02044cam a22002657 4500001000700000003000500007005001700012008004100029100002200070245013800092260006600230490004200296500001600338520091100354530006101265538007201326538003601398690005601434690007001490700002401560710004201584830007701626856003801703856003701741w14834NBER20170425071350.0170425s2009 mau||||fs|||| 000 0 eng d1 aCrucini, Mario J.12aA Model of International Citiesh[electronic resource]:bImplications for Real Exchange Rates /cMario J. Crucini, Hakan Yilmazkuday. aCambridge, Mass.bNational Bureau of Economic Researchc2009.1 aNBER working paper seriesvno. w14834 aApril 2009.3 aWe develop a model of cities each inhabited by two agents, one specializing in manufacturing, the other in distribution. The distribution sector represents the physical transformation of all internationally traded goods from the factory gate to the final consumer. Using a panel of micro-prices at the city level, we decompose the long-run variance of LOP deviations into the fraction due to distribution costs, trade costs and a residual. For the median good, trade costs account for 50 percent of the variance, distribution costs account for 10 percent with 40 percent of the variance unexplained. Since the sample of items in the data are heavily skewed toward traded goods, we also decompose the variance based on the median good on an expenditure-weighted basis. Now the tables turn, with distribution costs accounting for 43 percent, trade costs 36 percent and 21 percent of the variance unexplained. aHardcopy version available to institutional subscribers. aSystem requirements: Adobe [Acrobat] Reader required for PDF files. aMode of access: World Wide Web. 7aF0 - General2Journal of Economic Literature class. 7aF15 - Economic Integration2Journal of Economic Literature class.1 aYilmazkuday, Hakan.2 aNational Bureau of Economic Research. 0aWorking Paper Series (National Bureau of Economic Research)vno. w14834.4 uhttp://www.nber.org/papers/w1483441uhttp://dx.doi.org/10.3386/w14834