Monday, March 7, 2011

Increasing Return on Education Dollars

Paul Krugman's article in Sunday's New York Times is generating a lot of talk.  Krugman points out that a lot of jobs considered "white collar" are subject to automation and are currently being outsourced.  From this point he extrapolates that "education is not the answer", but instead the answer is to give more bargaining power to unions.  

Over at Restoring Dignity To The Law, J-Dog emphasizes that although increasing education may have led to a better life 40 years ago, the same may not hold true now.  He also mentions that education has shrinking economic returns.  Over at Law School Tuition Bubble, Matt also references the article.

Is Krugman right?  Should we just encourage our kids to be janitor with strong unions instead of going to college?  More below.

First of all, I will suggest that the division of "blue collar" vs. "white collar" jobs is is archaic and mostly meaningless in today's economy.  There are "blue collar" jobs that consist of a person sitting in an office in front of a screen punching in a code.  There are "white collar" jobs that require a person to go out and take soil samples in the mud.

A more useful metric - and the stand in for what I think Krugman was really referring to - is to classify jobs as to their ease of automation (and considering that the ease of automation is always increasing).  That is, I will postulate that all jobs - absolutely all - can be automated to a greater or lesser degree and at a greater or lesser cost, given time and development.  Instead what we should look at is the cost and difficulty differential of automating the job.  For example, take the UPS guy.  Instead of a UPS guy, we could automate the system using self-guided trucks leaving from the warehouse with packages with RF-ids that are automatically delivered to a building's automated reception.  From there, robots could read the RF-ids and they could be routed to the recipient's location.  We can pretty much do this now, but the cost would be enormous.  It is much more efficient and cost effective to just have a human deliver packages using the UPS truck.

Conversely, a checkout person at a grocery store is much nearer the "automation-point" than the UPS guy.  There are already self-checkouts in many grocery stores and store adoption and user comfort are increasing.

Additionally, for many jobs in the publishing industry, we passed the automation point long ago.  For example, we don't need a lot of English majors to spell check and grammar check new books - computers have been able to quickly do the work for us for years.

Note that viewing jobs with regard to their ease of automation is really neither pro- or con- education.  For example, the widely-held assumption is that further education leads to jobs that are tougher to automate, but that's not true in practice.  Consequently, why not look directly for "jobs that are hard to automate" instead of "increasing education" if we want to help young people prepare for their future?

However, even beyond the ease of automation, what Krugman's article is really getting at is a mismatch in supply and demand for people with certain skills - and the accompanying costs, both personal and societal.  I think that this is a real tragedy for our society - and even more frustratingly, a mostly preventable one.  (No, I am not saying that we need to adopt a distopian future like Aldous Huxley's Brave New World wherein everyone's career is chosen at birth and people are physically conditioned to their careers.) 

Instead, we need to get real and we need to get transparent with our hiring information.  Then we need to position our societal investments to achieve the most gain - without exercising outright control.  For example, when a high school student is considering college, we need to stop saying "do what you love and the money will follow" - it is a hollow, pathetic farce.  Instead, high school students should be presented with salary information and hiring rate information with regard to all prospective majors - and not the crappy, biased-to-hell NALP-style information, but true, useful information with strong enforcement and auditing behind it.

Next, students should be given some insight into what their life will be like at certain incomes.  That is, it is really hard for a high school student to grasp what a $30K/year lifestyle is like as opposed to a $60K/year lifestyle - both seem like a huge amount of money to them.  As a hypothetical example, they could be shown that if they major in English, their odds of getting hired in that field are low and they are likely to make less than $30K/year.  This will most likely prevent them from owning a home, they probably won't be able to take vacations or buy a new car, they will be scrimping and saving your entire career - here is what that feels like.  Conversely, if you major in accounting, you are likely to make more than $60K/year, here is what your life will fell like at that earning level.

