Sunday, September 1, 2013

Limited Time Offer! Temporary Sales and Price Rigidities

Even though prices change frequently, this does not necessarily mean that prices are very flexible, according to a new paper by Eric Anderson, Emi Nakamura, Duncan Simester, and Jón Steinsson. In "Informational Rigidities and the Stickiness of Temporary Sales," these authors note that it is important to distinguish temporary sales from regular price changes when analyzing the frequency of price adjustment and the response of prices to macroeconomic shocks.
"The literature on price rigidity can be divided into a literature on "sticky prices" and a literature on "sticky information" (which gives rise to sticky plans). A key question in interpreting the extremely high frequencies of price change observed in retail price data is whether these frequent price changes facilitate rapid responses to changing economic conditions, or whether some of these price changes are part of “sticky plans” that are determined substantially in advance and therefore not responsive to changing conditions.  
We use an exceptionally detailed dataset on retail and wholesale prices to investigate these questions. Our dataset has the advantage of providing accurate administrative measures both of the Regular Retail price and of the retailer’s marginal cost—i.e., the Base Wholesale Price. We find that temporary sales are unresponsive both to movements in the Base Wholesale Price and to movements in underlying production costs..." 
They provide some interesting institutional features of temporary sales and promotions:
"Due to the logistical complexity of holding successful temporary sales and associated promotional activity, manufacturers and retailers jointly set a schedule for temporary sales – a promotion calendar – through an annual planning process. This means that temporary sales follow sticky plans...Wholesale price drops associated with trade deals typically do not represent commensurate drops in the retailer’s marginal cost, since the retailers is “spending down” a finite trade deal budget.  
We believe that there are two other notable institutional features of trade deal budgets. First, the most common way that manufacturer trade deal budgets are determined is via accrual accounts, which are analogous to frequent flyer accounts. Just as consumers accumulate “miles” when they fly on, say, United Airlines, retailers accrue funds in a manufacturer trade deal budget for the total volume purchased from a manufacturer. Second, payment from the manufacturer budget to the retailer is typically contingent on execution of a trade deal. Retailers typically receive money after there is verification of a price discount, in-store signage or advertising of the manufacturer’s project."
They conclude that regular (non-sale) prices exhibit stickiness, while temporary sale prices follow "sticky plans" that are relatively unresponsive in the short run to macroeconomic shocks:
"Our analysis suggests that regular prices are sticky prices that change infrequently but are responsive to macroeconomic shocks, such as the rapid run-up and decline of oil prices. In contrast, temporary sales follow sticky plans. These plans include price discounts of varying depth and frequency across products. But, the plans themselves are relatively unresponsive in the near term to macroeconomic shocks. We believe that this characterization of regular and sale prices as sticky prices versus sticky plans substantially advances an ongoing debate about the extent of retail price fluctuations and offers deeper insight into how retail prices adjust in response to macroeconomic shocks."


  1. Even if the "promotional plan" is sticky, I still wonder whether the frequency (or probability) and the degree of promotional discounts does depend on macroeconomic factors, and if this helps provide a degree of flexibility in average market prices. Anecdotally, I would say that marketers are much more willing to engage in discounting and similar promotional tactics when they perceive economic conditions to be weak.

    Paradoxically, from a psychological point of view, the mystery is that prices are not MORE rigid than they are! This suggestion might provide an answer to that, if not to the converse economic question.

  2. Interesting local example of "sticky" planning: California has taxes based on retail store inventories at the end of January, so many stores have a big end-of-January sale, which is pretty much unconnected to the overall state of the economy. (I'm willing to consider that sale prices might reflect current economic conditions in some fashion, but I'm not enough of a wonk to seek out the details.)


Comments appreciated!