In any ranking system, at least 50% of top universities (top 100, for example) are American. What is special about American higher education?
It started in the 1930s, when Nazi Germany started to fire Jewish professors and other people who opposed the regime. Many others left voluntarily as well. Germany had the best universities before this period. The United States ended out getting more of the refugees than any other country (e.g. Einstein) and this started the ascendancy of American universities.
Then World War II happened. Germany ended out with their universities basically obliterated, through the actions of the Nazis and also from the war itself. Japan's universities were also very damaged. The US on the other hand, not only benefited from the refugees but also from the war effort. A lot of money and effort was put into the Manhattan Project and so on, and this infrastructure was used after the war. Other European countries did not really gain as much... some suffered extensive war damage and others like France were at least occupied. Britain's academic institutions were (basically) intact but they were losing ground to the Americans.
After WWII the Cold War happened, and universities were highly supported by the US government during this period, building on the earlier growth. The US had a lot of financial resources and were willing to use them to help universities. Also, American universities in recent decades have been quite welcoming to foreigners, and the increasing prominence of US universities in turn attracts immigrants. So this has become a self-perpetuating phenomenon.
Ranking is based on several features. To understand why many top universities are located in the USA, according to many rankings, you might want to look at each feature. Examples of features:
- faculty salary: this gives to USA-based universities an advantage over some of the countries where academics are less paid. (e.g. France)
- student satisfaction: the American culture tends to be more positive than in some countries.
- research grants: in some countries, academics don't focus on obtaining grants as much as in the USA.
- reputation: the large average size of US universities make them more visible than in many other countries. Also, (taken from O.R. Mapper in the comments), some American universities seem to run professional news outlets on their websites that report about everything noteworthy that happens in the university. In contrast, in some other countries, e.g. Germany, universities' newsfeeds tend to be on a comparably low priority.
- Americans speak English decently well, it helps to get published, give talk in international conferences, create MOOCs, etc.
- endowment: see Why do American universities receive lots of endowments?
Here are some factors that contribute and that aren't really mentioned in the other answers:
The USA is big. The USA has 300 million inhabitants. So of course the number of US universities is higher, and the number of US universities in rankings will be higher. It doesn't really make sense to compare the huge US to a smaller country.
Rankings favor English-speaking universities. For example, one of the criteria in the Shanghai is the number of articles published in Nature and Science; these two journals only publish articles in English. Another criteria is the Science Citation Index, which is also biased towards English. Similar criticism can be made for other rankings.
Rankings favor the Anglo-saxon model. In France for example, most scientists work in a UMR, meaning they are affiliated at the same time at a university and the CNRS (or some other public research institute). This means that for the purposes of counting Nobel prizes for example, only "half" goes to the university, and the other half goes to the CNRS. Some scientists even work directly for a public research institute and not in a university. The CNRS is not a university and doesn't appear in rankings. In the US, a researcher works for the university and that's it; almost all research is done at universities.
Rankings favor very big universities. This one is obvious: if you're just counting Nobel prizes, publications... A bigger university will naturally have a higher ranking. US universities are typically very big: 16 of the 61 (26%) biggest universities by enrollment are in the US.
This is due to many factors, and some can actually be tied to historical reasons. For example, the Cold War - this lead to high R&D spending by the U.S Gov and the creation of many new labs and research areas. Additionally, the U.S. is a country founded by immigrants, and so until this day, you have immigration from all parts of the world (less so from Europe now as it used to be), and as thus, the brain drain in other countries is affecting the U.S. positively. Many top notch researchers in the U.S.A are not/were not even American, but left their countries to pursue opportunities in the USA. Now in cases like Harvard, Yale, other Ivies, these were amongst the first institutions in the USA, and until this day keep this reputation as the best.
Here's an aspect that's mostly missing in the answers that have been posted so far: American universities -- at least the ones which value research output, which are also the ones that tend to show up in published rankings... -- care either very little or not at all about the national origins and citizenship status of the faculty they wish hire and of the graduate students they would like to attract. They try very hard to hire the very best research faculty and the most research-oriented graduate students, irrespective of national origin.
In many of the top departments in the U.S., especially in the sciences and in economics, more than half the faculty and more than two thirds of the graduate students are not native U.S. citizens. (Quite a few, of course, choose to become U.S. citizens if they stay in the U.S.) Thus, rather than thinking in terms of "American universities", it may be more useful to think about "universities in America" in order to understand their prevalence in the rankings. If universities in the U.S. are free to compete for the best talent from around the world, whereas other universities have to give preference to applicants from their own countries (or regions) or are barred outright from hiring non-nationals, it shouldn't be a surprise that the freer system ends up with more talent.
