Hacker News Reader

Comment by rdtwo | Hacker News Reader

Why would you want to get kids into science? It don’t pay for shit and career/life prospects are grim. Grind out a phd for 80k starting salary? Hard pass

View on HN

An understanding of and willingness to do science is a broadly applicable skill across many different disciplines (even if you don't get a degree in it), as well as being a useful life skill. It's something kids absolutely should be exposed to even if it's not their passion nor do they wind up doing something as an adult which would be generally labelled 'science' (which is a shorthand usually meaning natural sciences specifically).

As a scientist, I'd say this claim doesn't hold true for many scientific disciplines. I know scientists working in biopharma earning over $1M per year.

That said, I do think the salary mentioned is accurate as a starting salary in academia for a research faculty position (tenure track starting salary in USA is ~50% higher though), but academia pays much worse than industry. You give up salary for huge amounts of autonomy.

You also have to factor in how terrible post-doc pay and job security is, just to have a shot at getting on the tenure track. In the US it's half of that 80k number in bio research.

We also need to dispel the myth that tenure track academic roles are the only jobs for scientists. In my field, one can have a wonderful, decently-paying career working in some of the most desirable places in the US if they go the national lab route. There's also a thriving private sector which will just get bigger. This doesn't even factor in the exit ramp careers if, late in the game, a student realizes that "science" isn't quite what they want to do for their 9-5 - e.g., trading in all of those hundreds or thousands of hours grinding data analyses for their Masters/PhD for a career in data science or software engineering.

We should make sure that folks understand a modern career in science may look, from a salary and QoL perspective, quite like a doctor or other professional. High-stress, relatively low-pay, challenging from your late teens to late twenties, then a rapid increase in earning power and opportunity into your early thirties.

I completely agree, but I think most people who arrive at this perspective only reach this view after spending time in industry or adjacent to people working there.

A lot of the focus on tenure track jobs comes from professors and academics who dedicate their life almost solely in academic environments (so in their view, it’s a rather unhealthy attitude of academic tenure or bust). I’m not sure how feasible this is, but perhaps a work-study private sector summer internship could be an encouraged part of certain PhD, to widen perspectives and help both students and professors better understand the possibilities out there, which someone can be motivated to strive for, rather than settle for.

Absolutely. There have always been domains where industry roles have been just as sought after as academic ones - think Computer Science broadly, and all the opportunities one has with a PhD there. Other domains have been slow to catch on to this, but it seems like there has been an inflection point over the past 10 years where we're just so grossly overproducing PhDs in so many domains relative to academic jobs that there is mounting pressure on degree programs to revamp their training/prep for "alternative" career paths.

Myself and colleagues are definitely trying to encourage our alma mater to have more support for externships other opportunities in industry. I think things will continue to change - for the better, for early career scientists - over the next 5-10 years. If only because there is a workforce crisis in my field where skills like AI and software engineering are in extremely high demand, but even 3-4 years ago we actively discouraged PhD students from pursuing this type of work!

In my time as a PhD student (granted still there currently just at the finishing things up stage) I was actually part of founding an organization for students to basically do this. We invited speakers from different fields and had them talk about "alternative" career paths considering the current system basically assumes you want to become a professor. We also included a workshop every year where we would build in career skills and have professionals come and speak. You learn very quickly that there is an incredible range of activities one can do with a PhD.

> “You give up salary for huge amounts of autonomy.”

I heard otherwise from a staff scientist at a university in biological sciences, though I’m not sure how representative his view is. He said that non-professors lack research autonomy, and even then, there are limitations of what one can fund due to grants (though I’m not sure how accurate this is, as I used to assume that academic grants generally had fairly flexible conditions for using the funding).

This is all accurate. There is flexibility in funding, but if you use your grant money for a completely different project than you proposed, you better hit an absolute home run or you won't get any more grants.

> You give up salary for huge amounts of autonomy.

The "autonomy" is a facade in many instances. Professors have to do what can be funded. For a lot of professors that means doing what's trendy or what panders to moneyed interests rather than what they'd actually like to do and what would be more valuable.

