English Channel



去年时看过一篇文章，题目是Taxation and International Migration of Superstars: Evidence from the European Football Market，在网上就能找到pdf，应该免费



写的是个人所得税和欧洲足球明星跨国转会的事儿。提到了博斯曼法则、贝克汉姆法则等一些个人比较感兴趣的问题。正巧当时一课程让我们写份英文review，就有了这个，贴上来混论坛贡献了




Introduction
This paper evaluates Kleven et. al’s (2010) recent study which uses the European soccer market as a sample analyses the effect of taxation on international migration of high-income (highly-skilled) labor.




Although theoretical framework has been developed a few decades ago (Mirrlees, 1982), due to limited data Kleven et. al (2010) is still the earliest study on the effect of taxation on the mobility of people, especially high-skilled workers. The authors aim to establish a system which other studies on this topic can follow.




Section II of this paper will give a brief introduction of the findings by Kleven et. al (2010). Section III evaluates the importance of the findings. Section IV discusses the data used in this study. Being able to find appropriate data is a very important reason why this study is able to achieve the authors’ goals. And the final section concludes the evaluation.




Section II
Kleven et. al (2010) are able to identify strong and negative correlations between (a) the tax rate on foreign players and the fraction of foreigners in the national league, (b) the tax rate on domestic players and the fraction of native players playing in the home league, and (c) the tax rate on local players and the performance of first-league teams in the country. Case studies in different countries on the effect of special tax scheme led to similar findings.




In the second part of Kleven et. al (2010), a theoretical model of taxation and migration is developed. Depending on whether the demand of the soccer sector is flexible or rigid, the effect of taxation may be broken down to “ability sorting effect” (taxation has different impacts on workers with different levels of skill) and “displacement effect” (favorable tax rate on foreign workers hurts domestic workers).




The third part of Kleven et. al (2010) provides a structural estimation of the flexible- and rigid- demand models. Evidences of both ability sorting effect and displacement effect are found.




Section III
It is true that for the most part the migration of labor is not solely due to income tax differentials, however, clear identification of tax-induced labor migration can provide guide to policy makers. The sample used in this study, the European Soccer teams, can be generalized and the findings can hence provide directions on the economic development.




The soccer leagues in Europe are competing against each other for market share which ultimately leads to revenues to the League and players. This is mainly determined by the performance of the teams which origins from the quality of players. There is an interesting quote of Arsene Wenger, manager of Arsenal FC “with the new taxation system, (…), the domination of the Premier League (England’s top soccer league) will go, that’s for sure” (The Sunday Times, April 25, 2009).




If we view the performance of soccer teams as productivity, this is no different than saying that an economy with more efficient (skilled) workers can be more productive. According to the findings of Kleven et. al (2010), for two locations that are similar in all aspects, the one with lower income tax burden is more attractive to (skilled) workers. However, while lowering the tax rate to such workers can make the economy more competitive, it destroys the progressivity of the tax system.




Section IV
Considering the cost, it is almost impossible in nature to experiment on the effect of taxation on migration purely for academic purpose. Therefore one has to look for the data that matches the requirement. To examine the effect of personal income tax on labor migration, an ideal experiment requires the following qualities:




The first requirement is the worker must live in the same area which his or her income is taxed. This may be taken for granted in a cross-country case, but in a smaller geographic area such as at county level, it is possible that one works at a county with higher income tax while he or she lives in a county with lower rate. With such option possible it is hard to observe the correlation between income tax and labor movement.

Secondly, income tax differential should be the only factor that causes labor migration. For instance, a developed country tends to have better public service and infrastructure than a developing country and as a result, the labor migration between these two countries is likely to be caused by factors other than taxation. USA and Mexico are a good example, although the immigration of Mexican labor is not likely to be driven by income tax differentials.




In addition, the income tax differential should be large enough relative to the cost of migration to justify relocation. This argument is quite obvious. Although there may be income tax differentials in two areas, as long as the gain from relocation does not exceed the cost of job searching, moving, etc., labor migration will not be observed. Cross-country moving cost may be higher, since it includes converting currency, learning a new language and getting used to a new climate, etc. By this argument, in an ideal case there should be a very large income tax differential between the two areas, or/and a very low cost of moving.




There should be enough variation in taxation to allow observation. This is true for both cross-sectional and time-series analysis. For a cross-sectional analysis, a control group should be available. This control group, be it a country or a state, ideally should be comparable to the experiment group in all aspects except for the variation in income tax rate. For a time-series analysis, one should allow enough time after a variation in income tax for its impact to be observable.
Last but not least, there must be sufficient data available on the sample. One important reason why this area of research is almost unavailable is there lacks micro data such as citizenship that can help observe a causal relationship.



The international migration of top soccer players in European soccer market meets many, if not all of the requirements above.




Unlike running online stores, soccer players must train everyday and play games at home stadium frequently. This requires them live close to their team. Therefore soccer players usually pay income tax of the country in which they live.




At the same time, there is extensive data available. Like any world-class sports league, there is large amount of data on each player in the European soccer market. Such data includes income, statistics of the player, career path and other details. It is also available for almost all European soccer leagues over a long period of time.




Thirdly, the relative potential gain of international migration for European soccer stars is large. On one hand, soccer stars qualify for top level income group and typically have large proportion of income in the top bracket of income tax. Moving to a country with lower tax rate could mean huge tax burden cut. On the other hand, their cost of movement is low. Soccer as a sport requires relatively low location specific human capital. Playing in UK is almost exactly the same to playing in Germany. Besides, the cost of moving from a European country to another may be the lowest at international level.




There are enough sources of variation in taxation in the European football market. Over time, there are regulation changes such as the Bosman ruling in 1995. Across countries, there are different tax schemes offered to soccer players that allow them to pay less than what they would be paying in another country. At the same time, it is not hard to find good control group for a European country.




Conclusion
Kleven et. al’s (2010) study is a smart and successful attempt on finding the effect of taxation and international migration of high-income labor. It is smart because the authors have picked a very unique and appropriate dataset that allows them to conduct the study and at the same time generalize their findings. It is successful because the authors are able to identify some effects that may provide important guide to policy makers. As an early study in the area of taxation and international labor mobility, their study has concentrated on the high-skilled labor. Further studies may expand to low-skilled labor and domestic mobility. The findings may be very different.