Addendum to address @Emil's comment, about me allegedly claiming that non-US universities suppress cultural diversity. Please note that I have made no such claim. My answer was focused on issues of citizenship and nationality, and not on cultural diversity -- which, I believe, is a much broader concept. Since you asked, though, let me give a specific example of the effects that being able to hire faculty and attract graduate students irrespective of their national origin can have. In Switzerland, the country I live in right now (Switzerland), the immigration law that's still in effect has tended to give great leeway to universities in their hiring decisions. The ETH in Zurich and the EPF in Lausanne, in particular, have made great use of this freedom in recent years to attract stellar faculty and students, in the process either solidifying or strongly improving their international ranking positions. Unsurprisingly, then, a good chunk of the faculty and students, especially the younger faculty and students, of both universities are not Swiss citizens. (Switzerland is, after all, a fairly small country!) In early 2014, however, Swiss voters approved an amendment to the Swiss federal constitution which -- if implemented as some fear/expect it will -- will greatly curtail the freedom of the ETHZ, the EPFL, and all other Swiss universities to go after talent irrespective of their national origin or current citizenship status. Leaders of the ETHZ and EPFL have already warned that if and when the law that implements the constitutional amendment goes into effect, and if the law will be as restrictive as some fear it will be, the rankings of these two universities (and of the other Swiss universities) will suffer seriously and rapidly as many of their top faculty will choose to go elsewhere and as international students look elsewhere when they think about pursuing graduate studies.
@Emil also wrote:
I thought the causation could as well be the other way around: "Because US universities have a high reputation they attract more international students and faculty."
Isn't such an argument nearly circular? For sure, it doesn't explain how or why the rankings of European and Asian universities have finally managed to rise in recent years -- in many cases quite significantly so. What matters greatly, obviously, is having a good reputation. However, being based in the US does not, by itself, establish a good reputation. After all, it's rather well known that there are lots and lots of mediocre and poor universities in the U.S. as well.
I'm looking at http://www.topuniversities.com/university-rankings/world-university-rankings/2015#sorting=rank+region=+country=+faculty=+stars=false+search= and I see that 10 of the top 50 are British, vs 18 USA.
However the population of the USA is about 5 times that of the UK.
So the answer to your question is there's nothing special about the US education system other than the country has a larger population than some other countries with a better education system.
Here's a piece that is touched on in a few other answers but which seems to be unstated: the United States' university system has many of the worlds top schools because it is massively unequal.
The US has more than 4,000 colleges and universities — the "best" universities may be in the US but so are some of the worst. These two facts are not unrelated. The US higher education system is more effective at concentrating resources (top students, money, grants, and the best faculty) in a small number of universities at the top than any other national system.
In much of the developing world and in many small countries, there are simply few resources. In many developed countries, universities are funded by governments that attempt to distribute resources widely and fairly (if not equally) across universities in their systems. If, for example, most students go to universities near their hometown, it seems unfair to send the best faculty and most of the grant funding to the university that trains the folks who happen to live in the capital.
In the US, there are few pretensions. Most of the top US schools are private schools that compete with funds and that have a massively disproportionate share of the money, grants, top students, and top faculty. Because there are few checks on inequality in the US system, this leads to a feedback loop where universities at the top use their resources to capture more — aggravating inequality over time.
American child-rearing and American education are more tolerant of failure. Top American students have learned how to lose gracefully as well as how to win gracefully. They have been taught that risk-taking is worthwhile.
If you look at the distribution of academic scores in American primary, secondary, and undergraduate education, you will find that the American education system has institutions that are consistently very poor, institutions that are consistently mediocre, and institutions that are consistently strong. Depending on their resources and priorities, Americans gravitate toward the institutions that seem to suit them.
A top secondary student who wishes to attend a premier undergraduate school needs to focus on certain things. In Asian countries, the student should cram to pass tests. In America, the student should split their effort into essay-writing, math, science, and athletics. Most premier American undergraduate schools can easily fill their student bodies with students who have high academic marks. Athletic striving helps distinguish applicants as being "well-rounded".
American secondary schools have well developed inter-scholastic leagues in many sports. In each contest between schools, one school wins, and one school loses. Occasional failure is expected; effort and improvement are demanded; coaching is provided; and success is honored. Success is simultaneously measured on a personal improvement level, in contests between individuals, and on a team level.
Courses like MIT's "2.70 contest" have been replicated at several top engineering schools around the world. Instructors of these courses have noted the wide variety of approaches and successfulness of MIT's students, as compared to other countries' students. At MIT, it is normal for one-third of the students' machines to fail to score any points. On the other hand, some MIT students' machines are wildly aggressive and/or wildly successful. In prestigious European and Japanese engineering schools, nearly all students' machines score some points -- in identical contests -- but few students take the unconventional approaches needed to be wildly successful.
