The Real Cost of a STEM PhD?

My son just finished his first year of undergrad as a physics and math major. His goal is to graduate and apply directly to PhD programs in high energy quantum physics. I know that in the US physics PhD students will have their tuition and expenses waived and often be compensated to be a TA or RA. That compensation may or may not be enough to pay for housing and food depending on the location of the school.

My sense is that the real cost is more than the remuneration you’re getting from the university. A PhD student may want to attend conferences, go to national labs, may have additional expenses, etc. The question is, what is that cost?

I’d like to put some more money into his 529 plan to help with those expenses if possible but I don’t know how much to really plan for over the course of his 5-6 years as a PhD.


It is customary for the institution to pay for conference travel, at the one per year level. It would be unusual for a student to self-fund this.

The real costs are living expenses and, of course, opportunity costs.


It is very much dependent upon, as you mentioned, the location of the program and the amount given to TA’s. Daughter’s boyfriend graduated from a program at UNC-CH last year. He shared the rent on a house with 2 other classmates and worked as a TA. He also did some work for a local pharmaceutical company. He must have done well as he purchased 1/2 of a duplex last summer.

1 Like

An offer for a STEM PhD program that is not enough to cover a modest, but entirely reasonable, standard of living (including conference trips) is a big red flag- and he shouldn’t take it. In some circles that is considered a polite rejection.

One of the gradschoolkids (also a physics and math UG) is in a PhD program, and is- this very minute! on a plane en route to Hawaii, to deliver a paper at the big-deal conference in her field. Like every conference trip she has made, it is fully funded by her department. In first year she shared a house w/ 3 other students, and socked money away like a mad thing. In second year she & her bf (a math PhD candidate at the same U) bought a teeny-tiny house together, with no help from either set of parents. That won’t be possible everywhere (housing costs obviously vary by location), but the point is that their stipends are sufficient to fully support them. They do not have a lavish life (and the first year they had a third roommate to help with the costs of getting the house setup) but it’s not ramen noodles and freezing in the garret either.


To your point about the opportunity cost, it seems that 5-6 years of lost retirement savings (with all the years of compounded interest) would be a huge one. After doing some more reading, it would seem like that if their TA or RA funding isn’t specifically earmarked to be used for room and board expenses, it would make sense for me as the parent to pay for R&B with 529 funds and for my son to spin up a retirement savings account and sock away as much of the money their paying him as he can.


Blockquote An offer for a STEM PhD program that is not enough to cover a modest, but entirely reasonable, standard of living (including conference trips) is a big red flag- and he shouldn’t take it. In some circles that is considered a polite rejection.

At first blush I would totally agree with you but I had a very interesting discussion with my son over winter break that resulted in us looking at published data on application, admission and ultimately offer acceptance rates at several tier 1 research schools for physics PhD programs. It’s pretty clear that some schools just offer better compensation packages to their grad students than others even when discounting the cost of living. I’m not suggesting that pay should be the deciding factor but it’s certainly an input.

Anyway, it’s good to hear they won’t be eating ramen and keeping the heat off in the winter to survive, either. I don’t expect that he’ll be staying in a five star hotel and eating a Michelin Star restaurants but I do think knowing you have a nice place to stay and not worrying about skipping a meal until your next paycheck is helpful as well.

1 Like

This is absolutely true- and there are often sweeteners if they really want you (eg, no requirement for TA/RA in year 1, named awards that bump up the stipend, etc.- the gradschoolkid did some negotiating based on her other offers). But, an offer that would not support a student without outside employment (or that requires an unrealistic number of teaching hours, as happened with one of the bf’s offers*) is a red flag.

*interestingly, that program was much lower ranked than the program he ultimately accepted, so it’s not a straight line between rankings & offers either

1 Like

I’ve heard and read about such “sweeteners.” At this point in the game all I’m trying to do is take advantage of the fact that I have about 36 months to throw some additional $ in the 529 but if it’s truly not needed then I’ll add even more $ to his younger brother’s 529 since he seems to be more interested in going out of state.

I just don’t like the idea of sending my older son off to PhD program with his 529 depleted from undergrad and his only source of sustenance is a meager TA or RA-ship while also knowing that some graduate programs put rules in place which prevent grad students from working more than 20 hours a week.

Again, this isn’t about making sure my son is living life high on the hog or somehow cushioning him from “the real world.” I just figure there are clearly some better tax-advantaged ways to go about helping with grad school than others from a parental perspective.

You or your son should take a look at the PhD stipend survey:


For most programs, you don’t want him working more than 20 hours/week! Really, it should not be necessary.

