When productivity varies greatly among establishments, exit and entry of establishments becomes a critical way for the market to adjust to shocks and substitute capital for labor. When prices fall or wages rise, establishments with high labor costs or labor/output ratios are scrapped, while new establishments enter with low labor requirements. The result is a change in sectoral productivity and labor costs even when individual establishments operate with relatively fixed-labor usage coefficients. While Salter focused his analysis on how new technologies affect a sector, the model is a general one. We apply it to the shoe industry where the competitive pressure comes from increased price competition from imports rather than from new technologies.
To examine changes in the population of establishments in the shoe industry, we extracted all establishments in the LRD in SIC codes 3143 and 3144, men’s and women’s shoes. For the years ending in 2 or 7 we have a complete census. In intermediate years, we have establishments covered in the Survey of Manufacturers. Table 2 provides some detail on our data set. Column 1 shows the number of establishments in each year of the Census, when the count of establishments is relatively complete.
We are prepared to provide you with quick easy cash loans any time you feel like one. Most of our clients prefer short-term loans, but we are always ready to meet you in the middle based on what you need and how you want to repay your debt. Find out how easy it is here fully.
Column 2 shows the percentage of establishments that were in existence in each of the Census of Manufacturing years that had survived through 1992. Column 3 gives the average compound annual death rate of establishments: it begins at about 8% per year and rises to nearly 12%. Since even in a declining industry some new establishments enter, this does not give the number actually in existence in 1992. Columns 4 and 5 show the number of extant establishments that died between the specified census year and 1992 and the number of new establishments that entered the industry. Added to the number of establishments in the base year, these figures sum to the number existing in 1992.
Our results are consistent with the death rates of shoe factories for more than the past hundred years. For example, the annual death rates for shoe factories during the first half of this century was about 10%, which is similar to our estimates gathered through the LRD for the 1960s through 1990s (Davis, 1940). The overall decline in employment and the number of shoe plants we found in the LRD is therefore largely due to the decline in the number of new entrants. Given the structure of the U.S. shoe industry, new foreign plants seem to be able to produce shoes more efficiently than American firms based largely on a relative labor cost and productivity basis.
Table 3 records measures of productivity—sales and value added in 1992 dollars per hour worked and labor’s share of sales and value added in the shoe industry; average compensation for all employees; and hourly wages for production workers in constant dollars from 1967 to 1992. Productivity per worker has increased by 1.3% per year while valued added per production worker-hour increased by 1.5%. Earnings for all employees have been constant over the period while hourly wages for production workers fell by 15%. The share of labor in sales and value added fell by about a quarter.