The Decline of Harvard

 

Published in Substack 08 April 2025

Quite a few stories have come out of the Ivy League about how standards are collapsing. I used to think that this was just the perennial lament of teachers everywhere that today’s students are inferior to those of my day. But the stories are coming faster these days, and they seem to be consonant with declining cognitive skills throughout the West, a general disengagement by students, increasing rates of plagiarism, rejection of science and liberal values, and the ardent embrace of extremist ideologies.

Perhaps the most striking story was when Harvard introduced remedial math courses for some of its students. This resulted from the suspension of requiring the submission of SAT and ACT scores following the COVID-19 outbreak.

I suspect that the problem may go deeper than that, and remedial courses at Harvard and other elite schools may become permanent, although probably presented as enrichment programs or something like that.

But this is all anecdotal. Evidence from global rankings can provide more systematic data, which shows that Harvard is steadily declining relative to international universities and even to its peers in the USA.

Here is a prediction. This year, next year, or maybe the year after, Harvard will cede its position as the top university in the world in the publications metric in the Shanghai Rankings to Zhejiang University in Hangzhou.

The Shanghai Rankings, officially known as the Academic Ranking of World Universities (ARWU), have six indicators: Nobel prizes and Fields Medals for alumni, and faculty, papers in Nature and Science, Highly Cited Researchers, publications in the Science Citation Index Extended and the Social Science Citation Index, and Productivity per Capita, which is the sum of those five scores divided by the number of faculty.

When they began, the Shanghai Rankings placed Harvard in first place overall and for all the indicators except for productivity, where Caltech has always held the lead. However, in 2022, Harvard lost its lead to Princeton for faculty winning the Nobel and Fields awards. The coming loss of supremacy for publications will mean that Harvard will lead in just half of the six indicators.

This is only one sign of Harvard’s decline. Looking at some other rankings, we find a similar story. Back in 2010, when QS started producing independent rankings, Harvard was replaced by Cambridge, which in turn was superseded by MIT, which has held first place ever since. In the THE rankings, Caltech deposed Harvard in 2013 and was overtaken by Oxford in 2017.

I have no great faith in THE or QS, but this is suggestive. Then, we have the more rigorous research-based rankings. In the 2024 Leiden Ranking, published by the Centre for Science and Technology Studies at Leiden University, Zhejiang University took over first place from Harvard for publications, although not – not yet anyway – for publications in the top 10% or 1% of journals. In the SCImago Institutions Ranking, published in Spain, Harvard is now fourth overall, although still the leading university.

But when we look at computer science and engineering rankings, it is clear that Harvard has fallen dramatically in areas crucial to economic growth and scientific research over the last few decades.

Shanghai has Harvard in 10th place for computer science, the National Taiwan University Rankings put it in 11^th behind Wisconsin, Georgia Institute of Technology, Texas at Austin, and Carnegie Mellon, SCImago Institutions Rankings 61^st, and University Ranking by Academic Performance, published by the Middle East Technical University in Ankara, 35^th.

For engineering, the prospect is just as grim. The Taiwan rankings have Harvard 31^st, Scimago 42^nd, and URAP 71^st.

Fine, you might say, but the bottom line is jobs and salaries. Let’s look at the latest Financial Times MBA rankings, where Harvard has plunged to 13^th place. A major reason for that was that nearly a quarter of the class of 2024 could not find jobs after graduating. According to Poets & Quants, Harvard’s “placement numbers are below every M7 peer, including Stanford, Wharton, Columbia, Kellogg, and Booth, with only one exception: MIT Sloan which is equal to HBS.”

It seems that Harvard’s problems are entrenched and pervasive. They may have been exacerbated by the pandemic, but their roots go back and go deeper. So what is the cause of this decline? I doubt that the usual villain, underfunding by vicious governments or offended donors, has anything to do with it. However, the announced Trumpian cuts may have an effect in the future.

