If you want to be paid well in the United States, conventional wisdom suggests that going to college is a necessity. But Wharton management professor Peter Cappelli, director of Wharton’s Center for Human Resources, says it is not such a straight-forward decision. In his new book, Will College Pay Off? A Guide to the Most Important Financial Decision You Will Ever Make, Cappelli demystifies common beliefs about when — and whether — it makes sense to go to college, what to major in, and more. Knowledge at Wharton recently spoke with Cappelli about this life-changing decision.

An edited transcript of the conversation follows.

Knowledge at Wharton: Before we talk about the book itself, I wonder if I could ask you a quick question about the title. Normally when people talk about the most important financial decision of their lives, I’ve heard them talking about buying a house. Why is college number one in your mind?

Peter Cappelli: For many families, they are making this decision more than once, so their home could be certainly more expensive than what they spend on college. But if you’ve got a few kids, it begins to mount up. It’s probably fair to say that my view on this is it’s one of the most important decisions you could make because there’s a lot of risk involved. The information is not very good. The outcomes could really vary much more so even than most housing markets — which except for the recent past, have not been quite so bumpy.

Knowledge at Wharton: I want to quote something that you write in the book, which is that you wrote this book because of the proliferation of “unqualified statements about the big payoff to a college degree that are pushing so many students and their families who can’t afford to do so to jump into the deep end of college expenses, taking on debt that they cannot afford for experiences that are unlikely to pay off.” What is your basic counter-argument to these claims?

Cappelli: To rephrase it a little bit, the concern for many families is not whether your kids are going to go to college; it’s a bit of an entitlement in the middle class and upper middle class. For those folks, we can offer some advice about how to think about the economics associated with it. But increasingly over time, a much larger percentage of the U.S. population is going to college. For a lot of those folks, it really is a challenging decision to make.

“One of the things I worry about with the very specific, job-focused majors, is: What do you do if you don’t get a job in that major?”

The alternative, counter-argument to “everybody should go” is to say… are you offering to pay for the education of all these kids? You’re pushing to them go; how are they going to pay for it? Then it becomes quite a different question. If it’s free, of course you should go. It will be a wonderful experience for most people. But if you have to pay for it, then it’s quite different. The way we pay for college has changed a lot over time. Loans are up 150% over the last ten years. The typical costs of college have risen by about four times the rate of inflation over the last generation. So it’s much more expensive in a period where incomes are more or less flat. The kind of advice that many of us grew up with — college is a great thing, you all should go, it would be wonderful — might not be completely right in the current environment given the costs and given the realities of the job market….

Knowledge at Wharton: Coming to the job market, what are some of the most common myths that you found in your research about the link between college degree and well-paying jobs? Because that’s the normal argument you always hear: If you want to get a well-paying job, you need a college degree because the wage differential between those who have a college education and those who don’t is so high.

Cappelli: That is the one piece of information that you hear over and over. The average college graduate in the U.S. does make 60% more than the average high school graduate. Of course, the average worker in the U.S. economy is about 46 or 47, so they have been out of school for 20 years or so. It’s a kind of historical effect. That premium, that wage difference, is something that is bounced around a lot. In the 1960s and 1970s, it was almost zero, believe it or not. College graduates didn’t make more than high school graduates. There’s nothing that guarantees that it’s going to be at a particular level. It is now at a record all-time high. If you’re thinking this will persist, it bounces around a lot. That’s the first thing to think about.

The second thing though is that it’s a mistake to assume, as a lot of people do, that the college graduates are identical to the high school graduates except for the college degree. [What is left] unspoken after that 60% figure is, therefore, you should go to college because you will make a ton more money. The kids who go to college are systemically different than the kids who don’t go to college. They have a lot more things going for them typically: more supportive parents, more family wealth, all kinds of differences in maybe their abilities and motivation. They are different even before they go to college, and they are quite likely to do better even if they didn’t go to college than the average high school kid would do. The assumption that a high school graduate would do as well as the average college graduate is a big, big leap. It requires assumptions that we know are just not true.

Knowledge at Wharton: To make sure I understood you correctly, is it correct to say that if people who have gone to college make more money than those who didn’t, there are factors that existed even before they went to college that are responsible for it rather than going to college itself?

Cappelli: They are responsible for some of it, for sure. This is a really interesting problem…. Why do college graduates make more money than high school graduates? Some of it seems obvious. You can take on jobs that maybe require a college degree — that seems sensible; it’s just a requirement. But we know there are other differences as well. One is this big difference that the kids who can make it to college and complete college, which is one of the biggest issues. Most kids actually who go to college don’t seem to get through college. So getting to college — you’ve got to be different, you’ve got to have more support just to get to college. To finish it suggests a lot of achievement….