As part of this information, there should also be an estimate of the future for that career - are the number of people employed growing or shrinking?  Is the field likely to decline soon due to automation, etc.  Our goal in providing this information is to allow students to self-select careers that minimize the painful supply/demand imbalance.  For example, it does not do society any good to have 10,000 JDs who can't find a job.  If you consider that each of them took out about $200K in loans, then that's $2 BILLION of our money that has been wasted in a bad investment with no return.  Also, those 10,000 people are mostly pretty smart and could have been making great contributions to other areas of our society - and we have lost that, too.  That's bad for us - but terrible for them.

Second, we need to position our societal investments to help minimise the supply/demand imbalance.  For example, I think that it is absolutely insane that we don't take major into account when we are dishing out student loans.  For example, if the career data analysis above indicates that we need 100,000 accountants, but only 1,000 English majors, then we should structure our student loan program accordingly.  That is, for the 1,000 "English-major" loans, those would go to the 1,000 people with the top scores that want to major in English.  Other than that, if you want to major in English, then go right ahead, but you are doing it on your own dime.  Conversely, if you want to major in accounting, then we will have loan money available for you.  It should be clear to all participants that student loans are society's investment in education for positions that are going to be needed in the future - not just giving people a huge rope to hang themselves with or just giving them a license to indulge a useless hobby or hide out from the real world.

At first glance, to a freedom-loving person like me, such a system seems kind of autocratic and therefore undesirable.  However, the more you look at it, it's just society providing an incentive to get something of benefit for itself - and providing an ancillary benefit to the student.  There is no actual loss of freedom because people can still do whatever they like on their own dime.  Further, the results of the program should be better than the current regime because the skills of the students graduating from college should more closely match the needed skills - therefore unemployment should be lower. 

For example, consider law students.  Let's say the societal projection is that we will need 10,000 lawyers in three years, so 10,000 "Law" student loans are available.  Bob graduates undergrad and then applies to law school and is accepted, but his criteria are not high enough to get one of the 10,000 Law loans.  The most likely situation then is that Bob declines to go to law school.  Fast forward three years - under today's program, Bob would have gotten the loans and now he would have been in a world of hurt because he can not find a job because the 30,000 people who went to law school are competing for the 10,000 available jobs. Conversely, under the proposed system, Bob would not have the debt and would probably be working in another field.  Additionally, the opportunities for the 10,000 Law graduates would probably be much better due to the decreased competition.

Is this a loss of freedom for Bob or is this really more of a program that optimizes the returns for our society as a whole?  I submit that sometimes it is economically good for Bob not to get what he wants - both good for Bob's personal economics and good for the economics of our country as a whole.

What do you think?

UPDATE: Matt over at Law School Tuition Bubble just posted some data that indicates that the expected demand for lawyers from now until 2018 is only around 100,000 - but law schools will produce around 500,000 law graduates over that time.   Wow.  Assuming $200K in loans for each of the 400K non-hirable law students, that's $80B in wasted taxpayer money - not to mention 400,000 lives ruined by being saddled with non-dischargable debt that can not be paid off.  Again - $80 BILLION dollars.

This is crazy!  We absolutely need some sanity in the student loan market.  In the home loan market, we have protections in place so that people buying a home can't get a mortgage greater than a certain percentage of their income - it's a "loss of freedom", I guess, but people are accepting of it and it lowers defaults.  We also require a certain downpayment.  Further, the lender simply will not make the loan unless the property's value justifies it- regardless of the price that the buyer is willing to pay.  Also, we call these "protections", not "barriers to home ownership" because they are designed to protect both the purchaser and the market.

These principles seem directly relevant to the student loan market.  There needs to be someone who looks at the value of the degree (earnings and open positions) and will not make the student loan unless it make sense - even if the buyer/student really wants the loan.  Further, these should be thought of as "protections", not "barriers" - and should be put in place to protect the student and our society's employment as a whole.


