Integration of the US goods market, 2001–2015

A spatially dispersed market for a tradable good is deemed integrated if there are no barriers to trade between its spatial segments (except for geographical barriers, namely distances between the segments). However, perfectly integrated markets are not a common case; real markets deviate to some extent from this ideal state. Therefore, estimating a degree of integration is more helpful then an answer of the type “all or nothing” (whether the market is integrated or not integrated). In an integrated market, price for a good is determined in the national market as a whole, not depending on demand in its spatial segments. Hence, a dependence of local price on local demand (controlling for transportation costs) indicates a deviation from perfect integration, and its “strength” can measure the degree of market integration. Based on this idea, we estimate the annual integration degrees of the US market for an aggregated good (grocery basket) over 15 years, 2001–2015. The spatial segments are cities; our sample covers 66 cities from 39 states of the US. The results suggest that the US market is not perfectly integrated; however, the integration degree of the US market is fairly stable over time. We also compare results for the US with results of a similar study for Russia. With a reservation that the empirical material is not fully comparable, we can conclude that the US market is integrated more strongly than the Russian market and that the integration degree in Russia is more volatile.