The third problem in updating a piece rate system is that there is a significant information asymmetry about the “potential level” of production. Workers learn on the job about how best to produce, and will always be ahead of the firm in knowing the potential level of production. Like tax lawyers, they find “loopholes” in the piece system faster than (the IRS) firms can correct them. Hence, when the firm updates its piece system, workers will have an incentive to “hide” their knowledge by taking leisure on the job rather than by increasing output.
The firm cannot observe leisure on the job whereas it gets full information about potential production by observing actual output. We can model by making the cost term in the utility analysis depend on cumulated units produced as well as on current units, per leaming-by-doing models. The “loophole effect” exacerbates the ratchet effect: the more rapid the firm updates its incentive system, the greater will be the tendency for workers to apply their cumulated knowledge to leisure on the job than to production. in detail
In sum, the decision to update a piece system must weigh the direct cost of updating the system, the reduced incentive effect, and increased incentive for workers to take leisure on the job from a reformed system against the lower labor cost that a reformed piece rate system produces. Update frequently and you lose some of the incentive effect of a piece rate system. Update infrequently and the firm loses potential quasi-rents. Weighing these offsetting factors, the firm picks the most profitable number of updates or speed of updating. If the shoe industry was correct in shifting from piece rate to time rate, the loss of incentives from rapid updating of the piece rate system and the cost of updating must have been sufficiently high to make revising the piece rate system less profitable than allowing the system to become outmoded.
Survival in a Collapsing Market
In this section, we use data from the Census Bureau’s Longitudinal Research Database (LRD) to examine the characteristics of surviving and dying establishments in the shoe industry over time. The LRD was created by the Census to provide an establishment-level data from the Bureau’s Survey of Manufacturers. It contains information on shipments, value added, employment, wage bills, materials, and inventories for all manufacturing establishments with more than 250 employees and a sub-sample of establishments with fewer employees in all years except the years when the Census conducts its Census of Manufacturers every five years.
We use Salter’s (1966) classic model of the utilization of different production techniques in a competitive market to guide the analysis. This model highlights two aspects about production in a sector: the range of variation in practices among establishments due to the vintage of capital (and other factors); and the role of labor costs in determining that range of variation.