  1. "In the home loan market, we have protections in place so that people buying a home can't get a mortgage greater than a certain percentage of their income - it's a "loss of freedom", I guess, but people are accepting of it and it lowers defaults.  We also require a certain downpayment.  Further, the lender simply will not make the loan unless the property's value justifies it- regardless of the price that the buyer is willing to pay."

    Were you alive during 2003-2007, because this comment makes it seem like you were born yesterday. And, BTW, accountancy at the low end where you earn 60k is going the way of the Dodo - HTH.

  2. I'm surprised. I totally agree with this post, although I don't think we even need to call it "protections" for the student or society. We flat out are allowed to "protect" taxpayer funds from being wasted on something the country doesn't need (and I'm a huge liberal - but that just means I'd rather spend our tax dollars on something like universal health care, not 500,000 more law school grads).

  3. Thanks HTH - Actually, I thought about writing about this in the main article, but I felt it was getting long. I was indeed alive during 2003-2007 and I saw what happened when the lenders started ignoring the lending criteria - it created a bubble. When you ignore good lending criteria - when you just start handing out money without a good regard for the value that you are getting - when you believe in hackneyed phrases like "real estate always goes up" (sounds a lot like "education is the way to a higher salary", doesn't it?) - then the value of the asset becomes distorted by the lending process and no longer reflects reality. Then what happens? Sanity needs to re-assert itself somehow. Lenders need to return to solid criteria that generate a solid return.

    So why hasn't the education market "adjusted" like the home loan market? Because education is not a tranferrable asset like a house is. For example, take a 3L who has spent $200K on their degree and realizes that they are not going to get a law job. If education were transferrable, I could then buy that degree at a significant discount and the market would quickly find the right price for a degree.

    The terrible part is that student loans are not dischargable like home loans, so instead of people getting free and clear of the overvalued asset that they have purchased, they instead have that debt hanging around their necks for their entire careers. That's not good for them - or the country.

    If lenders had maintained good lending discipline and stuck to lending criteria, then the housing bubble would not have happened. From the housing crisis, we should know what happens when we abandon discipline and allow supply and demand to become distorted - why are we not already applying that lesson to the student loan market?

    One reason is that the losses in the student loan market are not as visible - shattered lives, lessened US competitiveness, lessened return on taxpayer dollars invested in education - actuall cash losses hidden by an accounting system that won't actually record the loss for 25 years for students who can't pay- all of these are not as visible and easily quantifiable as the housing market.

    But why do we have to wait? It's time to be honest and come clean and clean up our mess before it gets bigger. Discipline needs to be imposed. Imagine what would have happened if lending discipline had been imposed on the home lenders in 2003? If lenders had decided to be more rigorous in their criteria and had started using the criteria that they are using now? There would not have been a 2003-2007 bubble and frankly, the US would have been much better off.

    Today we have much the same situation with regard to student loans. We need to "man up" and impose discipline - because if we don't, the problem only gets worse. It might be 50,000 dissappointed law students today (which is a national tragedy), but it is going to be 400,000 disappointed law students by 2018, which is just stupidity - preventable stupidity. It will also have a vast economic impact on the country, both in lost tax dollars and lost economic gain from people remaining unemployed or underemployed. We need to align out educational loan activity with the market - like the housing loan system learned to its chagrin.

  4. I think your update has a better angle on how to limit bad student loans than your original post: rather than society determining a quota of required English majors, instead let lenders figure the expected earning potential and thus likely ability to repay the loan when deciding whether to give out a loan. For them to have the incentive to do this, student loans need to be dischargeable in bankruptcy, so the lender will feel the pain when they've made a bad loan. You are right to make the analogy with mortgages: the bank gives you cash with a physical property as collateral; it's the same with a student loan except that your human capital (skills, etc) is the collateral. If your human capital does not have a high market value (= your house doesn't have a high market value), you shouldn't get a big loan.

