Advertising Part 3: Advertising and Market Structure

This article will be the final installment of my advertising series and it will be about the empirical findings from studies of the relationship between advertising and market structure looking at data from three industries: cola, photographic film, and saltine crackers.

Works in this area saw a boom in the 80s and 90s and are still frequently studied under the broad banner of empirical industrial organization. The inner workings of the empirical models employed in these market structure studies are very complex. Hence let’s focus on what the findings and implications tell us.

First up is Gasmi, Laffont and Vuong’s (1992) pioneering work using data on the Coca-Cola and Pepsi-Cola markets over the period 1968-86.  The specified authors presented a novel methodology to study market structure and rivalry between firms. The general idea of their approach is to specify a demand function for each product (so that demand depends on prices and advertising intensity), and a cost specification for the firms. With these two sides of the market specified, the authors derive first order conditions (for different oligopoly games) and simultaneously estimate the demand functions and first order conditions. Within this setup the authors find that in the Cola industry advertising is quite combative and used as a form of competition: an increase in advertising spending did not seem to enlarge the market, but rather worked to redistribute the market share between the two firms.

Another paper of this sort is Kadiyali (1996). Kadiyali lays down a general model like the one explained above and derives the first order conditions. This paper however looks at the U.S photographic film industry over 1970-90. This industry was largely a monopoly of the Eastman Kodak Company until Fuji entered the market mid-sample in the 80s. Using a data set that includes firm-level price, advertising, quantity and input cost data from the pre and post entry periods, Kadiyali finds evidence that Kodak maintained their monopoly position with high levels of advertising in the pre entry period. During the 80s Kodak accommodated for the entry of Fuji due to the cost advantages of Fuji, and during the post entry period her estimates imply that Kodak and Fuji were colluding in their price and advertising strategies. This is quite a different finding from that of the previous paper.

Slade (1995), looking at different brands of saltine crackers finds evidence that the sales of a brand of crackers increases with its own advertising and decreases with rival brands’ advertising. This suggests a predatory role for advertising, however in a less profound manner compared to the findings of Gasmi, Laffont and Vuong (1992), since total sales for all brands seem to increase with advertising.

Depending on what industry and which market we look at, advertising may be a tool utilized for the purposes of mutually beneficial coordination, or a weapon for aggressive competition over market share. It also turns out that the role of advertising can be somewhere in between.  Once again, the results from various studies of advertising and market structure are equivocal.

After this series of articles on advertising, what have we learnt for sure? Can economists define the role of advertising? At this point in seems as if the answer is: “it depends”.



Gasmi, F & Laffont, J J & Vuong, Q, (1992),Econometric Analysis of Collusive Behavior in a Soft-Drink Market”. Journal of Economics & Management Strategy, Wiley Blackwell, 1(2), pages 277-311, Summer.

Kadiyali, V, (1996) “Entry, its deterrence and its accommodation: A study of the US photographic film industry”. Rand Journal of Economics 27 (3), pages 452–478.

Slade, M. E. (1995), “Product Rivalry with Multiple Strategic Weapons: An Analysis of Price and Advertising Competition”. Journal of Economics and Management Strategy, 4 ,pages 445-76.