July 7, 2016
The following is an interview of Dr. Anthony Carnevale, author of the report “America’s Divided Recovery: College Haves and Have-Nots.” The MacIver Institute previously detailed some of the report’s key findings.
Could you briefly summarize your findings and what you think is the most consequential part of your report?
Dr. Carnevale: What the study says in fairly straight forward terms, I think, is that the continued trend that we see — people usually date it in 1983 after the 80-81 recession — is the increasing dominance of the college educated in the workforce. Their dominant position in labor markets, it’s still there, and it looks even as if it’s gotten stronger after the recession. That is somewhat unusual. Generally in recessions the kinds of change that result in shifting skill requirements on the job are not that strong. Recession tends to slow down investment and structural change. A lot of institutions tend to get frozen, or are in decline, or are very tight with their money, so you don’t get much investment or change in the structure of companies, industries, etc. But in the last three recessions, it’s pretty clear that the shift towards a post secondary educated workforce has become stronger during the recession, and that between recessions that same trend continues.
A previous study adds another element to that, which is that not only are people with post-secondary education getting most of the jobs, but that they are getting most of the good jobs, in the sense that they are getting the jobs that are paying more than $50,000 a year. 70% of the jobs that post secondary folks get tend to have healthcare coverage, and maybe 30% have some sort of retirement coverage.
And there is another fact about the recession. One of the factors that’s important, at least to economists like me, is that there is something of a debate about the recession. Obviously this was very difficult recession, and we saw a fall off in all jobs, including college jobs. We had an employment rate among recent college graduates of almost 10%. But economists have argued that throughout the recession, even though things were bad for recent college graduates, they were worse for people with only high school degrees. In a sense, the structure of the economy and its preference for college workers held up that thesis. And then in the recovery the job growth confirms that. That is, once the economy started to grow again, the new jobs that were created were disproportionately favoring people with some college — that is a two year associate’s degree, a certificate, or a few years of college with no degree — and decisively favoring people with bachelor’s degrees and graduate degrees.
These results are an affirmation of a trend that has been widely studied and reported since the ’80s, in which technology change increasingly substitutes for any kind of repetitive task in a job and leaves the non-repetitive tasks to people. And those non-repetitive tasks require more skill. That’s really the simplest explanation [for the job loss in areas requiring a high school degree or less] — there are more complicated explanations than that, but the technology definition is the most parsimonious.
And the other headline in this report, I think, is that people with bachelor’s degrees and graduate degrees, or jobs that require bachelor’s degrees or graduate degrees, now outnumbered jobs that require high school degrees. The top is bigger than the bottom in the skill hierarchy.
Keep in mind that in the 1970s, before all this change started, all the way up through 1983, 70% of American workers had a high school education or less. That was a different economy in which a high school education was enough. And most of the people with high school degrees were solid middle class citizens. But that has been completely turned around, so that those are the people that have been left behind in the recovery. We see evidence of that in our politics, and the anger in our politics. The people who are getting ahead are the people who have moved on into post-secondary education.
The unemployment rate has improved since the bottom of the Great Recession, yet most of those jobs requiring a high school degree or less haven’t come back. Is there evidence that the cohort of lower educated people who lost the most jobs went to get college education, and that’s one of the reasons why the rate has improved?
Dr. Carnevale: It is the case that, in this recession and before this recession actually, this [the decline of jobs for people with a high school degree or less] has been going on for a while, I would argue. You could trace it back to the 2000-2001 decline, as well as the late 1990s. After a very robust middle ’90s, the economy hasn’t performed that well ever since. One of the problems is that if you compare the data from 1996-1997, everything looks like it has slowed down. One view is that the economy was overheated. We had all the funny money from Wall street floating around the economy, there was a lot of construction and buying of manufactured goods, the economy was simply running very hot. So you had relatively high earnings returns, and especially to people with more education. You have relatively high employment rates; the male employment prospects in the middle ’90s are really quite good. But it turned out that, when we get back to a more steady state growth rate, that the male economy — that is in manufacturing and blue collar jobs — that there was something of an artificial boost in the mid-90s. So males were hit especially hard in this recession.
