|
...extended flexible-price models disputed the empirical finding itself. In this paper, we estimate variety of alternative total factor productivity measures for a representative sample of Italian manufacturing firms and on average find a negative effect of productivity shocks on hours growth. More interestingly, using the reported frequency of price reviews, we show that the contractionary effect is stronger for firms with stickier prices and weaker or not significant for firms with more flexible prices. Price stickiness remains a crucial factor in shaping the response of hours after controlling for product storability or market power.
JEL codes: E30, E32, D24
Keywords: pricing behavior, technology shocks, hours.
IN RECENT YEARS, the comovement of technology and labor at business cycle frequencies has come under growing scrutiny. In standard flexible-price models the correlation is positive because, after a technology shock, prices fall, aggregate demand increases and hours worked rise. By contrast, in a widely cited paper, Gall (1999) reported a negative correlation between productivity and labor, and interpreted it as evidence in favor of sticky-price models. Arguably, after a technology shock, if nominal rigidities prevent prices from falling as much as they would with flexible prices, aggregate demand remains stable or increases only modestly and firms may satisfy it by employing a smaller volume of inputs, which have become more productive. Later work has emphasized that this occurs unless monetary policy fully accommodates technological shocks by lowering interest rates (Dotsey 1999).
Because of the important implications for business cycle theory, Gali's (1999) results have fueled a wide debate in the literature. On the one hand, a number of authors have provided evidence that corroborates the finding of a negative response of labor input to technology shocks. In particular, while Gall estimated a structural VAR on productivity and hours and identified technology shocks as those having a permanent impact on productivity, Francis and Ramey (2005a) extended Gali's identification scheme by imposing additional long-run restrictions and considering a wider set of variables. Basu, Fernald, and Kimball (2006) developed an extended production function framework with proxies for changes in unobserved capital and labor utilization. (1)
On the other hand, several contributions have either disputed Galf's (1999) empirical finding or challenged his theoretical interpretation. On empirical grounds, Christiano, Eichenbaum, and Vigfusson (2003) have argued that the negative correlation found in the above studies is driven by over-differencing of the hours worked data. If hours per capita are assumed to be stationary and the level of this variable is considered, a positive effect of technology on hours is found. This contribution, in turn, has stimulated a debate on the statistical properties of hours per capita (see Francis and Ramey 2005b, 2006, Fernald 2004, Gaff 2004). A positive effect of productivity shocks on hours is also found by Chang and Hong (2006), who use data of U.S. four-digit manufacturing sectors. Fisher (2006) allows for both neutral and investment-specific technology shocks and his findings are largely sensitive to splitting the sample. (2)
On theoretical grounds, a variety of alternative explanations of Gali's (1999) finding are consistent with flexible prices. One class of possible explanations refers to mechanisms through which the adoption of technological progress may somehow disrupt current production, eventually resulting in a decrease in worked hours. For example, reaping the benefits of productivity improvements may require the replacement of existing equipment (Cooper and Haltiwanger 1993), changes in the labor organization (Hall 2000), retraining of the firm's labor force (Campbell 1998), or reallocation of labor across firms (Davis and Haltiwanger 1990). Another type of explanation of Gali's finding, suggested by Francis and Ramey (2005a), calls for habit formation in consumption. In this model, aggregate demand is largely unaffected by technology shocks because consumers have inertial behavior. In principle, another alternative explanation of contractionary productivity shocks hinges on market power. A firm with a higher degree of market power is expected to face a less elastic demand. Even if prices are fully flexible, a relatively low price elasticity of demand may cause output to increase modestly after the price reduction induced by a productivity shock; accordingly, labor input, which has become more productive, might even decline. (3) Moreover, the relevance of price stickiness in the transmission mechanism of technology shocks has recently been questioned by Chang, Hornstein, and Sarte (2004). As originally suggested by Bils (1998), they argue that the labor response to a productivity improvement depends crucially on the degree of storability of the firms' products. In particular, following a technology shock, demand may increase only modestly because of price stickiness, but if goods are storable and the cost of holding inventories is not too high, firms may still choose to increase output and therefore employment.
We contribute to the debate by exploiting the unusual richness of a detailed data set on a representative sample of Italian manufacturing firms. In the whole sample, we find on average a negative response of hours worked to a productivity shock. However, this is not the main result of the paper. Indeed, the macroeconomic relevance of contractionary technology shocks can be better assessed in studies that utilize data spanning the whole economy. On the contrary, the advantage of firm-level data is that they allow us to investigate the microeconomic mechanisms underlying the aggregate result. In particular, by using information on the reported frequency of price reviews and other firm's characteristics we can discriminate between sticky- and flexible-price interpretations. We find that the contractionary effect on hours of technology shocks is strong for firms with stickier prices, while it is weaker or not significant for firms with more flexible prices. We also report similar findings when the number of workers is used as a measure of labor input.
