A Two‐Period Repeated Game Advertising Investment Model for Oligopolistic Markets with an Application to the Beer Industry*
提出了一个分析寡头市场广告竞争的简单模型，使用对数-对数销售反应函数，假设广告效果持续两期，并求解纳什均衡。用啤酒行业数据验证，比较了不同策略下的广告支出。
This paper proposes a simple analytical model of advertising competition in oligopoly markets. The widely used log‐log sales response function underlies the model specification. Advertising carryover effects are assumed to persist for one period following the period in which the expenditure occurs. Firms are assumed to be engaged in a repeated competitive game in which in every period advertising levels are set such that they maximize current and next period (i.e., two‐period) profits. A Nash equilibrium solution is sought for the game. Compared with previous empirical studies of advertising competition in a game theoretic framework, the proposed model offers the following advantages: (1) oligopoly, not duopoly, markets are analyzed; (2) industry sales is allowed to vary over time as a function of advertising expenditures; (3) non‐zero discount rates are used for the players. An empirical application is provided using data from the beer market on sales and advertising expenditures of Anheuser‐Busch and Miller Brewing. Comparisons are provided with policies that ignore the dependence of next period profits on current advertising levels, reaction function strategies and spending levels obtained from a market share game. Extension of the model formulation to multiple marketing instruments is briefly discussed.