Many economists call the current era of technology growth a boom era, not unlike previous gold rushes such as the Dot-com bubble. But the thing about bubbles is, they usually pop. And that has some people concerned. Is another bust on the horizon?
It’s not only tech employees who are paying attention to these patterns. In higher education, the number of computer science bachelor’s degrees follows market trends in finance and technology in particular—growing when times are good and plummeting when economies crash.
Since 2010, computer science majors have again been increasing, going from about 39,000 to more than 64,000 in 2016. And the Computer Science Research Center claims that the current enrollment surge has in fact exceeded previous CS booms. But what have we learned from these patterns? And what can it tell us about the future?
Mehran Sahami, professor and associate chair for education in the computer science department at Stanford University, has witnessed and followed the pattern closely. The conversation has been edited and condensed for clarity. You can listen to a complete version below, or on your favorite podcast app (like iTunes or Stitcher).
We’ve seen there are increases and decreases in CS majors that seem to match patterns in the tech industry, such as more CS majors leading up to the Dot-com boom, and then a real sharp drop off after that. How have you seen this play out in the CS department at Stanford?
Sahami: Well, it tracks pretty closely. About a decade ago, I did an analysis where I looked at the number of students declaring computer science at Stanford each year, and correlated that with the NASDAQ Composite Index, which is a tech-heavy stock index. And the correlation is actually extremely strong. You find things like the peak in 2000 for the NASDAQ corresponds exactly with the peak in enrollment at that time, then they both drop off.
But there’s an interesting thing. At some point, they do diverge, and there’s certain other reasons you can look at for that divergence. For example, there’s a drop after the Dot-com bubble burst around 2000. You fast forward a few months and you begin to see the drop off in student enrollment. Then at some point, when the NASDAQ begins to recover, you see the student enrollment continue to drop, and it’s because there was news around that same time that jobs were being offshored. There’s this perception among students that even though the economy may be recovering, that’s not necessarily translating into jobs for them domestically.
But then there are other turning points too, where starting around 2008, we start to see an increase, which grows pretty significantly, of the number of CS majors at Stanford, and that was kind of a leading indicator to the national trend, where we’ve certainly seen an increase in CS majors nationally.
Now that we’re in this moment of an uptick again for CS majors, does this tell us anything about the market for tech jobs today?
I think there’s clearly a lot of demand for tech jobs now. Depending on who you talk to, there’s an estimate that puts a ballpark of about 500,000 unfilled IT jobs nationally. And so, given that there’s a lot of opportunities there, it’s economically a very healthy field, and there’s a lot of job prospects. I think that does affect students.
Could this be any indicator of another tech burst?
Depending on who you talk to, some people say we’re in another bubble now. Some people say it’s a fundamental shift. I think if there is an economic downturn in the high-tech sector, there will probably be some accompanying effect with the number of students deciding to go into CS. But the numbers are so large now that even if they were to fall at Stanford, say by 50 percent, which would be a pretty extreme drop, our numbers of CS majors each year would still actually be larger than the peak at the 2000 Dot-com bubble, because our increase in enrollment has just been that strong.
Currently, there are more students studying CS at Stanford than ever before. How has this affected computer science professors? Is there a need for more, or have you been able to keep up?
Oh, absolutely, there’s a need for more. The growth in the faculty, for example, has not at all kept pace with the growth in the number of students. That has to do with a lot of factors. Part of it has to do with the way universities operate. In some sense, they’re slower-moving ships. They don’t respond exactly to the enrollment growth that happens from time to time. When you hire someone at a university, you’re potentially hiring them for life with the tenure system. And so, there tends to be less direct responses between enrollment and size of faculty.
But I would certainly say, we’ve seen the national trend that there’s a shortage of faculty candidates. And you sort of get the double whammy, in that when there’s more economic prospects and more people are interested in going into industry—as opposed to academia—which can potentially make it harder to hire. And then, at the same time, you have more students who are enrolling in the courses in the academy, so you have more demand.
At the same time we’ve also seen an increase in the number of coding bootcamps. How does that kind of fit in, when we’re seeing more and more CS majors? Do you think that that’s part of this boom period that we’re in right now?
