Upward Pricing Pressure in Two-sided Markets: Corrigendum

The original code producing the results in Affeldt et al. (2013) did not calculate the two-sided diversion ratios correctly. In this corrigendum, we present results that are obtained using revised code. The code is also provided as an updated replication package.

Tables 2 and 4 are affected by the coding error. Here, we present corrected versions of those tables. Comparing Table 2 to the original one in the published article shows that diversion ratios in the bottom panel and in the last column of the top panel are affected. In the article, we point out that two-sided diversion ratios on the advertising side are positive and explain why this is to be expected. This is still the case.

Comparing Table 4 to the corresponding one in the published article shows that results for two-sided upward pricing pressure (UPP) measures are affected mainly on the advertising side. The necessary efficiency credits are lower than previously reported. Nevertheless, there is evidence for UPP on both market sides, while we have already documented in the article that one-sided UPP measures only suggest this to be the case for the readership side. So, importantly, the main conclusion that one cannot, in general, use one-sided UPP measures in two-sided markets, still holds.