Foreign athletes in domestic leagues

The question

When a league admits players from abroad, two things can happen to the players already there. They can improve, because they now train and compete against better opponents and are exposed to techniques and habits from other systems — the spillover hypothesis. Or they can lose playing time and development to the newcomers — the crowding-out hypothesis. Both mechanisms are real; the empirical question is which dominates, and for whom.

Sport is unusually good for answering it, because national teams provide a clean measurement. In most sports a national team may use only players eligible for that country, so if the domestic players improve, the national team's results should improve too. Economists have used that fact to treat leagues as firms and players as co-workers, and to ask whether importing foreign talent raises the performance of the locals.

The basketball study

The most-cited study is by Jaime Alvarez, David Forrest, Ismael Sanz and Juan de Dios Tena, "Impact of importing foreign talent on performance levels of local co-workers", published in Labour Economics in 2011. It is about basketball, not football — a point that matters because FIBA Europe registers every foreign player in a member's top two divisions, so the count is exact.^[1] [2]

The panel covers 47 European countries observed "over more than twenty years": foreign-player licence counts from 1986 to 2007, and tournament results through the 2008 Olympics. Over that period the average number of foreign players per team in Europe's top two divisions rose from 2.11 in 1986 to 2.46 in 1995 and then to 5.62 by 2007 — the jump comes after the 1995 Bosman ruling.^[2]

The authors model two things: the probability that a country qualifies for the Olympic tournament, the World Championship or EuroBasket, and its final ranking once there, controlling for previous senior and youth results, population, GDP per head, host status and the number of NBA draftees. Their conclusion, in the journal's abstract:^[1]

"We demonstrate that, consistent with the spillover hypothesis, an increase in the number of foreigners in a domestic league tends to generate a subsequent improvement in the performance of the national team (which has to be comprised only of local players)."

Three qualifications belong next to that sentence, all from the authors themselves. The effect on qualification holds for the world-level events but not for EuroBasket; the effect on final ranking holds for both. There must be some upper limit — a league of entirely foreign players would leave only amateurs for the national team — but no turning point appeared within the range of the data. And crowding out is conceded as a real mechanism that is "unambiguously dominated" in the results, not absent. The authors' own policy sentence is blunt: restricting foreign players in English football and cricket, then being proposed, "would not be evidence based and may well prove counterproductive."^[2]

What the other studies found

The basketball paper sits inside a literature that runs in more than one direction. The table gives the main peer-reviewed results in the authors' own terms.

Milanovic's 2005 model of football globalisation predicted the pattern most of these results show: free circulation of players raises the overall quality of the game and the inequality between clubs, while reducing inequality between national teams, because poorer countries recover some of their exported talent.^[13]

How international the leagues have become

The counts are published by the leagues and by the CIES Football Observatory, which defines an expatriate as a player who grew up in a different association from his employer's and moved for football reasons — so a naturalised player can be an expatriate and a locally trained foreign passport-holder is not.^[14]

• Football. 15,310 expatriates in 135 leagues on 1 May 2024, a record, up 19.2% from 12,841 in 2020 and 24.3% of all squad members (29.3% in the European leagues surveyed). Across 2020–25 the largest exporters were Brazil (3,020 distinct players abroad), France (2,293) and Argentina (2,171). In Europe's 31 top divisions expatriates were 42.3% of players in October 2022, highest in Cyprus (74.0%) and, among the big five, Serie A (61.7%).^{[14] [15] [16]}
• Basketball. The NBA opened 2025–26 with a record 135 international players from 43 countries, 71 of them European; every team had at least one, and international players had won the previous seven MVP awards.^[17]
• Baseball. 249 of the 948 players on MLB Opening Day rosters in 2026 — 26.3% — were born outside the United States, led by the Dominican Republic (93) and Venezuela (60). MLB has published the series every year since 1995.^[18]
• Ice hockey. Of 726 NHL players on 2025–26 opening rosters, 41.9% were Canadian, 26.9% American and 32% born outside North America, from 20 birth countries.^[19]
• College sport. About 22,600 international student-athletes compete in the NCAA — 4% of all athletes, 7% in Division I. Among Division I entrants in 2019 the shares were highest in tennis (61% of men, 60% of women) and ice hockey, and among men in soccer.^[20]

