RESULTS OF PRODUCT-PRICE STUDIES: A Survey of Nine Product-Price Studies 14

Posted by Connie R. Aponte on May 25, 2014 in RESULTS OF PRODUCT-PRICE STUDIES |

Their empirical approach thus involves two steps. First, they decompose observed tfp into a component attributable to outsourcing and a separate component attributable to investment in high-technology equipment. They do this by regressing observed industry tfp growth on measures of outsourcing and computer investment and then using the parameter estimates to do the decomposition. Second, “By using each one of these components in place of total TFP … we will be able to estimate the change in factor prices consistent with that component alone” (p. 13). The two-step approach is similar to that of Leamer, who first decomposes observed product-price changes into a technology component and a globalization component and then regresses each component on factor-cost shares to estimate mandated factor-price changes attributable to each component. Thus, the mandated-wage regression of FH looks like this. payday loans that accept prepaid debit cards

The dependent variable is the estimate of tfp attributable to some measure of outsourcing; because this regressand is estimated FH adjust their standard errors on bi. There are two important comments to make about this dependent variable. First, FH use two different measures of outsourcing and thus have two different sets of outsourcing-related regressands. One “narrow” measure tracks imports of intermediate inputs only from the same two-digit industry as the good being produced. A second “broad” measure tracks all imported intermediate inputs regardless of industry. In their analysis FH actually use the narrow measure and then the difference between the broad and narrow measures. Second, for each of the two measures of outsourcing FH use two different constructions of (“Outsourcing” TFPj*). The first assumes zero pass-through from these outsourcing-induced tfp improvements to product prices. The second allows some pass through both to lower product prices and to industry-specific wage differences from economy-wide average wages. Unlike Leamer who assumes a single pass-through rate common to all industries, FH estimate the pass-through rate by regressing observed product-price changes less industry wage differences against actual tfp, outsourcing, and computer investment. Their point-estimate passthrough rates are 1.725 for the period 1972-1979 and 1.040 for the period 1979-1990. Thus, for each time period FH actually estimate four versions of the (FH) equation: two different measures of outsourcing and two different assumptions about whether or not outsourcing feeds through to product-price changes.

Tags: , ,

Copyright © 2013 - 2019 Blog About Business And Finance. All rights reserved

Home | Site Map | Contacts