Hacker News Reader

Comment by pacala | Hacker News Reader

The PhD system is a Ponzi scheme. First and foremost, a PhD program trains one to become a professor. During a professor career, one is expected to graduate 5, 10, 20 PhD students. It is mathematically impossible for the majority of these students to find a professor job. The explosion of higher education, internally and globally, masks the problem to some extent, just like a good old fashioned Ponzi scheme appears solvable for a period of time.

Edit: It is pretty well known that there is exactly "One True Path" for a research career. See [0] or [1]

[0] https://blogs.scientificamerican.com/doing-good-science/care...

[1] http://ethicsandscience.scientopia.org/2008/11/20/excuses-fo...

View on HN

Quite certainly, becoming a professor is not "first and foremost" what PhD leads to (nor is it what many people do it for).

It is certainly not my experience. I am a PhD student and since the very beginning my mentors (program directors, my PhD advisor, committee members) have been pretty clear about future prospects: "the opportunities in industry are such and such (compared to the opportunities for BS and MS)", "postdoc positions have those requirements", "the chances for getting a permanent research/teaching position is so and so".

While there are people who are dishonest with their students, it is borderline offensive to lump them up with all the rest. All the people I have worked with have been honest about job prospects and have worked hard to give me the best tools to work in those jobs.

Right. I'm the Chair of a (neuroscience and psychology) departmental graduate school and I also sometimes sit in on departmental PhD 'scholarship' interviews. One of the questions we ask is 'Given that your chances of gaining a job in academia are only about 1 in 8, what do you think you will do when you finish graduate school?'. Then, when they arrive, we show them, a discuss, a version of this picture

http://thegradstudentway.com/blog/wp-content/uploads/2013/02...

during the induction lecture on their first day. They are all still there on day 2. I like to think they are smart enough to know what they are doing and that the only people who are still writing phd 'glut' articles like this are those whose ideas about graduate training were formed a few decades ago.

When I was a grad student in the mid 80s, we got the same basic news on our first day. We were still there on day 2 too. I think the idea that we were duped into attending grad school by inflated expectations of academic employment is something we fabricate in our own minds much later. A lot of just enjoyed studying science.

Not to discount your story, but weren't the financial implications of a PhD in the mid 80s and the mid 10s drastically different? If the benefits, the joy of learning and chance of getting a job, are the same but the costs, the massive amounts of debt you need in modern times, are much higher, but the number of PhD students is the same doesn't that mean something had to have changed? Maybe learning is more enjoyable now, which is certainly plausible with the internet, but maybe students are being liked to a higher rate?

Indeed, my parents put me through college, and I finished grad school with money in the bank thanks to the NSF.

At the time, the conventional wisdom was: Don't get a graduate degree unless somebody else is paying for it, or you are in a job that guarantees a salary bump for doing it. I don't think that's changed.

Unless you live in a high COL area where even NSF grant money isn’t enough.

The NSF grant money wasn't paid directly to students. The departments still set the pay scale, and had to deal with COL disparities like all employers do. And if your prof couldn't support you, then you didn't work for that prof, or you got a teaching assistantship.

At least in my field, physics, lack of sufficient funding meant you didn't get students.

This whole business wasn't risk free. There were students who dropped out because their prof lost funding. In my view, the risk of not finishing at all is a bigger cost of graduate education than the pay scale.

Lol.. UCLA pays grad students 18k ish. Barely enough to survive. Yes not finishing is a real risk but it’s not true that students can generally survive on that without loans or savings. In other lower cost areas, they can. MIT, Columbia and other schools also have better financial packages / housing options.

I remember even in the 1980s, there were schools where the pay packages didn't look all that competitive. It's possible that there are grad students who have family support.

At least in STEM, tuition is practically free and PhD students are paid for their research work and teaching. There are even PhD student labor unions in most public and some private universities. (Whether the pay is good and whether unionizing makes sense are separate topics).

You will not get paid well as a PhD student, but there is no massive amount of debt involved either (and again, whether there are opportunity costs is a separate topic).

Too bad you can't (edit: very difficult to) get a PhD, in the US today, without going at least into an order of magnitude more debt (or in debt at all) while in undergrad when comparing to the early 1980's.

I did hear of a German uni that accepts people for PhD programs without an undergraduate degree, but iirc you need like 4 or more publications. But that seems like an exception to everywhere else.

I think you meant to imply that it was difficult to so, but you said "can't".

You absolutely can get a paid for ba and ma in the US. It isn't easy, but thousands do so.

Yes, but this is a problem unrelated to PhD programs (the majority of people go to college and college is expensive).

