An Examination of Productivity and Market Integration Under Volatile Market Conditions.

다국어 초록 (Multilingual Abstract)

The U.S. wholesale market for electric power demonstrates the role of mobility of goods and their factors in the integration of a market. Utilizing day-ahead wholesale prices at over 50 trading hubs across the U.S., we find the effective distance of being separated by a transmission interconnection to be small using traditional border regressions. Using Principal Component Analysis and a large Bayesian vector auto regression we document that much of the integration occurring across transmission interconnections is achieved from the mobility of natural gas, a primary factor of production. We conclude the regional wholesale markets are integrated despite physical barriers to trade in output markets because of the mobility of factors of production.
I suggest an alternative approach to investigate market integration. Specifically, I estimate the price transmission ratio across different horizons to evaluate the level of substitutability between two goods. The framework is applied to the U.S. wholesale electricity setting for two locations in the Western U.S. Over the period, 2001-2012, electricity supplied among the different regions of the Western U.S. are found to have high price transmission ratios at frequencies of a week or longer, while at shorter frequencies the price transmission ratio across locations are smaller.
Measures of productivity reveal large differences across producers even within narrowly defined industries. This paper proposes an explanation for the productivity differences. When adjusting production is costly, increases in demand volatility increase the unit costs of otherwise identical producers. Furthermore, temporal aggregation conceals mean preserving differences in demand volatility within the period of aggregation. Consequently, combining three factors, (i) differences in demand volatility (ii) adjustment costs and (iii) temporal aggregation, leads to differences in measured productivity. I document this effect empirically by comparing an industry with high adjustment costs (hotels) to an industry with small adjustment costs (airlines). Differences in annual demand volatility alone explain 35 percent of the measured productivity variation of hotels at the metro area-segment-year level. In contrast, differences in annual demand volatility have no effect on the measured productivity of airlines at the destination-airline-year level.