Identifying Price Reviews by Firms: An Econometric Approach

• Published date: 01 March 2020
• Author: MARK N. HARRIS,HERVÉ LE BIHAN,PATRICK SEVESTRE
• Date: 01 March 2020
• DOI: http://doi.org/10.1111/jmcb.12675

DOI: 10.1111/jmcb.12675

MARK N. HARRIS

HERV´

ELEBIHAN

PATRICK SEVESTRE

Identifying Price Reviews by Firms: An

Econometric Approach

Price reviews are a potentially costly activity.A significant fraction of un-

changed prices may stem from firms not reviewing prices, rather than from

obstacles to changing prices per se, such as menu costs. In this paper, we

disentangle these two causes of price stickiness by estimating an inflated

ordered probit model on a panel of French manufacturing firms. The results

point to a low frequency of price reviews, suggestive of the relevance of

information costs as a determinant of the observed price stickiness. In view

of the “inattentive producers” literature, pointing that the source of price

rigidity matters, this is suggestive of a large real effect of monetary policy.

JEL codes: C23, C25, E31

Keywords: price stickiness, price reviews, price changes, inflated ordered

probit model.

THE WIDELY DOCUMENTED EXTENT OF price stickiness (see, e.g.,

the survey by Klenow and Malin 2011) suggests that most firms do not systematically

change their prices every time there is a change in their environment.Such a f eature is

corroborated by surveys, that are able to control for changes in wage and other input

The authors wish to acknowledge the financial support of the Australian Research Council and the

Banque de France research foundation. The authors thank the editor, as well as Francesco Lippi and

several referees for helpful comments and suggestions. Theyalso thank participants to seminars at Banque

de France, Paris-Est Cr´

eteil University,Aix-Marseille University, Monash University and to the 26th JMA

conference (University of Dijon), the RCEF conference (University of Rimini), and the 16th International

Conference on Panel Data (University of Amsterdam) for their comments on previousversions. The usual

disclaimer applies. In particular, the viewsexpressed herein are those of the authors and do not necessarily

reflect those of the Banque de France.

MARK N. HARRIS is John Curtin Distinguished Professor at the School of Economics, Finance and

Property, Curtin University (E-mail: mark.harris@curtin.edu.au). HERV´

ELEBIHAN is Deputy Head of

Monetary and Financial Studies Directorate at Banque de France, and Visiting Researcher at Curtin

University (E-mail: herve.lebihan@banque-france.fr). PATRICK SEVESTRE is a Professor of Economics

at Aix-Marseille University, CNRS, EHESS, Centrale Marseille, and AMSE (E-mail: patrick.sevestre@

univ-amu.fr).

Received April 20, 2018; and accepted in revised form May 15, 2019.

Journal of Money, Credit and Banking, Vol.52, Nos. 2–3 (March–April 2020)

C

2019 The Ohio State University

294 :MONEY,CREDIT AND BANKING

prices changes for firms, as well as demand fluctuations (e.g., Blinder et al. 1998,

Fabiani et al. 2006, Vermeulen et al. 2007, Loupias and Sevestre 2013). Another well-

established fact is that most firms declare they review their prices only infrequently

(Fabiani et al. 2006, Blinder et al. 1998). Reviewingtheir price, as we interpret it in the

present paper—consistently with the relevant literature—is the process for firms of

collecting and processing the necessary information to compute their optimal desired

prices.1Having computed this optimal desired price does not however necessarily

imply that the firm will actually implement a price change, in particular if there is a

cost to implementing such a price change. Thus, the absence of price review is only

one of the possible causes of price stickiness.

This paper adds to the literature on price setting by providing a microeconometric

assessment of the respective roles of firms’ decisions to undertake price reviewsand,

possibly, a subsequent price change, for explaining the observed price stickiness at

the firm level. Tothe best of our knowledge, no such previous assessment is available

in the literature.

For that purpose, we estimate an inflated ordered probit (IOP) model (Harris and

Zhao 2007, Brooks, Harris, and Spencer 2012) on firm-level data. The structure of

this model is well suited to the modeling of the respective roles of price reviews and

price changes in firms’ price-setting decisions. Indeed, the absence of a price change,

a commonly observed outcome at the firm level at a monthly or quarterly frequency,

can be seen as the consequence of a two-step process: at any point in time, firms decide

whether or not to undertake a price review: “no review” results in prices remaining

unchanged (unless there is an “automatic” price-setting rule); while implementing

a review can lead to prices being consequently either changed, or left unchanged.

This sequencing is not observable by the applied economist: all she observes is the

large proportion of no-change outcomes, without knowing how these were arrived

at. The IOP model is typically appropriate in such a setup. In this respect, our paper

also adds to the applied econometric literature by providing one more example of the

capability of the IOP model to identify the respective determinants of the two phases

of an observationally equivalent decision process.

We use a panel data set of French firms, setup by merging the monthly business

survey ran by the Banque de France (BdF) with other data sources. A key feature of

this data set is that a number of time-varying firm-level determinants of price changes

are available. While a disadvantage of these data is that the information on price

changes is only categorical, the use of an IOP model precisely handles this feature of

the data.

Another advantage of our econometric approach is to circumvent restrictions that

would be embedded in a standard Ordered Probit (OP) model approach, for example,

and would be particularly harmful in a price review setup. In a standard OP model,

finding that a variable affects the probability of a price change entails restrictions on

the sign of its effect on the change in prices. For instance, if a positive coefficient is

1. Importantly, this definition of price review does not preclude that the firm may have access to some

information about its environment between two price reviews.