Moreover a good high school mentor will help you navigate to good financial aid (I have friends who came from the lower middle class who did not pay a penny for their Yale or Harvard or Cornell undergraduate degrees). They did pay for books though...

> …(the majority of people go to college and college is expensive).

This didn't used to be the case, so, yes it is true to say it is unrelated to the PhD programs if you set aside what it takes for people to enter them in the first place and the decisions that have to be made by all actors (of which only the self has control over).

>Moreover a good high school mentor will help you navigate to good financial aid (I have friends who came from the lower middle class who did not pay a penny for their Yale or Harvard or Cornell undergraduate degrees).

Yeah sure, my sister got a full ride for undergrad and for her masters and my cousin got a full ride for undergrad, but for every person like them, there are many more that even leveraging such knowledge of opportunities and trying to pursue them in earnest, doesn't net them opportunities.

It's laudable that the likely end results are presented so clearly and early. Have you ever done a survey immediately afterwards to see how many students goals align with the likely outcomes? It seems like if a student is in that position where they were very successful at being a student for the prior 16 years and succeeded at getting into your graduate school that they might think they are the 1 in 8 - maybe all eight of them for a given eight.

Is it? One way to recruit a certain kind of soldier is to build up the notion of “elite” and then ask, “are you up to it?” Young people often leap at that, and you can always say you told them it would be tough.

I don't know. It does seem that law schools don't do this based on the numerous post graduation interview/surveys, where if memory serves if one doesn't get admitted to a top 20 school or so one might want to rethink a legal education as one's avenue to financial success.

By the time of the induction lecture, won't most students be too invested already to be able to go back?

This is a fair question, but you should also keep in mind that there is no stigma to dropping out of a PhD program (if you do not mind the opportunity cost of preparing and applying it is not the worst way to "take a gap year").

Can you give some examples of what a neuroscience PhD end up doing that is not science? I hope those 66% don't cover Starbucks!

That is great and all but I would still classify this kind of behavior as predatory and taking advantage of the young.

Payday loan places are also pretty transparent about the situation.

I think this varies somewhat by field, but I promise you, in many fields essentially the only reason to get a PhD is for a career in academia. The idea of a someone in graduate school not aspiring to become a professor still sounds odd to me.

And while I never had anyone outright lie to me or tell me job prospects were great, there were also few people who gave me a really clear picture of just how bad they were.

Certainly it varies by field, I don't know a whole lot of English PhDs in industry, but are a multitude of industry jobs require STEM PhDs (e.g, research positions, quant positions in finance, etc.). Whether you agree those positions should require a PhD is a separate argument.

I hope that part of being accepted into a PhD programme involves the ability to realize that there’s not much room at the top.

Personally I loved my PhD experience even when I knew that the race to tenure was not for me. In many ways the reward of a science PhD on the jobmarket is large. At my current company we are treated as superstars among both engineers and management, and there is plenty of ways to make use of whatever lessons learned. (Disclosure: experimental particle physicist working in the European defense industry).

I'd say it varies more than somewhat, but rather a lot, by field (and maybe by school).

And even advisor - some have clear paths directly into industry jobs.

On the other hand, there are fields where the opposite holds. A family member studied Social Sciences, and the non-academia job situation (in journalism, museums, etc.) is pretty much "PhD or bust".

Depending on field, private sector R&D positions can require PhDs

A PhD is literally )as in that’s what the abbreviation stands for ) a license to teach. It’s origins are exactly in training people to pass the torch of knowledge.

Look at the Greats, or even professors until the 50s, they only had at most 6 grad students their entire career.

Many modern professors have as many students a year.

Plenty has changed since the word was created. Similarly, philosophy used to mean all of science which is certainly not the case today.

In other words, PhD does not mean "a license to teach" and it has not for at least a century.

I am not saying all these changes are for the good, but the premise in your comment is not true.

Sources for any of this?

> First and foremost, a PhD program trains one to become a professor.

You are grossly misinformed. PhDs programs are training for a career in research. Professorships are only one possible path... and just like you intuited, it's not even the dominant career path.

It’s not the dominant career path because there are too many applicants for too few jobs.

In many fields, academia is the option for a career in research, and a viable career in academic research means a professorship.

I think this is more of an indictment of how little real investment goes into R&D at private industry - even as the companies merge into megopolies who should have the finances to best justify it, they seem to become even less inclined to invest in fundamental research.

The big (and some even not-so-big) companies hire lots of PhDs -- roughly 8% of Google's employees have doctorates.

That doesn't imply that all those people needed a PhD for their job, though. I worked with multiple people at Amazon who had PhDs who were doing the exact same job I was.