The kids who come to college are different than the kids who go to high school. That makes a big difference. The other thing is that the kids who finish college have signaled something to potential employers about their abilities that is valuable. We know this pretty clearly from the high school level. For example, in high schools, we have a graduate equivalency diploma, the GED, and then you have regular high school degrees. Most people think that the test to get a GED degree is reasonably hard, or at least would be hard for the average kid who is not going on to college, particularly the lower part of the distribution. But people who get GED degrees make nowhere near as much money, and do much worse, than kids who have high school diplomas. Why is that? It’s because the employers see the high school diploma and they know that at least you have been able to sit in a classroom and get along with people and persist enough to finish. With a GED degree, you know that that’s probably not true. It’s not the cognitive ability or the classroom knowledge that seems to matter so much.

When we look at surveys of employers who are hiring college graduates, what they say is most important to them has nothing to do with the academic experience. It has to do with experiences that are work-like on the college campus. It has to do with internships and summer jobs and extra-curricular activities. That’s what they care more about than the academic material. So we don’t know for sure exactly why kids who graduate from college do better. But I think the point for us is you shouldn’t assume that the average kid who doesn’t go to college will just do as well as the average kid who made it through college.

Knowledge at Wharton: Your book also notes that schools, especially the for-profit ones, in their efforts to show that a college degree leads to a good job, very often end up offering courses that almost sound, as you say, like job titles. Did your research show any direct link between a four-year college degree and very direct job related training? Does that actually work?

Cappelli: Do they do better if you have got this very specific job training? Well, no one has actually looked at that carefully. It turns out that it’s quite difficult to get some of that data. Here’s an unfortunate thing for parents. If you’re trying to figure out whether a particular college will get your kid a job, there’s no standard data to look at. Colleges are not required to report that information. When they report it, they report it in different ways. They’re not always very honest in the ways they report the data. Somebody could be employed as a barista and maybe they count them as employed. They sort of are, right? So it’s pretty difficult to know whether or not particular colleges are actually getting your kid a job. You probably can’t trust the marketing information, which for the most part doesn’t give you data anyway. They just kind of hint that this is the career and here’s the education that will help you with a career.

One of the things I worry about with the very specific, job-focused majors, is what do you do if you don’t get a job in that major? You’re an international tourism major — which is a real major, by the way — and the international tourist agencies are not hiring the year you graduate. What are you going to do? You would have been better off with a liberal arts degree. You’re taking a huge risk when you pursue these very focused degrees. It’s not free in the sense that there are a bunch of other courses that you could have been taking and a bunch of other things that you could have been learning beyond, say, memorizing international tourism laws and taxation relationships and things like that. There’s an opportunity that you’re losing that might have broadened you in ways that would allow you to do better after just your first job, which is the other point about these programs. They are just focusing on the first job out of school. What happens after that?

“The hottest job in the U.S. for the last three years … has been petroleum engineers. A generation ago, or even ten years ago, those folks were waiting tables.”

Knowledge at Wharton: That’s very interesting. When it comes to trying to decide what people should study, STEM — or so called Science, Technology, Engineering, Math — programs seem to be all the rage these days. And it’s a very common belief that if you want to move into high-paying jobs those are the kind of courses to focus on. Did your research show if that’s true?

Cappelli: If you look at the evidence on STEM, the first thing to note about it is it’s a bunch of different majors that actually don’t have an awful lot to do with each other. What the evidence seems to show is that people with math degrees and science degrees don’t do very well. There’s an interesting study in Texas that shows that students with sociology degrees make more money than students with biology degrees. It’s not true that in the sciences or in math there are even very many jobs for these folks, and they don’t pay very well.

It is true that if you look for the best paying jobs in the U.S. right out of college, those jobs are almost entirely engineering jobs. But the thing about them is they are not the same job every year. And being an engineer is not a general qualification right? If you’re a petroleum engineer, you can’t switch and become an electronics engineer or computer science engineer at the point of graduation. They are quite different and they are not substitutable across.

The thing about these very top jobs is that they change a lot year by year based on what’s going on in the economy. To give you an illustration of this, the hottest job in the U.S. for the last three years or so — and which pays about 50% more than the second highest paid job — has been petroleum engineers. A generation ago, or even ten years ago, those folks were waiting tables. There was no work for them. What happened was completely unpredictable. That was the discovery of fracking, which suddenly unleashed a bunch of exploration, and these folks were just really in demand.