Yeah. I think coding bootcamps are what you would naturally expect to see, given the demand for these kinds of skills. And if there isn’t enough capacity in university to train students for these skills, then you’re going to see other kinds of educational opportunities like coding bootcamps emerge.
I think the big question that still needs to be answered, because they haven’t been around that long, is: Do the bootcamps actually provide enough training to give someone a foundation for a 40-50 year career in technology? Or are they really just trying to get enough skills du jour, of the day, to students to try to place them in a particular job? And again, there’s conflicting opinions on that.
If there were another tech burst, how might the coding bootcamps feel the effects of that, compared to a university CS program?
Well, in our CS program, we’re trying to give students that foundation for a 40-50 year career. I think if there were to be a bursting, in some sense, of an economic bubble, you might see our student numbers come down a little bit. But I think, given the growth as I mentioned before, we’re going to still have very healthy numbers in our program.
So that’s not something that we worry about from the standpoint of enrollment. I mean, there are other worries about the impact on the economy, the impact on people. But in terms of enrollment, that’s not a huge worry for us. In some sense, it would ease the pressure we have.
I think for the coding bootcamps, it’s going to vary a lot more, based on their quality and their reputation. In some sense, if there is a downturn, and there’s less demand for these skills, how are students going to decide whether a coding bootcamp is still right for them, or which is the right one for them? And so, in that sense, they don’t have the track record to be able to provide them some kind of buffer in a downturn.
But it depends when that happens, and what kind of reputation they build up. Where do their students go, and what kind of comfort do people have with respect to the quality of the education they think they would be getting there?
Another important question here, of course, is how women are largely underrepresented in CS programs, and tech careers. I believe the highest year for women CS majors was around 1984. Why has that gone down, especially amid this period where it seems like there’s some reflection in Silicon Valley about getting more women into tech?
You’re right, there was a larger number of women in the early-mid 80s in computing, and then that went down, and it’s going back up again now. And there’s different theories around it. One of the things we’ve seen at Stanford, for example, is as the number of people in computing in general increases, the percentage of women, not just the number of women, but the percentage of women, at least at the college level, goes up.
You get these effects where if the major gets larger, even if women were a smaller percentage, there’s more of them, and that forms a community. Some of the barriers that might have existed previously, in terms of the program appearing as though it wasn’t welcoming to a particular group of people, now becomes more welcoming. One of the things we’ve seen at Stanford is that there is also a pretty strong correlation between just the size of our program and the percentage of women. So, it’s been going up over the past few years, and at this point, it’s a little bit over 30 percent.
Is there anything else from your own research in this space that you think is important for folks to think about when they’re choosing a CS major, or for people who work in this space?
I think something that’s important for parents to think about for the younger generation is earlier exposure to computing. One of the things that we’ve seen at Stanford is that our introductory classes are almost 50/50 in terms of men and women. But then the percentage of women drops off in the subsequent classes, and it made us wonder, why is this the case? Is there something about the introductory class that’s not welcoming, which might be the initial thought if someone sees the percentages declining after that class?
What we found was very different, which is if you look at when students take that first class, men tend to take it earlier in their academic career, especially freshman year. So, if they like it, and they decide they want to major in CS, they have the opportunity still. Women tend to take it later in their academic career, junior or senior year, and if you’re taking your first CS class senior year, even if you like it, it’s impossible to finish the major in the remainder of the year.
So, I think the important thing is earlier exposure for everyone, so they can, one, be able to see whether or not this is a field that might be of interest to them to pursue later on. But two, also to get more of the technical skills that everyone is going to need in the future.
And for someone who is maybe wary of another tech burst, is it worth even starting that CS education early? What’s your advice for them?
I would say it’s important for someone to pursue something that they really feel passionate about, and that they feel like they want to pursue through upturns or downturns. In terms of making career choices, or those kind of life choices, it’s more than just what the economic opportunity is today.
When I was graduating around 1992, it was the first time IBM had layoffs, and there were people wondering what the future of jobs in technology and software was going to look like. But the group of us that were doing it then were doing it because we loved it.