Some people do their PhD while they're working. You might want to change career paths, e.g. specialize in some machine learning subfield.

I'm pretty sure Doctors specialize further, whilst working, as well?

Google is exceptional, and also in area of "growth capture". How about large companies in more mature industries? For example, do any of the telecom companies maintain the equivalent of Bell Labs today?

Google is infamous for valuing academic achievement over practical skills in interviews, 8% is actually much lower than what previously reported - I guess they lowered their standards of late.

If google is at 8%, everyone else of comparable size will probably be around 3% at best.

A PhD program trains one to become a researcher, not a professor. There is a large need for researchers in industry.

Well, there's a need for certain kinds of researchers in industry. How large that is isn't exactly clear. The difficulty comes to PhD programs when people are able to become researchers without a PhD.

It's also not impossible to become a professor without a PhD. That doesn't make PhD programs useless for that purpose, as long as they make that career path easier to attain.

It's pretty unlikely. In one case there was a tech where I went to grad school, who performed so well that tptb decided she should be a professor. But before they could do that, they felt the pressure to give her a PhD (and where I went to grad school was very much on the "unconventional" end of the spectrum), so they enrolled her in the PhD program, made her take classes, and gave her a degree after she had finished enough classes to cya in the eyes of the accreditors, made her staple her papers together for a thesis, gave her a PhD, and then promoted her to professor.

Part of this was because consider if she decided to leave our institution... There are certainly places where she would have been denied a job opportunity on the grounds that she did not have a degree, e.g. state schools where there are often legally or administratively mandated minimum requirements for professorship and bureaucratic decision-makers who filter out non-conforming applicants before they are even considered.

Also there's lots of government researchers as well. The US has its NIH, FDA, CDC, DOE, USDA, and so on, and other countries have their own versions (if generally not quite as large).

> First and foremost, a PhD program trains one to become a professor.

No it doesn't! It trains you to be a researcher. You can be a researcher without being a professor.

For many (I would go out on a limb and guess most) fields people may study in graduate school, there is no research outside of academia.

In most countries, being a professor is the only way to have a permanent position as a researcher in academia. You can be a researcher without being a professor, but it may involve changing your position (and country) every few years.

I can't understand why you think this is the case. You can get a full-time job in industry being a researcher. Why would you need to move every few years?

As the first step on the path to understanding, first consider: "in what fields do people get doctorates?" As a second step, consider for each of those "what industry will employ them?"

If you step over to the non-STEM (and to a lesser extent non-TE) fields, the answer to the second question tends to diminish to "academia".

I mean, yes, it is possible for me to do biochemistry in my garage without a professorship (and I do), but will people actually take you seriously? Will you get money to do your research? Where will you publish? Not all of us can be Peter Mitchell, and arguably it's harder now that government funded research has a near monopoly on the research industrial complex.

I guess it comes down to why are you doing the research in the first place. Is it for your own curiosity and self-fulfillment or is it for "people to take you seriously"? I'll admit your point about resources is completely fair and oftentimes you need the prestige to obtain the necessary resources. However there is plenty or research that can be done without insane amounts of resources (<15k). As someone who does research professionally in an area that requires minimal resources (a start up cost of 10k and reoccurring costs that are much less). I do about as little as I can get away with in terms of disseminating my work. I do the work for my own personal motivations and am not concerned about being taken seriously.

There are actually some benefits to this approach in that I don't have to work on "safe projects" that are deemed "serious research" by the community. It is actually quite liberating to work on things that others might scoff at and call you a "quack" or "wack job". And even if I am a "quack" I can honestly tell you that I am happy and feel fulfilled. Just my two cents. Life is too short to worry about everyone else.

Ah, but you can be a researcher (doing work at least as good as most PhDs) without a PhD.

Therein lies the rub.

You can also be a good, e.g., mechanical engineer without getting an engineering degree, but this is an exception to the rule.

Also, companies use the degree as an indicator of skill, so the good engineer who doesn't have a degree needs to somehow reliably demonstrate a solid understanding of the field to compensate. Obviously, from the company's point of view, one is easier to deal with than the other.

The same applies to PhDs and research. A PhD proves that the person has followed through with and understood a very precise topic of research for 2-3 years.

> You can also be a good, e.g., mechanical engineer without getting an engineering degree, but this is an exception to the rule.

That's an awfully incorrect analogy. Enrolling in a PhD program has absolutely nothing in common with getting an engineering degree. An engineering degree is conceived to give the students a pre-established set of skills and technical knowledge that meets industry demands.

A PhD program is set to help the PhD candidate research a very precise and specific topic, which more often than not is entirely irrelevant to any industry, so that in about 4 years he is able to produce some scientific findings and defend a thesis in front of a jury.

