بی ثباتی سهم بازار : استفاده از آزمون ریشه واحد برای صنعت سیگار

This paper uses market share data to infer the nature of rivalry in the U.S. cigarette industry over the 1934–94 period. Unlike previous studies, which measure rivalry from various constructs of market share instability, we examine the time-series properties of market shares to determine whether or not rivalry is evident. Our empirical results imply that a majority of firm-level market shares are martingales, suggesting market shares have been unstable from 1934–94. This result leads us to conclude that rivalry in the cigarette industry has remained strong.

Much of the literature in industrial analysis centers on uncovering the nature of rivalry within an industry. To this end, market share data has played a key role in understanding the level of intraindustry competition. For example, several studies (e.g., Hymer and Pashigian 1962, Caves and Porter 1978 and Sandler 1988, and Eckard (1991)) use market share data across all, or the largest, firms in an industry to construct industry-wide measures of market share instability. Because greater volatility in temporal market shares is indicative of the push-and-pull tactics of intense rivalry, these studies argue that greater instability of market shares coincides with greater rivalry in the industry. Another line of work due to Rhoades (1985) and Rothschild et al. (1991), among others, infer rivalry from movements in the market shares of individual firms.1 In this study we adopt an alternative technique to infer the nature of rivalry from individual market share data. Analyzing the time-series properties of market shares, we use unit root tests to examine whether market shares have conditionally converged, or are stationary in the statistical sense. Applying this procedure to the cigarette industry over the 1934–94 period, our empirical results indicate that the majority of market shares in the cigarette industry are martingales, or follow a random walk. This finding suggests that market shares of cigarette manufacturers were unstable, or did not revert to their long-run mean after a shock between 1934–94. As such, the ability of a firm to capture market share at the expense of rivals has remained strong in this industry.

This paper used time-series techniques to offer an alternative method of uncovering the nature of rivalry in an industry. Applying the concept of stochastic convergence to 1934 –94 market share data from the U.S. cigarette industry, our findings imply that relative market shares in the cigarette industry tend to be unstable, or follow a random walk. Hence, this suggests that rivalry has remained such that the ability to perpetually capture or lose market share remains intact. Additional research is warranted. Although our findings have important implications from a policy, theoretical, and empirical perspective, it is commonly understood that unit root tests have limited power to reject the null, even after allowing a structural break in the series. Consequently, we stress that our approach should not be used as the sole indicator of rivalry in an industry, but rather should be used in conjunction with other measures of industry behavior.