‘Voting with feet’ (part 1)

Tiebout model

Charles Tiebout first proposed the idea of "voting with feet" in his 1956 seminal paper “A Pure Theory of Local Expenditures” published in The Journal of Political Economy, as a non-political, market way to solve the free rider problem in local governance. 

The idea is simple. The model basically implies that people will choose where they want to live based on local public goods and services offered in different municipalities. People differ in their individual preferences towards different services and their willingness to pay for these services. They will choose where to live accordingly. A local provision of public goods will hence depend mostly on the taste of its residents – a logical proposition. 

The idea was even embraced by two of my favourite Nobel Prize laureates Friedman and Hayek: 

Friedman: "...The second broad principle is that government power must be dispersed. If government is to exercise power, better in the county than in the state, better in the state than in Washington. If I do not like what my local community does, be it in sewage disposal, or zoning, or schools, I can move to another local community, and though few may take this step, the mere possibility acts as a check. If I do not like what Washington imposes, I have few alternatives in this world of jealous nations."  

Hayek: "I'm inclined to give local authorities power which I would deny to the central government, because people can vote with their feet against what the local governments can do." 

However, the model Tiebout proposes is not without flaws. As with any economics model two assumptions exist which pretty much defy reality: complete information and no transaction costs. Economists do this for simplicity; it helps them avoid unnecessary analytical complexity and gives them an opportunity to reach a unique solution to the problem they observe, and presumably offer policy advice based on it. In the real world, however, even if the conclusions are perfectly sound, many assumptions aren’t. This doesn’t make models uninteresting or inapplicable nor does it prevent policy-makers in trying to overcome the problems in implementing the theoretical findings.

As for Tiebout model's assumptions, people often don’t posses complete information on which part of the country could be the most beneficial to them, even though thanks to the wonders of the internet, this information is becoming more and more accessible and less and less costly. However, transaction costs are likely to be significant in any case. Other assumptions include the rationality of communities, non-spillover of public good benefits across communities, an optimal city size and no issues with commuting. Even if we were somehow to accept these assumptions it would still be very farfetched. 

Nevertheless, the model appears to work. At least in the US. Here is an interesting paper that empirically assesses the Tiebout model and finds it applicable to the US local government (though it does so only on the basis of environmental quality). There is a vast majority of other research available that tend to support the voting with feet story. The graphs below emphasize this to a certain extent. While New York has mainly net positive migration (black lines are inward, red lines are outward migration), L.A. has the reverse trend - people are moving across California and all around the South-East. Forbes is running a detailed map of inner-US migration where one can find out very interesting information about these patters for each specific county. One of these interesting patterns is the migration to Florida, particularly from New York (I assume old people make up most of this data). As for the patterns after the crisis (shown in the two graphs on the right), their frequency seems to have dropped, which was expected as the transaction costs of moving suddenly increased (actual costs could have even decreased due to lower housing prices, but relative to total wealth which decreased, the costs of moving increased). 

Source: Forbes Map: Where Americans are moving; and
WSJ: "Recession alters migration patterns in US"

But these patterns don't tell us what makes people move - is it really the disatisfaction with local goods and services offered, or is it usually due to accepting better jobs, or changing careers which seems to be much easier to do in the US than Europe, for example. I guess it’s really the choice of whether the utility gain from the new location will be bigger than the transaction costs of moving locations. 

So basically as long as U(Δm) > f(c; m)

Where f(c; m) is an increasing linear function determined by the location of the move (the further away it is, the higher the costs – it could also be a convex function). We can also add a stochastic shock in the equation where an event like the crisis or the housing bubble burst increase the relative transaction costs even if location is unaffected. But this is where things get a bit complicated, which emphasizes my earlier point. 

What can migration patterns explain?

However, even if Americans on average care more on the utility they receive from local communities, in Europe people are much less inclined to move even within the EU (at least across borders). People are much less mobile in Europe, mainly due to two things: language barriers (within the EU) and immigration constraints (outside it). 

Reflecting upon this, can competition in local, national and international levels explain the trails of world migration, both now and historically? When people aren’t happy with their lifestyle in the local environment they will aim to switch it; whether due to ambition, a chance to earn more money, or simply to survive. 

This is the point of foreign direct investments if you look at the World globally and countries as local units. FDIs will flow to where the returns are highest and where the tax rates are the lowest. The perfect combination is security of the investment, a relatively high return compared to the alternatives and a low tax rate. This is why emerging markets succeed – their reforms offer more security to foreign investors and a high return due to an unexplored market. The best way to attract foreign investors is to incentivise them with low enough tax rates. Similar things attract people as well. 

This mobility can also be used to explain the growth trails of some nations. For example, a large amount of mobility between sectors can lead to a rapid industrialization and development. Human capital transferred to places where it can make the most of its value will very often delivers that value. This is why many students go to study in the US or the UK, hoping they could stay and ensure a better life and fulfill their ambitions. 

Perhaps this mobility can be one parameter in explaining the differences between Europe and the United States - in terms of ‘mentality’, the welfare state model, the realm of social justice, and value of work. I’ve written before on how religiosity can explain different levels of social spending. The willingness to inner-migrate and 'vote with feet' can be another good explanation of these differences. 

If this is indeed true, perhaps this finding can be used to solve the problem of health insurance provision in the US? Part 2 will deal with this issue.