In contrast with a previous contribution (Marchetti and Nucci 2005), where we focused on one particular approach to productivity measurement (Basu, Fernald, and Kimball 2006), in this paper we compute a variety of different total factor productivity (TFP) measures, that together span a large spectrum of theoretical assumptions and models. These estimates are the standard (revenue-based) Solow residual, the cost-based Solow residual, and a model-based measure proposed in the industrial organization literature (Olley and Pakes 1996). For comparative purposes, we also include the estimate of productivity based on Basu, Fernald, and Kimball (2006) already used in our previous contribution.
In assessing whether the response of hours to a productivity shock depends on the degree of price stickiness, we investigate whether our results survive once alternative explanations or additional transmission mechanisms are explicitly allowed for in the empirical framework. In particular, we analyze the role of products' storability and inventory holdings. Furthermore, we assess the relevance of market power versus nominal rigidity as an alternative explanation of the contractionary effect of productivity shocks. We do so by using survey data on the price elasticity of demand reported by each firm.
The remainder of the paper is organized as follows. Section 1 describes the data and documents the relevance of price rigidity across different sectors and degrees of market concentration. Section 2 discusses the various TFP measures used in the empirical analysis. Section 3 investigates the response of hours and employment to productivity innovations and the role of price stickiness. Section 4 controls for product storability and market power. Section 5 draws some conclusions.
1. THE DATA
1.1 Data Sources
We use comprehensive panel data on a representative sample of Italian manufacturing firms. The main source is the Survey of Italian Manufacturing (SIM), carried out annually by the Bank of Italy. The data are of unusually high quality, being directly collected by interviewers who are officials of the local branches of the Bank of Italy, and often have a long-standing work relationship with the firm's management. Each year since 1984 roughly 1,000 firms have been surveyed; because of entries and exits, the balanced panel consists of almost 300 firms. Sample composition is maintained by the statisticians of the Research Department of the Bank of Italy to ensure representativeness with respect to the whole manufacturing sector in terms of composition by branch, firm size, and geographical location. Data drawn from SIM include figures on employment and hours, labor compensation, investment, and capital stock, plus qualitative information on a number of variables that are crucial for economic analysis but are hard to find in the existing surveys. These variables include the typical frequency of price reviews, the extent of the firm's market power, and the degree of concentration of its main market.
Data on gross production (sales plus inventory change), purchases of intermediate goods, and inventories of finished goods are drawn from the Company Accounts Data Service (CADS), which is the most important source of balance sheet data on Italian firms. It covers about 30,000 firms and is compiled by a consortium that includes the Bank of Italy and all major Italian commercial banks.
Merging the SIM and CADS data sets resulted in an unbalanced panel of almost 1,000 firms and 8,000 observations, ranging from 1984 to 1997. The period considered includes three manufacturing-wide expansions (1985-90, 1994-95, and 1997) and two recessions (1991-93 and 1996). Further details on data sources and the definition of the variables can be found in the Appendix.
1.2 Empirical Regularities on Price Stickiness
The information on the degree of price stickiness characterizing the individual firms of our sample was provided by the replies to a question included in the 1996 SIM survey. Firms were asked the following question, with reference to their main product: "How frequently does your firm typically review selling prices?" The managers interviewed could choose from among five possible responses: "Several times a month," "Every month," "Every three months," "Every six months," and "Once a year or less frequently." The replies obtained from 955 firms are summarized in Table 1, first row. The survey found that roughly 30% of the firms typically reviewed prices every 3 months or more often, 35% every 6 months, and another 35% of firms once a year or less often. Therefore, the median frequency of price reviews is twice a year, as in the case of the U.S. firms surveyed by Blinder et al. (1998) and somewhat lower than the quarterly frequency reported for U.K. manufacturing firms by Hall, Walsh, and Yates (2000). (4) In principle, for the purposes of this paper, information on the frequency of actual price changes (or, better yet, on the time elapsed between a shock and the corresponding price revision) would be preferable as measure of price stickiness, since the frequency of price reviews is only one aspect of firms' pricing behavior, though an important one. Unfortunately, such information is not provided by the 1996 SIM survey. However, Blinder et al. document a strong positive correlation at the firm level between the frequency of price reviews and that of price changes (see also Hall, Walsh, and Yates 2000, Table 1). Indeed, the Bank of Italy interviewers who conducted the survey used in this paper reported that the re-examination of prices had often coincided with their actual...
NOTE: All illustrations and photos
have been removed from this article.
Looking for additional articles?
Search our database of over 3 million articles.
Looking for more in-depth information on this industry?
Search our complete database of Industry & Market reports by text, subject, publication
name or publication date.
About Goliath
Whether you're looking for sales prospects, competitive information, company
analysis or best practices in managing your organization,
Goliath can help you meet your business needs.
Our extensive business information databases empower business
professionals with both the breadth and depth of credible,
authoritative information they need to support their business
goals. Whether it be strategic planning, sales prospecting,
company research or defining management best practices -
Goliath is your leading source for accurate information.
|
%20%20&position=1)