In my somewhat extensive experience of Japanese universities vs American universities, the expectations of students and faculty are far higher in American universities. For example, in Japan it is enough for faculty to publish "research" in their university's own journals, few of which are peer reviewed. Plagiarism is a minor problem. In Japan, the student is simply asked to rewrite the paper. In the U.S. it is grounds for immediate expulsion.
In Japan, students spend much of their time doing extra-curricular activities, and preparing for the school festival. Academics generally take a backseat to the social development that is foregone in earlier schooling, when exam preparation is foremost in most students' minds. Furthermore, administrators generally make it difficult for faculty to fail recalcitrant students. Giving a social pass is generally preferred to justifying a failing grade.
I recently did an Academic writing class for third year students in a high level public university. I found that many students were recycling papers they had written in previous years, and once I took measures to obviate that, the quality improved markedly, but it was also clear the papers were written for, and edited by, professors of higher level classes.
The other answers are excellent, but I'd like to add some I haven't seen, in particular aspects where the American system differs:
American universities have to compete for funds. Even "public" universities have to compete for state tuition and certainly for research funding from places like NSF, NIH, DOEnergy, DOEducation... The best American universities are in fact the ones who have some of the longest history of competing for those funds and winning them. Faculty successful at securing external funding often receive higher salaries than those who do not succeed. (Some Europeans tell me their countries are changing over to this model now.)
American laws do not mandate a preference for American employees in higher education. There are some hoops to jump through for foreign labor, but they are largely a pure formality, also unlike some countries. (Again, some Europeans tell me that their laws on this are not mere formalities, which is why so many of their universities have a preponderance of natives in the sciences.)
Perhaps the most important reason: Americans send a much larger proportion of their population to university. Many countries send only a filtered group (e.g., to enroll in the sciences you had to take a certain curriculum at a science-oriented high school). That means a lot of Americans are attending college, which means a lot more money is going through the American higher education system, which gives both more resources to American universities to spend on things, and more incentive to spend it wisely.
Let's define a top university as a university that attracts top talent from all around the world because of its reputation.
Building a reputation as a top university takes generations. It requires a stable operating environment, a decent level of funding, proper management, and more than a bit of luck. If a university has a reputation as a good university, it tends to attract good researchers and good students, which further increases its reputation over time. Top Asian universities are rare, because most Asian countries haven't been that stable long enough.
Destroying the reputation (or the university) takes much less time. Continental European universities are still recovering from WW2.
We're having this discussion in English. The universities in English-speaking countries have an obvious advantage in recruiting top talent, because most researchers and students already speak English. It's much easier to move abroad, if you already speak the language.
Because of these reasons, it's easy to understand why US/UK universities are overrepresented in most university rankings.
Another explanation is the governance. American universities are governed by their alumni. It is alumni who care the most about quality since any reduction in their institution's reputation immediately reflects on them. It also helps immensely with fund raising and money helps a lot with making a good university.
In other countries, it tends to either be the faculty or the government who govern and they have different motivations.
There are two related factors that weigh heavily on most worldwide university ranking systems:
Global ranking systems tend to be heavily biased towards prowess in academic research.
The US conducts a huge, disproportionate amount of the world's academic research.
With regard to the first point, let's look at a few specific ranking systems. The US News & World Report has one such ranking system that is a significant metric for my university. Of the top 50 schools in their World Universities category, 36 are American (72%). Their methodology is exclusively research-based (some categories sound like they're not exclusively research-oriented, but if you read the specific criteria below you'll see that they essentially are). Not only are the rankings exclusively research based, they're almost entirely based on research-related publications. This ranking system totally ignores teaching, teacher to student ratio, grad student to undergrad ratio, faculty salary, percentage of courses taught by PhDs, etc.
The Times Higher Education ranking is another that comes up highly in Google results. Of the top 50 schools in that ranking, 28 are American (56%). This methodology scores based on teaching, research, and "broader impacts" criterion. However, at least 65% of a school's score is attributed to research by my judgement (research category, citations category, international collaboration, and industry income). This system does include metrics such as faculty salary and staff-to-student ratio, but other educational metrics such as number of PhDs awarded is arguably just another proxy for research productivity.
The third Google result is QS Top Universities. Of their top 50 schools, 19 are American (38%). This methodology is starkly different from the previous two. Only 20% of a university's total score is specifically research-related, however the first category (academic reputation) is probably going to be heavily slanted towards research productivity, so 60% of a school's score is directly or indirectly research related. Moreover, this system is different because a full 50% of a school's ranking is dependent on reputation- subjective opinions of the school formed by faculty and experts around the world.