A plausible hypothesis is that Harvard has drifted away from meritocracy in student admissions and assessment and, more significantly, faculty appointments and promotion.

Perhaps the concept of Harvard’s meritocracy has always been overblown. A few years ago, I was researching early American history and came across a reference to a prominent Massachusetts landowner who had graduated first in his class at Harvard. I was baffled because I thought I should have heard of somebody that brilliant. But it turned out that Harvard before the Revolution ranked students according to their perceived social status, a practice that ended with Independence, after which they were ranked alphabetically. The idea of sorting students academically seems to have become widespread only in the twentieth century.

Even after Harvard supposedly embraced meritocracy by introducing the SAT, the GRE, and other tests and linking tenure to publications and citations, it still included large numbers of legacies, athletes, persons of interest to the dean, and affirmative action.

It seems that Harvard is returning to its earlier model of subordinating academic performance to character, athletic ability, conformism, and membership of favored groups. It has appointed a president who is almost certainly the only Harvard professor in the humanities and social sciences not to have written a book. It has admitted students who are incapable or unwilling to do the academic work that elite universities used to require. And its global reputation is slowly eroding.

Substack

 I will be sending posts over to Substack. Here is the link to the Substack version of the previous post.

The Decline of American Universities: The View From Leiden, Ankara and Madrid

There has been a lot of talk recently about the crisis or crises of American universities. Certainly, if we look at the deteriorating financial situation, the thuggish behavior of demonstrators at Ivy League schools or big state universities, scandals about admissions, or fraudulent research then, yes, American universities do seem to be in a very bad way.

However, financial problems, violent extremism, corruption, and research fraud can be found almost everywhere. Is there a way to compare large numbers of institutions across international frontiers? There is no perfect mode of assessment, but global rankings can tell us quite a bit about the health or sickness of higher education and research.

When Americans think about university rankings, it is usually America’s Best Colleges published for more than four decades by US News (USN) that comes to mind. In the rest of the world, global rankings are more significant. The leader in public approval, if we mean governments, university leaders, and the media, is clearly the Times Higher Education (THE) World University Rankings. These rankings are characterised by bizarrely implausible results, sometimes dismissed as outliers or quirky statistics. In the last few years – sorry to keep repeating this -- we have seen Anglia Ruskin University and Babol Noshirvani University of Technology leading the world for research impact, Macau University of Science and Technology and the University of Macau superstars for internationalisation, Anadolu University and Makerere University in the global top ten for knowledge transfer. No matter, as long as the composite top fifty scores look reasonable from a traditional perspective and the usual heroes, Harvard, MIT, Oxford, are at the top or not too far away.

QS, another British company, was once THE’s data supplier but has pursued an independent path since 2010. Its rankings are more sensible than THE's, but it also seems to have an undue regard for the old Western elite. In its recent world subject rankings, Harvard was first in the world for all five broad subjects except Engineering and Technology, where the crown went to MIT, and Oxford was second in all but one.

These two, along with the Shanghai Rankings by virtue of their age, and occasionally the US News Best Global Universities, because of the fame of their national rankings, constitute the NBA of the ranking world. They are cited endlessly by the global media and provide lists for the appointment of external examiners and editorial boards and for recruitment, promotion, and admissions and even data for the immigration policies of the UK, Hong Kong, and the Netherlands.

However, there are other rankings based on publicly accessible data, transparent methodologies, and consistent procedures. They are largely ignored by those with power and influence, but they tell a coherent and factual story. They are published by universities or research centers with limited budgets and small but well-qualified research teams.

I will take three: Leiden Ranking, produced by the Centre for Science and Technology Studies (CWTS) at Leiden University, the Netherlands, University Ranking by Academic Performance (URAP) by the Informatics Institute at the Middle East Technical University in Ankara, and the SCImago Institutions Rankings (SIR) published by the SCImago Lab in Spain, which has links with the Spanish National Research Council and Spanish universities.

