Do companies which try to do new things succeed and why? Phin Upham dicusses a seminal work on the topic
In Diversification, Ricardian rents, and Tobin’s q, Cynthia A. Montgomery and Birger Wernerfelt present a study on the ability of a multimarket firm to diversify its resources (factors). Montgomery and Wernerfelt test whether a multimarket firm’s average rents decrease the further a field they transfer resources from their core business. The results of this study have an important affect for the logic of a multimarket firm which is trying to maximize profits.Further, the essay speaks eloquently to the effect of factor specificity on firm profit. The more specified the factor assets, the less valuable alternate use derived from then but the higher their rents, the less specified and the more valuable the alternate use but also the lower comparable rents in their original use.
This essay contributes to the theory of the economic literature in which it lies, but it also indirectly contributes to management literature and has some very interesting potential implications for business organization. If a firm has a core market and there is some slack or market failure for the firm’s factors in this market, the firm might do well to transfer these resources to another market. But two problems arise, both of which Montgomery and Wernerfelt describe and differentiate clearly. Firstly, the authors argue that the more the diversification the more average firm rents are expected to decrease. This is supported by the double point that 1) the wide diversification implies less specified assets which can be so diversified, and 2) the factors transferred should have some decrease in rent generating ability when transferred to a new market. The second problem with diversifying for a firm relates to the value of the factors. Montgomery and Wernerfelt make two claims about this: 1) the more distant the factor transfer form the old market (as measured by the critical factors which the new market differ from the old market) the lower the rent and 2) the more specialized the factors are to the original market both the more rent they extract and the greater loss they have in rent extraction capacity when transferred.
Two problems with this analysis arise in my mind. The first involves the reasons for a firm to diversify. The article suggests that the prevailing theory about reasons for a firm to diversify its factor inputs and become a multimarket firm are because of a market failure in its original market which force it to redeploy resources.Later in the essay, Montgomery and Wernerfelt suggest that an alternate and potentially viable reason for diversification is the tendency for firms to inefficiently invest excess money in other ventures. But there are, it seems to me, other viable reasons for multimarket firms. Firstly, a Coasian analysis ought to provide some room for a firm to internalize some of its inputs or some other related field that it might be able to use to cross-sell or use in conjunction with its original market products. But efficiencies of scale might also force the firm to sell some of the output of this second product externally. Secondly, and this topic is brushed over in p174, a firm can earn rents based on a name-brand. One can imagine that a multimarket firm might enter another market and leverage its name brand in the first market to extract even higher rents in the second market than in the first (thought it is its presence in the first that allows for this). For example, Harley Davidson might be able to sell jackets based on its strength in motorcycles. But it seems as if this is NOT the kind of diversification that the authors wish to discuss. Instead they wish to discuss what happens when factors must be TRANSFERRED in order to enter new markets.Lastly, and secondly, a firm which had experienced serious market cycle effects might develop another related business that it could transfer its resources to during downturns. If these two markets acted inversely, one could imagine a situation where rents remained even but factor transfer was often taking place.Nevertheless, the story of a firm who has a core capacity and does less well outside this capacity is a compelling one.
Montgomery and Wernerfelt use Tobin’s q as a measure of rents.They are dissatisfied with the inability of accounting measures for rents to consider differences in systematic risk, temporary disequilibrium effects, tax laws, and arbitrary accounting conventions. But, assuming the efficient market hypothesis, market price will not be adequate for the author’s purposes since they are interested in the changes of levels of value in the firm not the changes in firs value per se. Thus, Tobin’s q, “the ration of market to the replacement cost of the firm, is a more appropriate measure” (177). Tobin’s q is a balanced mix of accounting and capital market data, which helps to offset the distortions inherent in each part of this ratio. The authors decompose the numerator in q into the sum of the firm’s capitalized income streams, which are available to the authors.
Using Lindenberg and Ross’s (1981) estimations for q in a random sample of 246 firms, the authors combine this data with financial data from various sources. This reduced their sample size to a respectable 167. With some mathematical manipulation and some assumptions about the nature of the Ricardian rents, the authors derive results that support the hypothesis that as firms diversify their profits decline. This does not imply that they are not maximizing value, though, only that their Rents are decreasing relative to their old rents (when they were used in the previous market at time t1). I still have some reservations on how to interpret this result. Clearly, the alternate use of the factors for the firm are more profitable in the adjacent market than in the original market. If the firm had continued to produce at full capacity in the old market, it would have sold its goods at a rate which would have been below the alternate use of these resources. So what we are really saying is that the new use is less desirable than the old use at the old price. But it seems to me that this is comparing apples and oranges.The previous price is no longer available, why is it being used as a benchmark? Nevertheless, the results are shown to be powerful with a .293 and .296 R^2. Montgomery and Wernerfelt conclude by discussing alternate hypothesis, all of which they find implausible except the previously mentioned one in which firms are investing in industries needlessly because they have extra cash.
Samuel Phineas Upham has a PhD in Applied Economics from the Wharton School (University of Pennsylvania). Phin is a Term Member of the Council on Foreign Relations. He can be reached atphin@phinupham.com.