    Of course, the trouble with this is it's difficult for a bank to assess the future human capital of a loan applicant. Do we really want to give admissions committee powers to bankers?

  5. (1) Yes, “blue” and “white” collars are archaic terms. They had to do with *literacy* not routineness. Barteleby the Scrivner was a white collar worker, even though he was easily replaced with typewriters and carbon paper. Incidentally, literacy is why law has traditionally been considered a prestigious occupation. People are more literate and law practice has diversified. This is another problem with legal education: it needs to be better specialized along practice lines. Sure, some core formal education is necessary, but the one-size fits all juris doctor is a relic.

    (2) As to informing high schoolers, it also occurred to me recently that we need to educate people with better “labor literacy.” It reminds of something Jim Hightower recommends: a labor section in newspapers to complement business sections.

    (3) Student loans? We could just get rid of them entirely. I mean, we had higher education without them in the past, and the toxic excuse that they allow access for the poor and the diverse are misguided because the poor and diverse need jobs and lower education before they need access to higher education. Student loans are premised on the assumption that if you get the education now you’ll earn more in the future. If we agree that education is decoupled from income, then I don’t see much of a point to them. It also dovetails with another belief of mine: Aside from the truth that higher education isn’t for everyone, the other myth is that higher education should occur directly after high school. Many people are more focused when they’re older and consequently benefit. I recommend Martin Hutchinson [] and Richard Vedder’s ideas [] [].

    (4) “Is this a loss of freedom for Bob or is this really more of a program that optimizes the returns for our society as a whole?” Bob is also free to become a heroin junkie, yet I’m in favor of taking away his freedom by making heroin sales illegal. This isn’t out of tyranny, just a recognition that the outcomes of certain choices are so frequently detrimental to the people who make them and produce negative externalities for everyone else that it’s worthwhile to foreclose those choices. Except for the poppy growers.

  6. Hi Andrew, Thanks for the comment! I agree with you in general with two caveats: 1) we have seen that bankers do not always behave rationally for the long term (witness 2003-2007 bubble). We want to avoid a recurrance, so we want to have good compliance with reasonable lending standards. 2) A lot of the student loan activity is federal now - and they have a hard time resolving profitability. You may be arguing to return the student loans more to private lenders (which may be good to insert the profit motive, but bad if it raises costs on lending for careers that we want). However, even if we keep most lending federal, the imposition of better loan discipline should still be helpful.

    With regard to your question about giving admissions powers to banks, banks would NOT have admission powers. The admissions committee would make its decision and inform the applicant just like now. The applicant would then attempt to obtain the loan for their desired major - if they are not successful, then they can switch their major and apply for loan dollars in the new major. Additionally, if they are turned down for their initial major, the lender could provide them with a list of majors that would be acceptable and they could choose one of them.

  7. Hi LSTB, thanks for the comment. I'll respond to each item below.
    (1) Interesting!
    (2) Interesting idea. Although I'm not sure how "labor literacy" meshes with personal finance or what it is, really. I also think that a full year course in personal finance should be required for a HS diploma. We have too many bright high schoolers with no clue about the world. They have been too sheltered and need to learn the realities before they make bad decisions through ignorance that impact the rest of their lives.

    (3) I agree that often people that are returning students are more focused then the "straight-to-college" student, but I think that there is a fair amount of selection bias at play - once people are already working, only those that are more focused then average return to school.

    In a larger sense, I have to disagree with you with regard to student loans - at least with the cost of public education being as high as it is. I think that student loans are useful in that they provide a positive return on investment for our society in general. (Note that this is in the context that we choose to make paying for public education the responsibility of the student rather than making it free as some other countries do). Although as we point out above, there is a considerable amount of waste in the current system.

    If you kill off the student loan program, you are going to see a massive decline in graduates - college, law, medicine, etc. That might be OK in certain fields, but there are many fields that we don't want to have a decline - or maybe only have a small one. For example, without loans I would estimate that the number of medical students would crash about 90%. College students maybe 80%. Now, it wouldn't be so bad if the number of english majors crashed 90%, but what about engineers, accountants, scientists, teachers?