There is a relationship between college-going and economic cycles. It is commonly understood that when a recession starts, people go to college. It’s quite a sensible decision if you do it right. If the jobs aren’t there, the smart thing to do is go get more human capital, go get more skill, wait it out. Both because you wait for the job market to come back, and it gives you a better start…
But the other reason is that you put yourself in another labor market, you put yourself on a different path once you get more education. What we always see in recessions, and particularly in this one since it was so deep, is that people stop working and start going to school. We saw people who went into college beginning in 2008 — because 2007 was kind of the transition year — the year the recession began. And then college enrollment went way up, and then it started coming down again. I’m pretty sure it’s in ’12 and ’13 that it starts coming down. It’s been going down since, and probably will continue to go down for a while. This is not unexpected.
You mentioned politics earlier. Will this report have any effect on the political debate about free trade? The argument made by some politicians that we need to be more protectionist to bring these jobs back?
Dr. Carnevale: I think this report and a lot of the stuff we do that tracks the economy, tends to support the narrative, or the story, that is, provides empirical support, for what lies beneath much of the current political anger. The pushback on things like trade, and that is this process which began in the early ’80s and onward where college degrees are more and more required, part of that story is that there was a shift in the jobs structure in the United States. Huge productivity increases occurred in manufacturing, transportation, utilities, etc. Basically, it reduced the number of workers required. That’s been the major factor in this. Probably 70% of the jobs we have lost in manufacturing and other blue collar industries have been due to technologies. In modern parlance, it’s about robots. About 30% is due to foreigners, that is, trade. So trade is not insignificant, but it is not the dominant cause. This report, and probably a lot of what everyone has been doing looking at the economy, it does explain this, but it doesn’t necessarily tell you what should be done about it…
It’s pretty clear that trade benefits a society on the whole. But one of the things that is also clear in the data and has been for a long time is that the people who benefit from trade are rarely the same people who are hurt. Trade has a very specific impact. It helps a lot of people, and harms very few, but the few it harms, it harms greatly. The people it harms tend to come from particular parts of the country, from particular communities. They tend to be high school educated workers. There is a very concentrated impact, it’s very disruptive, in certain communities, especially smaller communities driven by manufacturing, single manufacturing plants and that sort of thing.
Again, this report helps explain the economic trend. It doesn’t tell you much what to do about it.
At the end of the report, you say, “The nation must face up to a need to train more of its workers for the growing high-skill jobs that play an increasingly central role in the post-Great Recession economy.” What kinds of policies would accomplish this goal for you?
Dr. Carnevale: Well, there are a couple of things. There are things that cost money, there are things that cost less money. One is that now that we are a nation with a post secondary education system that runs at about $4-5 billion a year — that’s a lot of money, and a lot of it is public money. Even private colleges have not-for-profit status, which derives from special consideration that allows those institutions to be not-for-profit because they work in the public interest. That’s what not-for-profit status derives from, from that determination, that legal determination. More and more, either through tax credits or tax deductions for more affluent families, or direct student aid through Pell grants, the whole system of federal loans is now become a major part of paying for the cost of college. So it’s an institution where there is a lot of public money.
There is an argument, then, that it ought to serve public purposes. And one of those purposes is increasingly pretty clearly that we are using our higher education system to develop our workforce. What we are suggesting is not that we just spend more money on people going to college — although I think in the end it costs money, although the alternative argument is that it can be done a lot more efficiently — but also that the colleges themselves, which are now our major workforce development institutions, that colleges be more mindful of the impact of their curriculums on their students’ careers.
I don’t mean that everyone has to go to college and become and engineer, but whatever the major or program offered by the college, the college could take more of an interest in what happens to their graduates after they complete their degrees. Does the education help them 1) participate more fully in their own lives? Does it make them more robust people? That is a primary concern. But 2) it is also a concern that the kind of education they get helps them leverage some kind of job. Because in the end, you can develop people in very robust ways, but if they end up jobless, the notion that they will continue to develop is, I think, you can be very pessimistic about that. You can know a great deal, but you are not going to be a lifelong learner unless you become a lifelong earner in this society. You can be developed fully, but you’re going to develop much less if you are living under a bridge.
The employment power of college is something that I think colleges have to take some responsibility for. Especially given the fact that they are more and more expensive, and that more and more is being paid, in one way or another, by the community — that is, the nation, or in the states, or the communities of students and parents who are paying private cost to go to college.