The key difference is that an engineering degree is tailored to teach marketable job skills while the PhD program is the exact opposite.

> That's an awfully incorrect analogy.

I would say "not entirely accurate", rather than "incorrect". Are you familiar with what happens in PhD research and the skills you can acquire during a PhD?

> [.،.] so that in about 4 years he is able to produce some scientific findings and defend a thesis in front of a jury.

And that's exactly what having a PhD proves: that you can start from a cursory understanding of a topic and end up contributing something novel to the field.

Given that industrial research requires you to do practically the same thing -- albeit in a more applied fashion -- a PhD is usually a good indicator when hiring for such positions. Again, there may be more suitable people for the position who don't have PhDs, but that is once again an exception to the rule.

> Are you familiar with what happens in PhD research and the skills you can acquire during a PhD?

Yes, very familiar. I've worked as a researcher in a public university and enrolled in a PhD program, but I dropped out after it was made patently clear that the whole thing was a complete waste of time.

I even had my advisor pressure me to continue working for the research group with scare tactics such as warning that getting a job would be virtually impossible and working for the research group was my only viable option.

> And that's exactly what having a PhD proves: that you can start from a cursory understanding of a topic and end up contributing something novel to the field.

...and that on its own is worth absolutely zero. You either have marketable skills or you don't, and a PhD degree does not help a PhD candidate develop marketable skills, It's years of drudge work for a research group that managed to get funding for areas which more often than not are entirely irrelevant. To develop marketable skills the PhD candidate needs to devote extra time that he doesn't have to go out of his way to make himself relevant as a job candidate.

"The PhD system is a Ponzi scheme."

"It is mathematically impossible for the majority of these students to find a professor job."

Assuming that PhD's are 'smart' people, I would gather they understand all of this.

And of course 'prof' is only one role for PhD's in the world.

Well... there's intellectual understanding and then there's emotional understanding. These are people who were the very top in their high school and the very top as undergrads. They've always made the cut when most everyone else didn't, so why should a tenure track position be any different?

This. Plus psychological investment in identity as a scholar. Makes it very hard for some people to hear “it’s not a good idea” and harder to see when it’s time to cut losses.

"So why should a tenure track position be any different"

Economics and reality come into play as we are adults.

There's basically no excuse for 'the smartest among us' to not have a basic grasp of the dynamics of how their education will play out into their careers.

But they've always been the smartest in a group of the smartest people. It's not that they're not being smart or not being adult. It's that they feel this is just another step in a road they've been traveling since they were kids.

> The PhD system is a Ponzi scheme. First and foremost, a PhD program trains one to become a professor. During a professor career, one is expected to graduate 5, 10, 20 PhD students.

In the US at least, a substantial proportion of professors teach at institutions that are undergrad-only or where only some departments have graduate programs. I am a tenured professor of computer science and have not supervised any graduate-level research (nor will I be expected to do so).

...which is not to say that the numbers aren't all messed up. They are. And I've had many lucky breaks (and also some unlucky ones) to get where I am. But your Ponzi claim presumes that the only outcome of a PhD is an R1 professor position, which is not the case.

Maybe in the US, here in the Netherlands you get a reasonable wage and usually finish withing 4 years during which you learn to write papers, keep yourself busy and determine the direction of a research project by yourself (at least more and more over the 4 years). Afterwards you can make the step to industry, quite comfortably.

During my first year I was already told, 1 of the 100 of you will become professor so I had no illusions.

Few PhDs in my field go into academia, most go on to laboratory jobs. It's a relatively small area though (nuclear engineering) and demand is fairly high for PhDs. A large portion of PhD students are working on projects jointly with national labs, so there is a lot of preparation for jobs beyond being a professor. A significant portion also go into unrelated areas, e.g. someone I know just took a job at Github.

More like the entire higher education system. This is what happens when businessmen run an education system and not True academics.

There's a strong opposite argument, you should probably familiarize yourself with it to have something to think about.

http://www.thenewatlantis.com/publications/saving-science

What does an education system have to do with academics? It makes way more sense when educators provide education and education managers run the system.

We could build new universities. Take up the slack in all the excess PhD's, increasing the supply of good secondary education, and help stave off exploding tuition costs. Build it from scratch like Carnegie Mellon, Stanford, or Olin. Who is "we"? I dunno, some rich guy maybe.

and each department at said university will employ what, 20-30 PhDs in each department? who themselves will then go on to produce a pile of new PhDs in their career?

at what rate would new universities have to be constructed in order for this to do anything meaningful about the issue?

Reply on news.ycombinator.com