Training abroad as policy

The mirror image of importing players is sending your own athletes into foreign training environments, and the Olympic Movement funds exactly that. Olympic Solidarity's scholarship guidelines offer a "training centre option" under which a scholarship-holder is placed abroad in a high-level centre through agreements with partner National Olympic Committees, international federations and individual centres, as an alternative to training at home.^[21]

For Paris 2024, 1,560 athletes from 195 NOCs held scholarships; 604 of them competed and won 75 individual medals, among them Botswana's first Olympic gold (Letsile Tebogo) and Pakistan's first individual gold (Arshad Nadeem). The IOC raised the programme's budget to USD 650 million for 2025–2028, describing the result as "levelling the playing field".^[21]

The rules written in response

For EU nationals inside the European Union the answer is settled law. In Bosman (Case C-415/93, 15 December 1995) the Court of Justice held that the Treaty "precludes the application of rules laid down by sporting associations under which … football clubs may field only a limited number of professional players who are nationals of other Member States." The Court rejected each justification offered: that a club's identity is tied to its country (no more, it said, than to "its locality, town, region"); that quotas create "a sufficient pool of national players" for the national team (national-team players "need not necessarily be registered to play for clubs in that country"); and competitive balance (quotas do not stop rich clubs buying the best national players). Nationality rules survive only for matches "of sporting interest only, such as … matches between national teams."^[22]

What survives is the training-based rule. UEFA's home-grown rule, phased in from 2006–07 to reach eight home-grown players in a 25-man squad by 2008–09, defines home-grown by three years' training between the ages of 15 and 21 at a club in the same association, "regardless of their nationality"; UEFA's stated purpose was to counter the post-Bosman "hoarding" of players by rich clubs and restore incentives to train youth. The European Commission called it proportionate in May 2008 while calling FIFA's proposed "6+5" rule — six players eligible for the club's national team in the starting eleven — "directly discriminatory and therefore incompatible with EU law". FIFA's Congress voted for 6+5's objectives in Sydney that month; the plan was shelved in 2010 and was never implemented anywhere.^{[23] [24] [25]} The Premier League's version limits a 25-man squad to 17 players who do not meet its home-grown definition, "irrespective of nationality or age".^[26]

Outside the EU nationality quotas are ordinary. Japan's NPB may register at most four foreign players on the active roster, no more than three of them pitchers or three position players, with no limit under contract — the configuration in place since 2002. South Korea's KBO allows three foreigners per club, at most two pitchers, with temporary injury replacements permitted since 2024. Turkey's Süper Lig cut registered foreigners from 14 to 12 from 2024–25, falling to 11 in 2027–28, and abolished its matchday limit. China's CBA introduced a "four quarters, seven appearances" cap on foreign players in 2024–25.^{[27] [28] [29] [30]}

The reading

Taken together the studies agree on more than the headlines suggest. League quality rises when a league opens. Countries that export players benefit, and the benefit is largest for countries without strong clubs of their own. Countries that import players benefit in the one sport where the measurement is cleanest, basketball, and Greek club results point the same way. Whether the balance between national teams narrows, as Milanovic's model predicts, is not settled: the football studies find small or no change.

The dissenting results are about individuals, not teams: native players in Spain lost playing time and, more durably, lost performance; English and Italian players' minutes fell even after the home-grown rule. Those are real losses to real careers, and the basketball authors do not deny them — they find them outweighed at the level of the national team. The mechanism the studies propose is the spillover hypothesis itself: training beside and competing against better and different players raises what the locals can do, and the national team is where that shows.

Two things the evidence does not establish. It does not say how far the process can go — the upper limit is assumed, not observed. And it does not say that foreign players make a club better within a league where every club may buy them; there, money decides. The policy response has followed the law rather than the evidence: inside the EU, where quotas on other member states' nationals are unlawful, rules now reward where a player was trained; outside it, quotas remain the default.