"Most economic activity occurs in cities. This creates a tension between local increasing returns, implied by the existence of cities, and aggregate constant returns, implied by balanced growth. To address this tension, we develop a theory of economic growth in an urban environment. We show that the urban structure is the margin that eliminates local increasing returns to yield constant returns to scale in the aggregate, which is sufficient to deliver balanced growth. In a multi-sector economy with specific factors and productivity shocks, the same mechanism leads to a city size distribution that is well described by a power distribution with coefficient one: Zipf's Law. Under certain assumptions our theory produces Zipf's Law exactly. More generally, it produces the systematic deviations from Zipf's Law observed in the data, including the under-representation of small cities and the absence of very large ones. In general, the model identifies the standard deviation of industry productivity shocks as the key parameter determining dispersion in the city size distribution. We present evidence that the relationship between the dispersion of city sizes and the variance of productivity shocks is consistent with the data"--National Bureau of Economic Research web site.