But, I get your overall point. And I thought your ideas of amping up the 529 with an eye towards room & board or funding an IRA/Roth were fine.

But: your son is at least 3 years away from grad school- and the road may twist in other directions between now and then (it happens). As you can transfer a 529 balance to another child, or hold it as an investment for a future grandchild, why not just keep funding the 529, and see what kid needs what?

1 Like

My son is a 4th-year aerospace engineering student at Michigan. I cover his car insurance and his phone is on our family plan. His stipend has been ample, including fully-paid health insurance and tuition reimbursement, with no TA requirement. He saves enough to fly home twice a year and has a social life. He is going to a conference in Boston this summer on Michigan’s dime. He is adding a PhD in Scientific Computing designation, which has required more classes than most PhD students, but all have been completely covered.

The big issue is the opportunity cost of missing five years employed as an engineer. I would put that at $250,000 or so. But he should do well financially as a PhD (no interest in academia) and loves what he’s doing.


To be transparent, older son’s 529 is fully funded for four years of in state undergrad (even though he’s somehow mapped out a plan to graduate in 7 semesters with a double major and plenty of research experience). As a result, I’d took my foot off the 529 gas.

When my son was 15, I took him to visit Fermi Lab, a Dept of Energy lab near Chicago. The whole time his eyes were as wide as saucers and he walked out saying, “This is where I want to work for the rest of my life.” I periodically touch base with him and ask if he’s still loving all the physics and math because, to your point, majors can and often do change, but his course has been straight as an arrow. Thankfully, for him anyway, his math and physics abilities have matched his professional aspirations.

The younger son’s 529 is nearly funded unless he chooses a crazy expensive private school and then we’d need to have a talk. So luckily, this isn’t really about a tradeoff but more about how we might be able to help our “kids” out as they get their start in lives.

1 Like

We had a moment like that with the PhD kid also :slight_smile: It’s magic

1 Like

While I was the one who mentioned opportunity costs - and they are real and should be considered - if you are getting a PhD for the money, you should probably reconsider.

:“I came to Casablanca for the waters.”
“The waters? What waters? We are in the desert.”
“I was misinformed.”


I agree with you that OP should consider opportunity cost, which is THE real cost of a STEM PhD. However, OP and his son don’t appear to be in it for the money (or even weighing “ROI” as some on CC apparently do).

I remember three thoughts flashed through my mind when he said that:

  1. “You could do a lot worse, kid!”
  2. “I sure as heck wish I knew with that sort of certainty and your ability, what I wanted to do for the rest of my life at your age!”
  3. “So this is where my wife and I will be driving to to visit our son (and perhaps future wife and grandkids?!? I’m cool with that!”
1 Like

The son who wants a physics PhD has always been my saver and frugal one. As he’s told me many times, “I could probably make a lot more just getting a masters and then going into industry than doing a PhD and working at a DoE lab or in academia.” He’s in it for the challenge of solving the very hardest problems, not the money.


To me that is the right reason to get a PhD. Basically “I want to do it” is the reason to do it. Not everything in life needs to be focused on maximizing the amount we leave to our children. Of course, if economic value was the primary goal, we might not have any children to leave money to.

However, the initial question was how much to put into a 529 to help with PhD expenses. I have been leaning towards intentionally running out of 529 funds before our children are done with graduate school so that we do not need to figure out what to do with the left over money. My understanding is that for a fully funded PhD the “extra money” that you need will be limited, and if a parent wants to help we can just use our regular bank account.

Thus at least my wife and I are planning to have the 529 run out before either daughter can be properly addressed as “Doctor”.


This could be the correct answer…that there really is no reason to have any substantial money set aside for a fully funded PhD because it really is “fully funded.” I have two graduate degrees - one obtained in England where I paid for it 100% and the other paid for by an employer while I worked for them. Thus, I don’t have any direct experience with a fully funded PhD program though I have a healthy level of skepticism about something sounding too good to be true as in PhD AND fully funded…hence the question.

I have no idea what your personal financial situation is but I heard many years ago that for those people who max out their employer sponsored savings programs, individual IRAs, etc., that putting money into a 529 and then pulling it out in retirement could be a viable tax strategy. Whether that’s still sound advice, I have no idea. To me, any money in a 529 is a sunk cost so the idea that some leftover money sits in there for decades, compounds and then some future grandkids are the beneficiaries is kind of appealing. But as they say, “You’re getting ahead or yourself.”

There may be conference trips that are only partially funded by the professors’ grants. Perhaps everyone on the project goes to one or two but beyond that there might not be as much funding depending on the lab. It’s a nice idea to have some money in the 529 for this. 529s also pay for computers, etc and internet service.