Read chapter 4, titled "Assignable Curiosity", in the book Disciplined Minds for more detail.

https://disciplinedminds.tripod.com/

(Note: There's a lot I disagree with in this book, but I think this particular chapter is accurate.)

This is true if you need large amounts of resources ($) to achieve your research goals. At least for me, I have not found this to be true.

However, what I was thinking about when I wrote that was that I control my schedule, calendar, research objectives, and have a yearly check in on my performance. In contrast, my experience in industry was VERY different across the board.

> This is true if you need large amounts of resources ($) to achieve your research goals.

My experience is that the quoted situation is the exception in US academia. Despite paying a pittance, graduate students are expensive enough to make professors dependent on the approval of funders. And that's where the loss of autonomy comes from.

It seems to me that certain parts of academia that are more funded by teaching like pure math may be more immune to this, but STEM broadly seems accurately described by the chapter I recommended.

> However, what I was thinking about when I wrote that was that I control my schedule, calendar, research objectives, and have a yearly check in on my performance. In contrast, my experience in industry was VERY different across the board.

I agree with what you said, aside from that I don't think you have full autonomy over your research objectives. If you think you control your own research objectives, I recommend reading the chapter I mentioned. As far as I can tell, your interests are well aligned with what's supported. I looked you up, and was not surprised to see that you work in deep learning, one of the most supported and hyped fields around today. People (like myself) whose interests aren't aligned with what's supported often find academia to be quite hostile. The relative research autonomy you feel is real, but if you decide to move outside of deep learning, you may no longer find the same sense of research autonomy.

Probability distributions are the difference.

There are people earning $200k, $500k, and $1M+ in many fields.

The probability of getting there (and quality of life while getting there and after you get there), are the important bits of information.

For example, if you have to grind out my whole 20s (some of the best years for the human body) doing dissertations and working 80+ hour weeks for near minimum wage (per hour worked), only for a 10% chance to make it to $200k per year or more, maybe it is not worth it.

Especially if other options exist that provide much higher probabilities of attaining an income you want with similar levels of dedication.

Unfortunately, in the USA that's true across PhDs and MDs. It requires about 10-15 years post high school to make more than a living wage.

At least for everyone I know, they all ended up with comfortable salaries after finishing their PhDs. Most people spend our best years spending the majority of our time working, and at least for me, scientific research makes it not feel like work. Not everything is about money and a professor can live well in many places on a $90K salary, speaking from past experience.

The biggest cost is probably the impact it has on trying to have a family since one doesn't typically have much salary or time until around age 35 with the typical trajectory.

Yeah that can be an issue since you basically only have 2-3 years left to have kids if you want them and are a woman.

Also 90k ain’t what it used to be. It’s probably equivalent to 60k 3-4 years ago.

While the ROI on MD has gone down quite a bit, it is not as volatile as PhD. The floor for MDs is quite a bit higher, and the path to high income is well defined (less random to achieve it if you go through the steps).

https://www.medscape.com/slideshow/2022-compensation-overvie...

grad school wasn't a grind in my experience. Definitely easier and more fun than a real job, apart from the sparse high-stress deadlines of course. I think that's generally the norm in the US (except in chemistry I hear?)

Not all people have the motivation and self discipline to force themselves through a career in something that they hate, solely for the money.

I learned programming as a kid in 1981 and fell in love with it. I knew about the demand for programmers, as my mom was teaching programming at a nearby tech school.

My college internship was at a large computing facility, which employed a number of programmers. I formed the impression that programming as a job was incredibly repetitive, stressful, and boring. I majored in math and physics, and ended up with a PhD in physics.

I do plenty of programming today, but on my own terms. I work at a place that has a large programming department, and my opinion of programming as a job has not changed. On the other hand, if someone loves that kind of work, or can at least grind their way through it, they're welcome to the money.

It may have taken me a bit longer to reach a decent salary level. Today, I'm the highest ranking "technical track" worker in one division of a F500. Had I gone into programming, I might have reached a plateau, or burnt out.

More than half the American families earn less than 75k household income[1]. So the career/life prospects are grim either way.

[1] https://www.statista.com/statistics/203183/percentage-distri...