One approach we used in Wisconsin to help boost workforce participation among those currently without jobs is to start workforce training programs. Can these accomplish the same things as a college education, or are they insufficient?
Dr. Carnevale: It all depends. What is fairly optimistic about that kind of strategy is that two things have happened with education training since the ’80s and been developed over long term. One, the value of education beyond high school, big and small bites of education, has grown enormously. The college wage premium, which in the ’70s was below 40%, rose to 84% and has essentially stayed there. We know that going to college, getting skills, small bites or big bites, tends to be a good thing.
But the other thing that has happened that in some ways is even more powerful is that it all depends on what you study. What you take is probably the most important factor in what you make. There is a huge variation in the value of college degrees. If you become a petroleum engineer, with a bachelor’s degree, you’ll make almost $130,000 a year, although those wages are going down now from trouble in the oil fields. But if you get a bachelor’s degree in early childhood education, you’re going to make more like $30,000 a year. There is almost a five fold difference between those programs, but they are the same bachelor-titled degrees.
It’s also true that 30% of two-year degrees — not so much the associated arts degrees but the associated science degrees, the associated applied sciences, those are more career specific degrees — and probably the certificates that are a year or less also provide very high earnings, and earnings above what the average college graduate makes.
The answer on the workforce development programs is, it all depends. What you need to know is that it can be the case that a three-month or six-month program can do you a lot of good, but the proof of the pudding is in the tasting of it. Everybody has to have the data that shows that the people that participate in that program, before you joined it, did it do them any good? If it didn’t, I would suggest that we not fund the program. Especially with narrowly conceived programs. It’s one thing to get a general four-year degree in philosophy — you’ll probably be ok in the labor market because you’ve got a four year degree — but if you’re going to get a three month certificate, or a year certificate, you need to know that it will get a job at a decent wage which will justify the cost of the program. The only way you know that is to record the data on what happens to people. For the most part, there are now 44 states in America that have built information systems that will tell you that. Probably only a dozen of them use the information aggressively, but the building of the data basis is relatively recent.
There is increasingly a move afoot in states and at the national level to require that colleges, universities, training academies, whoever supplies the education and training, if they get public money for that education or training, they need to tell the students what the job impact will be. That is the regime that I think we are moving into more and more, state by state, and ultimately, I think, nationally.
Which is not to say you won’t be able to take philosophy, it’s just to say that there needs to be, there is a growing demand for transparency: the notion that the student has a right to know what this program costs and what it’s going to do for them.
Pew Research detailed how the middle class has decreased as a portion of the population. Does your report help explain that trend?
Dr. Carnevale: Yeah, it is, the middle class has decreased slightly, which is different from the narrative that says it’s gone. And I think the reason that this has occurred — and it shows up in our data, not on this recession, but more long term data — is relatively clear. All those highly paid high school jobs, those have melted away. There are fewer and fewer of those. And those have been disappearing faster than the new middle skill, middle wage, middle income jobs — which tend to be things like like being an LPN in a hospital or being someone who understands computer language. The new middle skill jobs have not been growing as fast as we are losing the old jobs that were not middle skill. They were high school educated, but they paid a solid middle class wage. So in order to get that wage now, you have got to have at least some college.
There is another issue here. There is one view that the middle class is sort of declining in size, and that I think is essentially true, although I think the decline gets overstated. It’s due to the disappearance of that industrial economy, or the downsizing of that industrial economy. It provided solid middle class jobs to people with only a high school degree. Now to get a middle class job, with the exception of maybe 15-20% of males with only high school degrees who can still find middle skill jobs in construction, manufacturing, utilities, transportation, and the old blue collar economy, but now to get a middle class job, you probably need to get some kind of post secondary education that leads to some occupation that pays well. If you were lucky enough to have a place with a union, it would be better, but getting a blue collar job you could do pretty well that way. That avenue is getting narrower, narrower. The question ultimately in the long term is: the slight decline in middle class wages and jobs, is it just because we are not producing the new middle class jobs as fast as we are losing the old ones, or is it a permanent decline in the size of the middle class in the United States? None of us really know the answer. The evidence so far only says we are losing the old middle class jobs, the high school educated middle class jobs, faster than we are making the new ones.
This interview has been edited and condensed.