clock menu more-arrow no yes mobile

Filed under:

Sperm Counts Are Falling All Over the World—Why?

Average sperm counts have fallen by 50 percent worldwide, Harvard’s Jorge Chavarro is here to discuss the data behind this major shift

“Mini-Mondes En Voie D’Illumination - Mini Worlds On The Way To Illumination” : Light Festival At Jardin Des Plantes In Paris Photo by Edward Berthelot/Getty Images

In the last 50 years, average sperm counts have fallen by 50 percent. This isn’t just happening in the U.S. or Europe or Asia. It seems to be happening everywhere. If the current rate of decline continues, researchers concluded, the average male sperm count will fall so low that the typical guy in every advanced economy will be infertile by 2050. Harvard’s Jorge Chavarro, a professor of nutrition and epidemiology, breaks down the data on declining sperm counts and tells us what it means, what might be causing it, what men can reasonably do to avoid it, and how bad it could get.

If you have questions, observations, or ideas for future episodes, email us at You can find us on TikTok at

In the following excerpt, Jorge Chavarro explains the significance of the data accrued by a worldwide meta-analysis concerning sperm count levels that indicate an alarming rate of reduction, possibly occurring since around the mid-1900s.

Derek Thompson: Let’s talk about the most important finding of this meta-analysis, and you can draw back from that and tell me what the most important fact in this field is. What should we know? Is it the decline in sperm quality? A decline in sperm production? What is your headline here?

Jorge Chavarro: I think the headline is that over the course of the time period covered by the meta-analysis, roughly from the 1980s to 2020—so over a 40-year period or so—total sperm count has reduced by about half worldwide.

This is after taking into account some issues that weren’t addressed in previous versions of these studies, including: Is this only happening in rich countries? Is this only happening in North America and Australia, New Zealand and Western Europe? So this new meta-analysis includes data from South America and Central America, from Africa and from Asia, and it appears that the same thing is happening worldwide.

Does it have to do with the type of people who end up giving a semen sample for research and, therefore, there’s data that eventually gets published on semen quality? That is addressed. That was a bigger problem in the first version of this paper and it is absolutely a non-issue in this version of the paper.

Then what was probably the biggest criticism in the 1990s? The 1992 version of this paper and then the 2000 version of the paper is: Is what you’re seeing differences in what’s happening at the lab? There really wasn’t a consensus on how it is that you should analyze the semen sample until the late ’90s. More recently, the WHO has established guidelines that have become pretty much universal on how you process and how you evaluate a semen sample. But in the older papers, people were using all sorts of different methods for counting sperm. So that was a big issue in some of the earlier versions of the paper.

The newest one only includes papers in the meta-analysis that used the standard WHO protocol so, again, it is less of an issue.

Thompson: That’s really interesting, because I did a lot of reading about this subject in the last few days, and it seems like there’s been a controversy in the last two years.

Some people think that this issue is being catastrophized, that it’s not nearly as bad as the meta-analysis suggests. You’re saying that in previous years or even previous decades, people said this is self-selection bias. People are only turning in sperm samples if they have reason to believe that they might have less sperm concentration or if they’re having some difficulty with fertility. Maybe it’s self-selection bias, maybe it’s mostly a phenomenon of lab analysis.

But you’re saying this meta-analysis we should trust because it resolves or directly answers a lot of these criticisms that had been levied against this kind of research. Is that basically right before I move forward?

Chavarro: That’s basically right. I think with each different version of the paper over the last 30 years or so, this group has gotten closer to answering the question, Is there a downward trend in sperm counts? I think this most recent version of the paper gives us the best possible answer to date, which is probably yes. Probably sperm counts are going down.

Thompson: Tell me about when this decline happened, because you’re looking at a difference between the 1970s, early 1980s. You said there was a meta-analysis published in 1990, and now we’ve got an update published in 2020. When did this decline happen? And is it a decline that’s over, or are sperm counts continuing to fall in these populations?

Chavarro: So when is it happening? I guess it depends on how much you want to believe previous versions of the meta-analysis.

If you’re willing to say— okay, maybe the methodological problems that were present in the 1992 version of the paper are not that big of a deal, and we can use that as an initial guide of how long this has been going on: That paper included papers from the late 1930s. So you could say, in the worst-case scenario, the answer is, “It maybe has been happening since the late 1930s all the way through today, to the 2010s.” The last paper published that’s included in the meta-analysis is 2019, so it’s probably 2010s.

If you want to be more stringent as to what is the type of data that we should be allowing to be included in this type of analysis, then you would say, well, it’s probably something that’s been happening from the 1970s forward.

So definitely at some point in the second half of the 20th century, maybe ... a little earlier, and definitely still happening today.

Now the question—like you said—“Is it going to continue to happen?” We don’t know, and in my mind, it really depends on why. So yes, it’s going down, but why is it going down? Is it going down because whatever is explaining this downward trend has saturated the possible effect that it can have on our population level and semen quality? Or is it something that can continue to impact sperm concentration and sperm count at a population level?

That question, in my opinion, cannot be answered today because we do not have a very good understanding of what is causing this downward trend.

This transcript was edited for clarity. Listen to the rest of the episode and follow the Plain English feed on Spotify.

Host: Derek Thompson
Guest: Jorge Chavarro
Producer: Devon Manze

Subscribe: Spotify