thirty-five cities with circulation above 200,000 newspapers reporting a price for a standard advertising unit (sau) for black/while retail advertising were included in the data set. Twenty five of the cities included were monopoly cities, while the other ten cities were competitive. Before disclosing my results, however, I will give a brief overview of the previous literature.
George Stigler's 1939 study, A Theory of Oligopoly, investigated newspaper advertising rates as a function of the number of evening newspapers in a city. Using the equation:
Log(milline rate) =
5.194 - 1.689 Log(circulation)
Stigler tabulated the residuals of fifty-three cities. He found that the milline rate (the cost per line of reaching one million readers) was five percent above average in one newspaper towns and five percent below average in two newspaper towns.
In 1969, in a PhD dissertation entitled, An Intra Industry Approach to Measuring the Effects of Competition: The Newspaper Industry, John Henry Landon found through a regression of sixty-eight cities that monopoly newspapers charged higher rates for classified advertising. Landon's independent variables included the city's level of income, the product market concentration, the per capita retail sales, and the circulation of the newspaper.
In 1970, in a PhD dissertation entitled, Changes in the Ownership of Daily Newspaper and Selected Performance Characteristics, 1950-1968: An Investigation of Some Economic Implications of Concentration of Ownership, G. L. Grotta compared the advertising rates of thirty-four cities in which a monopoly structure replaced a duopoly, to a control group of forty cities which had remained either a monopoly or a duopoly throughout the relevant time period. The milline rate increased more in the experimental group than in the control group. Grotta acknowledged that costs would be lower for the monopoly firm, but found that resulting advertising rates were actually higher.
IV. Empirical Evidence
It is well documented that advertising rates increase with increased circulation but at a decreasing rate. Newspaper with greater circulation should, therefore, charge a lower price per reader for advertising space. I attempted to see if advertisers were paying more to advertise in two smaller competitive papers than in one large monopoly paper.
I ran a regression with the sau, standard advertising unit, as the dependent variable. Competition was included as an independent, dummy variable, taking on a value of one in a competitive situation and a value of zero when the city was non-competitive. When the city had more than one newspaper, I combined the data to form one observation. For example, in a two-newspaper town, if one paper costs $120 with circulation 200,000 and the other $60 with circulation 100,000, the city was entered at sau=$180 and circulation=300,000. This method assumes that no one will read both newspapers. Although this is clearly not the case, it does not seem that this assumption discredits these findings.
Therefore, if advertisers received no price reductions, the dummy variable should have been insignificant with respect to the sau. Gross household income per household was included as an independent variable, as it seemed logical that firms would pay more to advertise in wealthier cities.
When all thirty-five cities were included in my regression, the results were as follows:
Log(sau)= - .483 + .830 Log(circulation)
+ .119 (competition) + .146 Log(gross household income per household).
A second regression, only including the twenty-nine cities with circulation between 200,000 and 600,000 circulation, was run with the following results:
Log(sau)= .130 + .624 Log(circulation)
+ .134 (competition) + .306 Log(gross household income per per household).
The results support three conclusions. First, there is greater variance among monopoly newspapers. This my mean that other media forms are not suitable substitutes in some cities. For the 25 cities with a monopoly structure, the sau/circulation range was much larger (.2779 to .6703) than the range for competitive cities (.3379 to .5146). In addition, the standard deviation for monopoly cities was much higher (.1064) than for competitive cities (.0485). The standard deviation for monopoly cities was greater in all observed circulation ranges (200,000-400,000, 400,000-600,000, 200,000-600,000, above 600,000, and all) except 400,000-600,000.
The second conclusion is that papers in competitive cities did not charge less for advertising than monopoly papers did. In fact, in all of the circulation ranges, the mean and median sau was greater for the competitive newspapers with only one exception (the exception being the median in the 200,000-400,000 range). In both regressions, advertising prices were positively related to competition, although the coefficients were insignificant (t = 1.196 for all data, t = 1.176 for 200,000-600,000).
Finally, there may be diseconomies of scale experienced in larger cities. The data set includes six cities with over 600,000 newspapers sold daily, five of which are competitive. As the relevant market grows it may be more costly to further expand. Distribution costs rise as the geographical market grows, and additional printing equipment may eventually become necessary. Workers may begin to require overtime. Therefore, it may be imprudent to state that there is no limitation to the efficient size of the firm.
This paper demonstrated, both intuitively and empirically, that advertisers actually benefit from the geographic concentration of the newspaper industry. Because fixed costs dominant variable costs, the newspaper industry is most efficient under the monopoly case. This result, however, is certainly not unique to this industry. Industries which incur high fixed costs are apt to experience economies of scale, resulting in lower average costs. Each industry, therefore, must be evaluated separately and regulators must analyze all alternatives. The myth that all monopolies are evil should finally be put to rest.
[TABULAR DATA OMITTED]
Compaine, Benjamin M., Who Owns the Media? (White Plains, NY: Knowledge Industry Publications, Inc. 1982). Grotta, Gerald L., Changes in the Ownership of Daily Newspaper and Selected Performance Characteristics, 1950-1968: An Investigation of Some Economic Implications of Concentration of Ownership (unpublished doctor's dissertation, Southern Illinois University, 1970). Landon, John Henry, An Intra Industry Approach to Measuring the Effect of Competition: The Newspaper Industry (unpublished doctor's dissertation, Cornell University, 1969). Owen, Bruce M., Economics and Freedom of Expression (Cambridge, MA: Harvard University, Program on Information Resources Policy, 1981). Rosse, James N., The Evolution of One Newspaper Cities, discussion paper for the Federal Trade Commission Symposium on Media Concentration; Washington, DC, December 14-15, 1979. Simon, Julian L., Issues in the Economics of Advertising (Urbana: University of Illinois, 1970). Standard Rate and Data Service, Newspaper rates and data, Wilmette, Ill.: Macmillan, Inc., July 12, 1990). Stigler, George J., "A Theory of Oligopoly," Journal of Political Economy, LXXII, 1964.
|Printer friendly Cite/link Email Feedback|
|Date:||Mar 22, 1992|
|Previous Article:||An economic approach to abortion demand.|
|Next Article:||Is capitalism kaput?|