Zipf's law and urban scaling are two fundamental paradigms researched in urban science. They have mostly been investigated independently and are perceived as disassociated matters. Here we present a large scale investigation about the connection between these two laws using population and GDP data from 96 countries. We empirically demonstrate that both laws are tied to each other and derive an expression relating the exponents. This expression captures the main tendency of the empirical relation between both exponents, and simulations yield very similar results to the real data after accounting for fluctuations. Our research puts forward the idea that urban scaling of GDP does not solely emerge from intra-city processes. Instead, we show that population distribution affects the scaling of urban GDP such that countries with less small cities and more metropolises (small Zipf exponent) have less pronounced increasing returns to scale of urban GDP than countries with a more uneven number of small and large cities (large Zipf exponent).
