S.1. INTRODUCTION THIS SUPPLEMENT CONTAINS DETAILED DERIVATIONS and supplementary material for the paper. A separate Technical Data Appendix, located together with the replication data and code, collects together additional empirical results and robustness tests.Section S.2 of this supplement presents a more detailed analysis of the theoretical model. We report the technical derivations of the expressions reported in the paper. We also establish a number of results about the properties of the general equilibrium with exogenous location characteristics and endogenous agglomeration forces. Section S.3 calibrates the model for known parameter values and shows that there is a one-to-one mapping from these known parameters and the observed data to unobserved location characteristics. Therefore, these unobserved location characteristics correspond to structural residuals that are functions of the parameters and the observed data.Section S.4 turns to the structural estimation of the model, where both the parameters and unobserved location characteristics are unknown and to be estimated. We derive the moment conditions used in the estimation and review the Generalized Method of Moments (GMM) estimator as applied to our setting. We discuss the computational algorithms used to estimate the model and report the results of a grid search over the parameter space that we use to characterize the properties of the GMM objective function. Section S.5 uses the model to undertake counterfactuals for the effects of division and reunification. Section S.6 contains further information about the data sources and definitions. S.2. THEORETICAL MODELIn this section, we develop in further detail the theoretical model outlined in the paper. We present the complete technical derivations for all the expressions and results reported in the paper. In the interests of clarity and to ensure that this section of the supplement is self-contained, we reproduce some material from the paper, but also include the intermediate steps for the derivation of expressions.We consider a city embedded within a wider economy. The city consists of a set of discrete locations or blocks, which are indexed by i = 1 S. The city is

A central prediction of a large class of theoretical models is that industry location is not necessarily uniquely determined by fundamentals. In these models, historical accident or expectations determine which of several steady-state locations is selected. Despite the theoretical prominence of these ideas, there is surprisingly little systematic evidence on their empirical relevance. This paper exploits the combination of the division of Germany after the Second World War and the reunification of East and West Germany as an exogenous shock to industry location. We focus on a particular economic activity and establish that division caused a shift of Germany's air hub from Berlin to Frankfurt and there is no evidence of a return of the air hub to Berlin after reunification. We develop a body of evidence that the relocation of the air hub is not driven by a change in economic fundamentals but is instead a shift between multiple steady-states.

Why do borders still matter for economic activity? The reunification of Germany in 1990 provides a unique natural experiment for examining the effect of political borders on trade both in the cross-section and over time. With the fall of the Berlin Wall and the rapid formation of a political and economic union, strong and strictly enforced administrative barriers to trade between East Germany and West Germany were eliminated completely within a very short period of time. The evolution of intraGerman trade flows after reunification then provides new insights for both the globalization and border effects literatures. Our estimation results show a remarkable persistence in intra-German trade patterns along the former East-West border; political integration is not rapidly followed by economic integration. Instead, we estimate that it takes at least one generation (between 33 and 40 years or more) to remove the impact of political borders on trade. This finding strongly suggests that border effects are neither statistical artefacts nor mainly driven by administrative or "red tape" barriers to trade, but arise from economic fundamentals.

This paper develops a quantitative model of city structure to separate agglomeration forces, dispersion forces and fundamentals as determinants of location choices. The model remains tractable and amenable to empirical analysis because of stochastic shocks to worker productivity, which yield a gravity equation for commuting flows. To empirically disentangle alternative determinants of location choices, we use Berlin's division and reunification as a source of exogenous variation in the surrounding concentration of economic activity. Using disaggregated data on land prices, workplace employment and residence employment for thousands of city blocks for 1936, 1986 and 2006, we find that the model can account both qualitatively and quantitatively for the observed changes in city structure.

We examine the geography of cotton textiles in Britain in 1838 to test claims about why the industry came to be so heavily concentrated in Lancashire. Our analysis considers both first and second nature aspects of geography including the availability of water power, humidity, coal prices, market access and sunk costs. We show that some of these characteristics have substantial explanatory power. Moreover, we exploit the change from water to steam power to show that the persistent effect of first nature characteristics on industry location can be explained by a combination of sunk costs and agglomeration effects.