Canned tuna is among the most popular seafood products in the U.S. in terms of per capita consumption. There are currently three major brands of canned tuna in the U.S. retailing industry. These are: Bumble Bee Foods (BB), Chicken of the Sea International (CS), and StarKist Seafood Co (SK). The canned tuna category is noted for a high degree of competition among the brands. However, retailers have alleged that the three brands colluded to fix prices. The U.S. Department of Justice (DOJ) initiated investigation into price-fixing allegations, and there are reports the companies involved have reached agreements. In this paper, we analyze the strategic price responses among the top brands in the canned tuna market using national-level barcode scanner data.
We use nationally representative brand-level data on monthly sales quantity, prices and promotional efforts for canned tuna in U.S. superstores (stores with annual turnover of at least $2 million) for the July 16th 2005 to June 12th 2010 period. The data are barcode scanner data on canned/ shelf-stable tuna products, procured from The Nielsen Company. Data consist of information on the three major brands (BB, CS and SK) and a composite brand representing all store brands (SB) or private label brands. Names other identifiable features of the brands are not shown in the results to protect confidentiality.
Figure 1 shows the average unit prices for the four brands. There are three brands (brands 1, 2 and 3), whose prices move more closely together and show larger variations. Whereas, the price of brand 4 is the lowest among all brands and shows smallest variations.
Our econometric model is a system of demand functions and price reaction functions that incorporates both short-term and long-term dynamic responses. We use impulse response functions (IRF) to characterize the response to a one standard deviation shock in the price of the prices. We use Forecast Error Variance Decomposition (FEVD) to decompose the change in a price index into contributions of innovations in the different endogenous variables of the system.