This assumption can't be considered universal: One thing that can happen within a nation – and not across nations – is that the workers in that region can migrate to other regions and therefore partake in the benefits of trade that accrue elsewhere. That is how, for example, Southern states in the United States adjusted to the industrial dominance of the North. This isn't the case in countries such as Indonesia or China (and many other places) where domestic migration and human movement is controlled. Similarly, this: Another thing that happens is that there is an overarching state that will engage in transfer payments and other policies that aid the lagging region. This makes the assumption that the 'overarching state' will inevitably distribute wealth evenly, whereas the disparities between urban and rural revenue distribution can be stark, thinking specifically of Thailand (where it is used to prop up a middle-class voter base for Thai royalists) and, historically, Indonesia, where most of the 14,000 islands that aren't Java (including Sumatra, geographically the largest island) have for the most part been ignored and suffered economically as a result. I think this all boils down to the assumptions in this statement: The boundaries of a nation are defined by shared sense of collective purpose, as embodied, in part, in that nation’s common laws and regulations and in its instruments of solidarity. This is true of a strong state, but the fact is the majority of the world's states are in fact weak. Governments are not accountable, they are absent rule of law, democracy is ineffective. They are not defined by a 'collective purpose'; they are merely defined by an arbitrary political boundary. In other words, Professor Rodrik views a nation state as politically homogeneous; whereas in many cases they are rife with division that can be equal to or worse than international political divisions.