نمونه های آزاد، سود و رفاه اجتماعی: اثر ساختار بازار و حالت های رفتاری

This paper addresses an important and underresearched issue in the economics and marketing literatures: what are the managerial and social consequences when firms use business models that are based on the dissemination of free samples? We develop an analytical model of free samples for both digital and physical goods that addresses three fundamental managerial and social questions. First, what is the effect of different market structures (i.e., monopoly and oligopoly) and cost structures on optimal marketing policy and prices? Second, what is the effect of different behavioral modes on prices and free samples? Third, how do different market structures and behavioral modes affect social welfare? The main conclusion is that a number of standard results do not hold when firms have the option of selling products and of distributing free samples. For example, the optimal strategy for oligopolists who produce homogeneous goods and coordinate their marketing policies is to increase – not decrease – the quantity of sold output. Similarly, under well-defined cost and demand conditions, monopoly can lead to a socially inferior outcome to competition. From a policy viewpoint, the managerial and social welfare implications of free samples depend on the type of market structure (monopoly or oligopoly) and the behavioral modes chosen by the firms in an industry (e.g., whether to coordinate their free sample policies or to behave non-cooperatively).

Business models in a wide spectrum of industries are based on marketing strategies that rely heavily on the distribution of free samples of their products. These industries cover the gamut and range from those that produce physical products (e.g., shampoo and cold medications) to those that sell digital goods (e.g., software). Indeed, today this “two-sided market” model based on free samples has emerged as the revenue engine for all of the biggest Web companies, from Facebook and MySpace to Google. For a succinct discussion see “The Economics of Giving It Away,” The Wall Street Journal, 1/31/2009. Consider the following examples. The Wall Street Journal currently offers a free two-week subscription to its online version, after which subscribers need to pay. Reportedly, the Wall Street Journal is pursuing a strategy of blending free and paid content on its Web site. Tapoulos is a popular music game for the iPhone that is distributed free. The goal is to induce millions of people to try the free version and to encourage a significant fraction of them to upgrade to paid versions. More generally, many online firms base their entire business models on giving out the product for free and obtaining revenue solely from advertisements on their Web sites. These examples bring up a number of critical managerial and social issues. From a managerial viewpoint, will the dissemination of free samples lead to higher or lower prices for paid products? Will the dissemination of free online products on a Web site lead to higher advertising revenues for those Web sites? What are the effects of the market structure (monopoly or oligopoly) and different behavioral modes (e.g., partial or full cooperation) on the quantity of free samples and prices? From a social viewpoint, is it true that, if firms are allowed to distribute free samples, monopoly will always lead to socially inferior results to competition? Are the managerial and social results different for digital goods (those with zero marginal costs) and physical goods (those with positive marginal costs)? Despite the importance and growing use of business models that are based on distributing free samples, the analytical marketing and economic literatures in this area are sparse. In particular, this literature on sampling appears to have focused heavily on digital goods (i.e., products for which the marginal costs are zero and consumers purchase at most one unit) — not on physical goods (i.e., products whose marginal costs are positive). For example, Peitz and Waelbroeck (2006) developed a music download model that allows for a special type of product differentiation in which the profit-maximizing multiproduct monopolist charges the same price across all products (p. 909). Our focus is significantly different from Peitz and Waelbroeck (2006). In contrast to their model, our model focuses on homogeneous products. However, it considers both monopolistic and oligopolistic models in which firms can pursue different game strategies (e.g., partial or full cooperation). Our model considers both digital goods (whose marginal costs are zero) and physical goods (whose marginal costs are positive and vary across both trial and regular sizes). Our model focuses on how firms' marketing policies affect both profits (firm-level and industry-level) and welfare. Finally, our model allows competing firms to choose the intensity of free sampling. Peitz and Waelbroeck (2006) assume that the firm's sampling decision is binary (i.e., whether or not to offer free downloads to consumers). The key managerial and welfare questions we examine are as follows. First, what is the effect of different market structures (i.e., monopoly and oligopoly) on optimal free sample policy and prices? Second, what is the effect of different behavioral modes (e.g., whether firms coordinate their free sample policies or not) on the firm's marketing policy, and how do the results vary across market structures? Third, what is the effect of different market structures and behavioral modes on social welfare when firms distribute free samples? As noted previously, for generality the model considers both digital goods and physical goods. The main finding is that a number of standard results in the literature do not hold when firms have the option of selling products and of distributing free samples. For example, when oligopolists coordinate their free sample policies, both the total quantities of free samples and sold output increase. Interestingly, there are well-defined cost and demand conditions under which monopoly can be socially superior to a competitive market structure. Finally, the managerial and social welfare results depend on the type of market structure (i.e., monopoly or oligopoly) and the behavioral modes chosen by firms (e.g., whether to coordinate their free sample policies or to behave non-cooperatively).

This paper addresses an important but neglected question: what are the managerial and welfare implications of the use of free samples under different market structures (i.e., monopoly and oligopoly)? The main finding is that a number of results in the literature do not hold when the standard pricing model is extended to the case where firms distribute free samples. For example, when oligopolists coordinate their marketing policies and have the opportunity to issue free samples, the optimal policy is to increase, not decrease, sold output. Similarly, if firms can distribute free samples, under certain well-defined conditions, monopoly may lead to a socially superior outcome than competition. In general, the managerial and social results depend on the market structure (monopoly or oligopoly) and on the behavioral modes chosen by firms in the industry. Finally, future research is necessary to address a number of important managerial issues not considered here. Our analysis assumes that all products are homogeneous. Future research should extend the model to allow for general forms of product differentiation by competing firms in an industry. In particular, these models should explicitly analyze the joint effects of product quality and free samples in an oligopolistic market. In addition, the free sample model should be extended to include other demand-generating tools such as advertising and promotion (see Anselmi, 2000). As noted earlier, previous research has examined special cases of product differentiation in a monopoly context (see Peitz and Waelbroeck, 2006). Our model implicitly assumes that consumers adjust their equilibrium behaviors immediately when they receive free samples. In general, a multiperiod model will be necessary to allow for partial adjustments by consumers and incomplete learning over time.