    Hmmmm... You may actually be suggesting that with the reduction in student demand, schools would be forced to lower their tuition, making it more affordable for everyone. That sounds good in theory, but the counterpoint is that there is a bottom limit to how low tuition can go and many schools will probably end up closing - and that's a lot of people out of work.

    I will admit that there are certainly some issues with regard to schools raising their prices because federal loans will pay the new higher prices. However, the overall principle that society invests in the education of its brightest and most hardworking people so that those people can provide a better return for society seems like an effective principle to me. The current way this gets done is the student loan program - which has its pros and cons. However, if you have some alternate way to get this done, that would be very interesting.

    (4) Interesting. I am a big believer in personal freedom, although I recognize that actual (or in some cases a high probability of significant potential) harm to others may justify curtailing personal freedom. It's easier for me if I approach this from the other side - that tax dollars should be carefully spent to provide a benefit for society, not just handed out with no regard for return.

  8. MP,

    Thanks, as always, for the thought-provoking post. You made a point in the comments with which I disagree:

    "If education were transferable, I could then buy that degree at a significant discount and the market would quickly find the right price for a degree."

    The market does this, either by compelling new law grads to work as unpaid or underpaid apprentices or by pushing them into other fields like teaching or sales. To the extent a law grad holds out for a market rate legal job (based on her own skewed view of the "market"), she is assigning a value to her degree equal to the $200k she paid for it.

    Education is transferable, but price discrimination is at work. The challenge is to devalue the degrees in the minds of their holders. Debunking the benefit of tangentially career-related educational goals in the Western world is a cultural problem that will require an ideological shift over several generations. The Western ideal for all of written history is "I study war so my son can study mathematics, so his son can study music."


  9. 12:59 - Interesting. I think that you are right that the market exposes the reality of the worth of the degree to some extent. However, the degree itself is not transferrable. To try to put the parallel into the housing market, it's as if once people bought a house they could never sell the house and had to keep paying on it forever. I think that you would agree that in such a situation the true market might still be somewhat distorted.

  10. The personal freedom to hang yourself with the chains of debt. Gotta love that personal freedom, I tell ya.

  11. 11:02 - LOL - Yeah, it's a tough issue. You want to protect people, but still preserve their free will. However, one thing that I know plainly is when the information that people are relying on to make their decision is not accurate and is misleading, then their free will has already been corrupted. Also, if we are going to make the student loan debt nondischargeable - unlike any other kind of debt - then it would seem like stricter scrutiny is appropriate.

  12. "Do we really want to give admissions committee powers to bankers?"

    This isn't exactly true. The schools could still admit anyone they want. The difference is that banks could only distribute funds for 1,000 grads, because that's all taxpayers want to pay for.

    Right now, we have admissions committees distributing tax dollars with no strings attached and no financial criteria - that's not better.

  13. Good point, Liz. Also, apparently they already do this in Canada - here's a quote from a conversation that I was having over at The People's Therapist

    "As a Canadian, we do not, despite our liberal reputation, allow everyone who wants to go to law school to go to law school. In the US, there is some horrific hybrid — not quite socialism not quite free markets. Unlike Canada, the US Fed govt doesnt control the schools or keep costs down, but it doesn’t allow the market to rule either – no, instead it has the Department of Education and gives out grants, loans, helps subsidize private loans, and is waist-deep involved in the educational system, creating the market distortions of socialism without any of the benefits of government regulation of the market. If I may – Canada has been spared the worst of the US legal system not because it had Trudeau instead of Reagan, but because the CBA and the Canadian government won’t allow law schools to multiply throughout all the land on some idealistic notion that anyone who ‘dreams’ of being a lawyer is entitled to pursue that higher education."