Blockbuster or Niche? Competitive Strategy Under Network Effects

Yinbo Feng, Ming Hu
2017 Social Science Research Network  
We provide a theory unifying the long tail and blockbuster phenomenon. Specifically, we analyze a threestage game where the firms first make entry decisions, then decide on the investment in its product and lastly customers sequentially arrive to make purchase decisions based on product quality and historic sales under the network effect. We analytically show that a growing network effect always contributes to the demand concentration on a small number of products. However, product variety and
more » ... uality investments, as an outcome of firms' ex-ante competitive decisions, may increase or decrease, as the network effect grows. When the network effect parameter is smaller than a threshold, the increasing network effect would shift more demand towards the products with higher qualities, preempting more products from entering the market ex ante and inducing firms to adopt the blockbuster equilibrium strategy by making larger quality investment. When the network effect is stronger than the threshold, the increasing network effect would make the market easily concentrated to a few products. Even some low quality ones may have chances to become a "hit." Interestingly, in this case, the ex-ante equilibrium product variety would be broader and firms adopt the niche equilibrium strategy by maker smaller quality investment. We empirically test the theory with the movie box office data and find strong supporting evidence. * We are grateful for the financial assistance provided by the NET Institute, 1 digitized world. For example, Chris Anderson pointed out in his book that a significant portion of's sales come from obscure books that are not available in brick-and-mortar stores. The same seems also true for the music industry. However, some other cultural industries may tell a different story. Elberse (2008) argued that the importance of individual best sellers in the homevideo industry is not diminishing over time. The concentration of most popular titles is growing. Evidence emerges against the long tail theory in the movie and video game industries as well. Stewart (2013) reported that studios have been shifting their resources towards the big-budget movies. That is what is known as a blockbuster strategy-focusing the limited production and marketing resources on a small number of likely best-selling products. This strategy is based on the notation that the "hit" product still plays an important role in today's world. In this paper, we aim at providing a theory unifying the long tail theory and blockbuster phenomenon. We postulate one key factor behind these phenomena as the network effect-the effect that the consumption of a good by one has an impact on the value of that product to others (Economides 1996) . Intuitively, the blockbuster phenomenon has a lot to do with the (positive) network effect. For one thing, people who are apt to purchase the popular product due to the network effect would make the popular even more so. For another thing, producers, like studios, are willing to put more resources, e.g., advertising budgets and production resources, to produce a possibly blockbuster anticipating that the product will become a "hit" in future due to the network effect. However, no one can exactly predict a blockbuster. A high-budget, mainstream productdesigned to cater to a broad range of tastes (Sun 2012) does not necessarily turn out a blockbuster. A low-budget, niche product-designed to cater to only a small group of customers (Sun 2012) may be a surprising blockbuster ex post, if it becomes popular sufficiently early on due to the market uncertainty and strong network effect. This can attract more and more niche products to enter the market while very few of them may gain significant market shares. In this sense, the (positive) network effect seemingly can also contribute to the phenomenon of long and flat tail. Thus, given the increasing network effect over time it is not surprising that in the iTunes App Store the downloads are more concentrated to a handful of top games while a rapidly increasing number of games sell very rarely or never. Roughly estimated, the number of available gaming apps in the iTunes App Store increased from 151,460 in 2013 to 631,091 in 2016. The sales of the100th game at the yearly grossing top increased from $2.8 million in 2013 to $6.3 million in 2016 but the sales of the 500th game decreased from $0.38 million in 2013 to $0.25 million in 2016. Around the 5000th, games' sales are very close to zero in 2013-2016. Nevertheless, how the network effect exactly leads ∞
doi:10.2139/ssrn.3049370 fatcat:wdtlys7jcnhqjbwh4ifrwd7pk4