But the thing that we know is, as soon as something like that really becomes hot, students pour into those majors and fields. That’s what happened in the last couple years. They are pouring in, and now oil prices have collapsed. The interest in oil exploration is also beginning to collapse and will shortly. The demand for these folks is clearly going to collapse as well because demand is falling and the supply of new graduates is going to pile into the market.

One of the things about pursuing these jobs which are at the top of the list is that you’re taking a lot of risk with those jobs because they may not be hot. It’s not all engineering jobs that are always hot. The second thing to think about with a lot of those jobs, particularly the IT jobs, is they don’t seem to last very long. People don’t stay in those jobs very long, so you get a great-paying job out of college; three years later, what are you going to do? You can do another programming job, and you can be a programmer forever. But they don’t lead anywhere long term. A lot of people give up on them altogether.

Knowledge at Wharton: What would be a rational choice for a student to make who’s considering a career choice? Would it be to ignore where the demand is at the present moment and just focus on their own interests? Or should they look at other factors as well?

Cappelli: We way overplay the job market in terms of pushing kids in particular directions, partly because we can’t predict it very well. If you think about it, a kid going to school now, applying for college this year, won’t graduate from college for five years, if they’re lucky. Here’s the shocking statistic for parents: only 40% of full-time students in four-year colleges graduate on time. Only 60% graduate in six years. I know people are saying, “Well not my kid.” Well, sometimes you can’t control that stuff. Neither of my kids graduated in four years. It had nothing to do with the fact that I wasn’t telling them to graduate in four years. So, it’s a long time away, and the job market is surely going to be different by the time they get out. That’s the first thing to consider.

The second is kids don’t do very well in fields where they don’t feel they’re very good at them. If you have somebody who really wants to be a history major and you’re leaning on them to be an engineer, it’s a very good chance they’re not going to be a very good engineer even if they get through the program. We shouldn’t focus too much on it, particularly at the entry level. You should try to go to a school where you can change your mind. Kids change majors all the time. The second part of that is you should go to a place where it is not only easy to change majors, but the majors are not that complicated. One of the main factors that cause kids to delay graduation, and one that is completely preventable, is that it is difficult to get the courses required to complete your major, unless you take them absolutely in lock step. If you move into a major, you’re suddenly in some trouble.

Here are things I would look out for, too, on a short list. If you’re applying to a small college and they offer, as many small colleges do, 70 majors or something like that, you have to ask yourself what’s their ability to actually deliver these majors. Typically what they are doing is bringing in a lot of adjunct professors, people from outside, practitioners sometimes, to teach those as well, which should give you a little pause. What are you paying for if it’s not regular faculty?…

“Here’s the shocking statistic for parents: Only 40% of full-time students in four-year colleges graduate on time. Only 60% graduate in six years.”

The second thing to look at is the actual requirements of the major. Of course a major that has many requirements is a little problematic. But the biggest problem is if the major requires that you take courses in a particular sequence. The reason that is a problem is if the professor who teaches this course is on sabbatical that year, maybe you can’t get it until the spring, in which case you’re already behind, and you can’t take the other courses.

Majors are quite complicated. It is completely preventable. We have basically designed them with the faculty in mind rather than the students. This is also an issue where public and private differ a lot. The state universities, which have really been squeezed by budget constraints the last few years, are having more trouble getting kids to complete majors because they’re having trouble getting faculty lines, filling positions. It’s an issue when you’re thinking about which college to go to.

Knowledge at Wharton: I sometimes feel, looking at the educational options and the high cost of college, that this is almost like the golden age for the autodidact where anyone who wants to create their own bundle of education experiences, say from the Ivy League institutions, can just sign up for MOOCs from Coursera or edX or Udacity, and get credentials of some sort from Ivy League universities in courses that they want, and create their own personal educational portfolio. Since many MOOCs are offered for free, they could probably also gain these credentials without piling up huge debt. Do you think this could be a viable option for people who might either be thinking of an alternative to conventional education or someone who just can’t afford it?

Cappelli: I think you would get something very different with a MOOC. As you know, they are actually not doing that well. The initial bloom is off the rose. Hundreds of thousands of kids sign up for these, or adults do, and four complete them. It’s something that actually was always available in the form of a textbook. Completely portable. You can take it anywhere. You had the best ideas of the faculty laid out in a sequence. You could stop and repeat things. It’s just as [good], or maybe better, than a MOOC. I’ve never heard of anybody who learned organic chemistry in a dark room by themselves with a textbook, right? The thing that universities provide is a context that forces people to learn. And MOOCs don’t have that. Access to the information has never really been the issue.