The last organization on the first page of Google results (and where I'll stop) is the Center For World University Rankings. Here, out of the top 50 institutions are 33 American schools (66%). If you look at their methodology, essentially 50% is directly research related (points 3-8), and a further 25% is indirectly research related (quality of education, since this essentially measures the number of students who go on and win research-related prizes). This system also does not explicitly rank any of the educational quality metrics mentioned here, with the remaining 25% being determined by the number of students who go on to be successful in business.
With all that said, you should be convinced of point number one above. Research plays a huge role in global university rankings. Even in the systems that attempt to factor both research and educational outcomes, such as the second and fourth system, the ranking is heavily weighted towards research. Both of those systems place two to three times more importance on research than on education. If you play the numbers game, a school that scores zero on educational stats but about 50% on research stats will be ranked similarly to a school with perfect educational stats but zero research stats. Also note that the one ranking system that places the least emphasis on research productivity, the third, also contains the greatest number of non-American universities.
Now, with respect to the second point above, the US is the world's academic research powerhouse. This is not to say that other countries do not contribute or that they do not produce good work. What is meant is that the US produces the largest volume of academic research as an individual country, and by virtue of that alone receives a tremendous boost in research-oriented ranking systems.
The simplest and best explanation for this in my mind is the amount of funding available for academic research. The top ten countries by expenditure on research in higher education (all funding sources), according to OECD data:
Country Academic R&D Funding in 2013 USA 64.7 Billion USD China 24.1 Billion USD Japan 21.9 Billion USD Germany 18.4 Billion USD France 12.1 Billion USD UK 11.0 Billion USD Canada 10.5 Billion USD Italy 8.0 Billion USD Australia 6.8 Billion USD South Korea 6.3 Billion USD
As you can see, the US expends a significant amount on academic R&D. More than the next three countries combined, and a large fraction of the top ten countries (35%). To generate a graph such as the above you need to view the data set itself and customize:
Measure: PPP Sector of Performance: Higher Education Source of Funds: Total Years: 2013 Table Rows: Country (in the layout customization)
The difference in funding doesn't explain the difference in rankings fully, but it goes a long way. In my opinion it is the primary source of the problems that other countries face when trying to engage in academic research- 50.6% of the top-ten funding occurs in English speaking countries, while only 28.5% of the top-ten funding occurs in Asian countries. Compared to individual countries, which is perhaps what matters most for your question, the US spends about two and a half times as much as China, three times as much as Japan or Germany, six times as much as France, the UK, and Canada, etc.
That funding disparity, combined with the institutional inertia possessed by the US universities (how that happened is a different question altogether), ensures that the US will be the most fertile single nation for academic research for some time to come.
If you consider research to be a input/output situation, then funding dollars is the primary input. However, all of the ranking systems above are more concerned with research output, as expressed through academic publications. A slightly dated Reddit thread aggregates this data from a few different sources. If you take that data at face value, the US publishes about 35% of the world's papers. The next competitor (China) publishes about 16%. The third place country (UK) publishes about 10%.
Bear in mind that most of those ranking systems also emphasize quality of papers over quantity, so this is a crude estimate to answer your question. But, different ranking systems will define quality in different ways (citations vs awards and honors, and so on).
As many others have pointed out, there are a host of other interrelated factors that influence this as well. But, I think the overall situation is best explained by the two points I have presented above.
There is no simple answer. I would think that it is a system dynamic thing. Top universities have a tendency to be sinks for the resources. Best people want to work at the best places and they produce the best results. The resource pool in US is big and there is not much friction; it is also that the system is made easy for talented foreigners so the friction from borders is low.
It is also game-theoretic. Governments have their own objectives like producing educated work force, having equally good universities around the nation, and having relatively cheap specialists to do some research for them. If the universities are left free, they can optimise for the rankings. Trivially you get what you make. For a privately funded university the rankings are one of the best commercial. I remember reading about sensitivity analysis paper where Rockefeller University ranked high in some of those lists, but "low" on another. It is important to look into many rankings while trying to figure out "the true ranking".
US universities is not a singular educational system as in many nation they are; thus they can not be analysed as one. In general, US science has a really bad reputation in my country. It is a synonym for "lobbed click-bait propaganda". It can even be used as a counter-argument in a debate to nullify opponent's information. As such it is often attached the name of the US university as a source, if the university is a respectable one. It is a custom to reference the source somehow, often nations or capitals are appropriate in a public debate because the source needs to say something for the audience. US is shunned more than an unknown researcher.