They won’t be Able to make it to a career in science. Since you need a PHD or at least a masters to do it you need to go through some top tier universities and invest some serious $$ into it. Science is for love is only a marginally better career choice than art.

My family made noticeably less than that, even considering inflation since then. In my home town the average family income is around half that in today’s dollars. I’m a high income SWE now. Growing up in a <$75k HH income family doesn’t imply grim prospects, that’s typical middle class. Everyone I grew up with would consider $75k high income bordering on rich.

Because it gives them tools to understand the world. Because not everything you teach kids has to lead to a PhD or a specific career? Because 'science' plays into engineering, finance and tech sector careers?

This has got to be some sort of satire.

To me, getting kids into science is not just a path to PhDs, but to engineering and technology in general, and that is a very lucrative area money wise.

I'm not able to tell if this comment is satire and it worries me.

I don't think it's satire.

STEM is pushed hard, but realistically, science and math dont have that great career prospects. For one thing, a PhD, and the 4-7 year opportunity cost that comes with it, is pretty much required to get anything more than a lab tech job.

After that PhD, where do you go? You can try for an insanely competitive tenure track faculty position. Or a very limited number of industry researcher jobs.

The E in STEM (traditional engineering) isn't exactly tech in terms of salaries and job availability, but it mostly guarantees paying the bills with a decent WLB after "just" a bachelor's.

I mean, usually you just end up trading stonk at Renaissance Technologies or PanAgora. I'm not studying anything finance related and haven't even finished my phd and they've still been actively after me to sell my soul for mid six figures. The worst part was, I did the interviews to get an offer and everyone I talked to was tangibly brilliant- like could really be changing the world instead of this. I almost did it for the money + chance to work with a bunch of actually smart people, but the moral implications barely won out.

Knowing what I know now, I would have jumped ship in your shoes in an instant. After 5 years of that work you could be free from working ever again and do whatever you feel like. I've heard that RT for example can pay you up to 1M per year for entry level, and looking at their profit margins I don't really doubt it.

Had to look it up.

The starting salary for a research scientist is $170 - $205K.

https://www.rentec.com/Careers.action?researchScientist=true

At those places the real money is made by being allowed to participate in the employee-only investment funds.

RT has a yearly return averaging around 40%, and the employees only portfolio averages around 70% per year (at least this was the case a few years ago still). I would imagine the major part of the compensation comes from bonuses and the opportunity to invest in the employees only fund.

170k-200k is not that different from ordinary banks, I can't imagine it's all RT would pay.

Edit: had to look up the annual returns and they seem to be a bit different depending on where you look, so take those numbers with a grain of salt. In any case they're consistently the most profitable hedge fund out there.

What moral implications? Quants at RenTech are basically just card counters in the biggest casino in the world.

I came to the conclusion that most likely, taking the job was morally neutral- barely hurting nor helping anyone outside the firm (not specifically RT- don't want to out myself); but that on a larger scale our country is rotting because our most capable thinkers are doing this instead of anything morally positive (solving healthcare, solving the climate crisis, leading the country), and I didn't want to participate in that brain drain.

Looking at my own motivations, one of the big draws of the position was that everyone I talked to at the firm was competent and pleasant, while in domains solving real problems (again for example healthcare and climate) 70% of people I've interacted with are stupid, petty, or both. But stepping back, isn't that state of affairs existentially horrifying? If that status quo persists, will we even survive as a species?

Fair. It's really a shame that society incentivizes its smartest to go into finance rather than a positive-sum endeavor. If it's any reassurance, the world has pretty much always been this way.

And I'm pretty surprised everyone you met was pleasant -- that was certainly not my experience at a hedge fund.

The current labor market is paying much better than before for scientists with just a B.S.

I do agree that there is little economic value in most Ph.D. programs.

From my experience working in biotech I would say certainly a B.S. in a scientific field (like biochemistry, etc.) There is often a hard cap in the career trajectory without a PhD. Often there are enough PhDs out there that moving into management in industry often requires this vs. Stagnating as a lab scientist without a huge upward spike in income/responsibilities.

