Rigid Prices: Evidence From U.S. Scanner Data

Abstract

This article uses weekly scanner data from two small U.S. cities to characterize time and state dependence of grocers' pricing decisions. In these data, the probability of a nominal adjustment declines with the time since the last price change. A store's price for a particular product typically goes through several price changes in rapid succession before settling down. We also detect state dependence: The probability of a nominal adjustment is highest when a store's price substantially differs from the average of other stores' prices. However, extreme relative prices typically reflect the store's recent changes instead of changes in average prices.

Sticky Prices and Costly Credit

“…We account for realistic durations, large average changes, many small and negative changes, and repricing behavior that depends on inflation without such devices. We get a decreasing hazard, which is problematic for other models (Nakamura and Steinsson, ), and price dispersion at low or no inflation, consistent with evidence but not other models (Campbell and Eden, ).…”

Logit Price Dynamics

“…We see that with low costs (squares), the monthly hazard rate reaches a minimum of 3% when the price is 4% above the 45° line. The hazard rate rises steeply away from this minimum; indeed, selection is strong here compared with the evidence of Campbell and Eden (, figure ) or Eichenbaum, Jaimovich, and Rebelo (, figures –). In contrast, with high costs (stars) the hazard rate is flatter overall, reflecting a weaker selection effect when the volume of sales is lower.…”

“…Many alternative models, including the Calvo () model, instead imply that adjustments get larger, quickly, as the time since the previous change increases. Also, we find that the highest and lowest prices are more likely to have been set recently than prices near the center of the distribution (Campbell and Eden ). Finally, prices are more volatile than costs, as documented by Eichenbaum, Jaimovich, and Rebelo (), whereas the opposite is true in both the Calvo and FMC models.…”

Real Rigidities and Nominal Price Changes

“…State-dependent pricing is a key feature of our investigation. Recent evidence for state-dependence includes Eichenbaum et al (2011), Gagnon et al (2012, Campbell and Eden (2014), and Alvarez et al (2015). Our use of state-dependent pricing is in contrast to much of the literature incorporating real rigidities, which has assumed exogenously time-dependent pricing.…”