Leiden Ranking

Let’s start by taking a look at Leiden Ranking. The publishers decline to construct any composite or combined ranking, which limits its popular appeal. The default metric, which appears when you land on the list page, is just the number of articles and reviews in core journals in the Web of Science database. Back in 2006-2009, Harvard was in first place here, and other US universities filled up the upper levels of the ranking. The University of Michigan was third, and the University of California Los Angeles (UCLA) was fifth. Chinese universities were lagging behind. Zhejiang University in Hangzhou was 16^th, and Tsinghua University in Beijing 32^nd.

Fast forward to publications between 2019 and 2022, and Zhejiang has overtaken Harvard and pushed it into second place. The top twenty now includes several Chinese universities, some now world-famous, but others, such as Central South University or Jilin University, scarcely known in the West.

Much of this decline is due to China's advance at the expense of US schools, but that is not the whole story. UCLA has now fallen behind Toronto, São Paulo, Seoul National University, Oxford, University College London, Melbourne, Tokyo, and Copenhagen.

You could say that is just quantity, not quality, and maybe we should be looking at high-impact publications. In that case, we should look at publications in the top 10% of journals, where Zhejiang is still ahead of Harvard. It is only when we reach the top 1% of journals that Harvard still has a lead, and one wonders how long that will last.

That is just the number of publications. Academics tend to judge scientific quality by the number of citations that a work receives. Leiden Ranking no longer ranks universities by citations, perhaps with good reason, but does provide data in the individual profiles. Here we see Harvard’s citations per paper score rising from 13.31 in 2006-2009 to 15.71 in 2019-2022, while Zhejiang’s rises from 3.38 to 11.43. So, Harvard is still ahead for citations, but the gap is closing rapidly and will probably be gone in three or four years.

 

URAP

Turning to the URAP, which is based on a bundle of research metrics, Harvard was first in the combined rankings back in 2013-2014, and the best-performing Chinese institution was Peking University, in 51st place. Now, in the recently published 2024-2025 rankings, Harvard is still first, but Peking is now tenth, and Zhejiang and Tsinghua have also entered the top ten.

Other elite American universities have fallen: Berkeley from 5^th to 54^th, Yale from 18^th to 38^th, Boston University from 58^th to 151^st, Dartmouth from 333^rd to 481^st.

The relative and absolute decline of the American elite is even clearer if we look at certain key areas. In the ranking for Information and Computing Sciences, the top ten are all located in Mainland China and Singapore, with Tsinghua at the top. Harvard is 35^th.

Some American universities are doing much better here than Harvard. MIT, which I suppose will soon be known as the Tsinghua of the West, is 12^th, and Carnegie Mellon is 15^th.

In Engineering the top 25 universities are all located in Mainland China, Hong Kong, or Singapore. The best American school is again MIT in 37^th place, while Harvard languishes in 71^st.

 

SCImago

These rankings are quite distinctive in that they have a section for Innovation, which comprises metrics related to patents, and for Societal Factors, which is a mixed bag containing data about altmetrics, gender, impact on policy, web presence, and the UN Sustainable Development Goals. It also includes non-university organisations such as hospitals, companies, non-profits, and government agencies.

When these rankings started in 2009, and before societal factors were included, Harvard was in second place after France's National Scientific Research Center (CNRS). MIT and UCLA were both in the top ten, and the best-performing Chinese university was Tsinghua, in 80th place, while Zhejiang and Peking lagged way behind at 124th and 176th, respectively.

In the latest 2025 rankings, Harvard has slipped to fourth place behind the Chinese Academy of Sciences, the Chinese Ministry of Education, and CNRS. Tsinghua, Zhejiang, and Peking are all in the top twenty, and MIT, UCLA, and the North Carolina schools have all fallen.

Looking at Computer Science, the world leader is the Chinese Academy of Sciences. The best university is Tsinghua, in fourth place. Then there are some multinational and American companies and more Chinese universities before arriving at Stanford in the 24^th slot. Harvard is 64^th. 

In the next post, we will look at the causes of all this.