A PhD in maths is _incredibly_ valuable. Anything abstract algebra related, and the likes of the NSA are all over you. Anything numerical analysis related will find plenty of opportunities in finance, and the industry just calls statistics specialists "Data Scientists".

> NSA

> finance

You're not selling it.

Yeah, I remember my Physics degree: "With a Physics degree you will have lots of employment opportunities... but not in Physics!"

Yeah, to be fair, my perspective is from Canada, which I probably should have mentioned.

I mean, we do have the RCMP and Bay Street, but I don't think it's comparable (in number of positions, salaries offered, etc) to NSA and Wall Street.

I heard that if you studied math with applications to computer science, you can find great opportunities with technology too (if not moving to the US for places like Google, there are quantum computing startups like Xanadu in Canada, along with other technology startups in Canada’s big cities too).

In addition, outside of the RCMP, various ministries of the Canadian government could need the skillset too (Statistics Canada being the one that first comes to mind, though any teams that do economic analyses could benefit from someone with the background).

GS pay isn't that competitive

The NSA doesn't pay that much compared to the opportunity cost of the wage lost getting that PhD. Their salaries are capped by government standards, they can't pay above that.

There are a fair number of phd’s employed in biotech/pharma. Who do you think makes the stuff? Bs/ms are their techs. The ads in the states publishes salary history every year for those who care. It’s not SV salary but it’s way more useful than selling ads and scarfing data - you can actually end up curing a cancer or other diseases…

And before you go about just curing symptoms, remember that many real cures would require gene modification and people aren’t all that interested in being a gmo. Ask RA sufferers about drugs like Humera re curing symptoms.

Definitely not a satire but the hard truth. Your kid might end up having a strong existential crisis about why pursuing a "science" field ends up with 80k salary jobs when with that effort you could just go to FAANG as a SWE with a much higher salary with a bachelor's degree. If you don't get there as a bachelor's, just go there with a master's.

The "strategy" is to get your kid interested in science enough so that they make a strong resume to a good college and have a strong foundation, then sway them to learn CS and get a SWE job.

Because it's interesting.

This is the sort of comment that makes me really take a hard look at capitalism. The way I see it, scientific work is rewarding in and of itself. I don’t _need_ to make huge amounts of money if I’m doing something intellectually rewarding that contributes to our collective knowledge. Sure, add monetary compensation to your list of optimising constraints. But to make it the SOLE optimising constraint in your life? That’s tragic, IMO. Make enough. Do interesting things. Cultivate rewarding relationships. Be happy.

PhDs are massively taken advantage of. The whole "do it for passion" thing is literally used as argument to abuse them.

Little money, long hours, effectively up or out pyramid where most people gonl out, expectation that you will move in few years and partner will follow.

I suspect interacting with the real physical world and its realities and to realize one can affect it would be good, no matter what career they'd pick later. Picking a career in software development has been a good choice for bright kids for several decades now. In the long term view, the past is full of "good career choices for bright kids" that at some point no longer were not.

The physical world is not going anywhere.

Its probably more reliable income though. Screw up in tech (not even big tech), or have some kind of life misfortune, and your salary can pretty quickly go to zero. Knowing how to reverse a tree on a whiteboard is not some kind of important skill that employers want.

I had the same question, but my response was: "at the end of the day, authoritarian politicians will order you what to think, and their media/technology lackeys will enforce those orders".

I think your assessment is not based on the current labor market. I am aware of scientists with a fresh B.S. making that much. I do agree that there is little economic value in most Ph.D. programs.

an 80k starting salary is hardly grim and if you happen to become a decent scientist or technician along the way there is usually work to be found, but yes, the phd mill incentive matrix is a blight and demonstrates broader misalignments in microcosm

This is a pretty techy group. Out of curiosity what do you do and what was your major?

This is a very much valid point. Science is amazingly cool, but gives you a pretty shit life.

I'll encourage my kids towards science as much as needed and let their develop their thinking while they're at it, and congratulate them for the enlightenment when they come tell me that working as a scientist sucks. It's still really good for them to pursue that career early, if it's what they enjoy.

Reply on news.ycombinator.com