The other part of that is it’s not clear employers care very much about that. So having a MOOC, being able to pass a test like the GED degree at a high school, they don’t seem to care that much about that. What they care about is the college experience. In the U.S., the more traditional college experience where you are working in a community, you’re getting along with other people, you’re learning from the context as well as just the academic material. That’s what they seem to care about. And MOOCs don’t give you that.

Knowledge at Wharton: If you were to look at the conventional U.S. schools, how do you think they compare to international schools in preparing students for jobs?

Cappelli: They are very different in the way they compare. The prototypical U.S. college experience, at least the way we thought about it, was four years on a leafy campus someplace where you’re walking in the quad and bump into your professor. That’s almost never been the case in colleges elsewhere. Much more common, particularly in Europe, is you are living at home and going to the local university during the day. You’re riding the bus. You’re not walking across the quad. You’re in a big lecture hall. You’re taking giant, standardized exams with other people. It’s a little more MOOC-like than the U.S. experience.

The U.S. students are getting something quite different on the traditional U.S. campus than you’re getting abroad. Employers seem to like that. You have to talk more. You have to get along with people. Those are life skills that seem to matter. If you talk to people in other countries, which I do quite frequently about these issues, they’re complaining about their own campuses, their own education, not just the college, but high school as well, that they don’t do the kinds of things that we do in the U.S.

On the other hand, in the U.S., we pay a heck of a lot for it. The average American parent pays seven times as much for a college education as the average parent in the rest of the industrial world. We’re paying an awful lot for it. We’ve got the second worst graduation rate, too. Not that many people are getting through this college program in the U.S., which is something we might want to think a little bit about.

Knowledge at Wharton: In answer to the question does college pay off, you say in the book it depends. What does it depend on?

Cappelli: It depends on the cost side. Of course, that’s just how much you have to pay. That’s where you have to be quite careful. But I think the focus that we’ve heard a lot about, which is about trying to game the system for financial aid and things like that, is really kind of a waste of time. For one thing, it’s quite likely going to be illegal if you’re trying to game the system in order to get more financial aid. That’s probably not the smart thing to do. But paying attention to simple things like the graduation rate: Will my kid get out on time? There are very big differences between colleges with respect to that. There’s a big state university/private university difference of about ten percentage points in the typical graduation rate between state and private institutions.

“The single most important fact that I would look at is the graduation rates across colleges…. It varies like crazy.”

For example, in the State of Pennsylvania … Penn State’s graduation rate is 65%. At the University of Pennsylvania — we think the most elite private school in the state — the graduation rate is 98%. If you’re thinking about probabilities, is it worth it to pay more to get the kind of support and systems and major design and everything that gets kids out in four years? Yeah, probably, it absolutely is. The first thing to think about are things like will my kid graduate on time? Financial aid matters a lot. State universities, private universities — now, the average difference in tuition between private and state for kids getting financial aid is only $9,000. Public institution tuition has gone up 50% faster than private. Financial aid in lots of places has gone up pretty high. Thinking that public institutions will be cheaper is not necessarily true.

The cost side matters a lot. Graduation rates matter a lot. Being able to get the first job matters, but I think getting the first job may not be quite so dependent on your major as we used to think. We know employers care a lot about internships more than they seem to care about the college curriculum that you’re taking, work experience on campus, and maybe some of these certificates that you can get online in fields where those count, like computer science. Maybe that matters.

For example, a kid who can delay picking their major until late in the program and get a sense of where the job market’s going to be when they graduate can take very practical courses that seem to be hot at the last minute. The year before they graduate they take big data classes in Python software and things like that, and get some work experience outside. They might do just as well as somebody who actually had a major in one of those fields. They could have a philosophy major, and they might do just as well.

Knowledge at Wharton: Interesting. One last question, if you are the parent of a child that is about to start looking for college, what’s the most important piece of advice that you can offer them?

Cappelli: The single most important fact that I would look at is the graduation rates across colleges. They have to report that. That information is pretty reliable. It varies like crazy. I think the single most important thing to think about is maybe on the parents’ side a little bit, and that is maybe to not worry quite so much about the job market side. One of the things we know about kids is they don’t do particularly well in colleges where they don’t want to be, so trying to force them into a particular major — they may not be good at it anyway — probably isn’t the best thing to do. You can become preoccupied with the job market early on in your college career in ways that distort what you’re doing.

One of the interesting statistics about college students in the U.S. now is that the amount of time spent studying seems to have declined quite sharply in the last generation. What has gone up, fortunately, is not beer pong time, but time spent in extracurricular activities. Maybe that’s useful. But it’s worth thinking about that as well, depending on the colleges you want to go to. What is the culture of the place like? Is this a place where kids are serious about studying and academics? Is it a place where they are spending all their time early on just worrying about the job market?