The great tuition debate: A rejoinder

This is a response to this, which is in turn a response to this and this, which was a response to this, which was a comment on this old E-Lab post of mine. This is old, well-covered ground, so I'll be as brief as I may.

Here are some problems I see with the case for free tuition as expressed in Dr. Dawg's post.:

1. "Education, in brief, is a public good." This is incorrect. Education does not satisfy the criteria of what it means to be a public good. To be sure, there are positive spillovers: for example, university-educated people generally earn more, and so pay more in taxes. This is why I have no problem with the notion of a partial tuition subsidy. But most of the gains to PSE - monetary and non-monetary - are captured by the student. See here for more on this point, and see here for Frances Woolley's "Why public goods are a pedagogical bad."
2. Tuition is not the only cost of PSE. Implicit throughout the post is the presumption that if tuition were free, the university wouldn't cost anything, and all potential students would be on an equal footing. This too is incorrect. The opportunity cost of PSE also includes the wages foregone by the student. This sacrifice is larger than tuition; depending on your circumstance, tuition accounts for roughly one-third to one-quarter of the total cost of a university education. Bringing tuition down to zero does not make university 'free.'
3. "Gordon appears to assume that the current demographic snapshot of university students is fixed." No. The stated goal is to change those demographics; the debate is about how to best achieve that goal. The claim that a policy in which 90% of spending consists of giving free money to people who are already going to university will somehow change those demographics requires more evidence than a flat assertion. 
4. Concerns about student debt. Most graduates either do not have student debt or are able to pay it off without undue inconvenience. It doesn't make a lot of sense to reduce tuition for everyone in order to address the problems of a small minority. To the extent that government action is necessary, it should be a targeted to those who need it.
5. The European example of free tuition. There is no evidence that I'm aware of that suggests that this model has a measurable effect on access to university.
6. Universality. I don't see how or why the notion of univerality should be endowed with a moral weight comparable to that of vertical equity (redistribution of wealth). Universality is a compromise, the price that progressives are sometimes willing to pay in order to implement a watered-down version of a their preferred, more strongly progressive policy. This compromise may be for political economy reasons (redistributing the gains farther up the income distribution to ensure the support of the median voter) or to reduce administrative costs. I'm not aware of a convincing argument in which the universal nature of a program justifies directing public funds to the top half of the income distrubution.

 There are other points, but that's enough for now.

Comments

[Sean Hunt]

Thanks for writing up my thoughts more coherently than I could.

[Dr.Dawg]

I posted my response at my site, here: http://drdawgsblawg.ca/2014/10/the-great-tuition-debate.shtml#comment-1621716599

[Neil]

Question: what do you think of alternative tuition models? Not free, but deferred. I've heard of somewhere (Ireland? Australia?) where by attending university you sign up for a 2% surtax for life, but pay nothing up front. This seems like it addresses access issues, and better distributes cost to those who gain the most from attending. It also leaves open more opportunity for graduates to pursue lower paying public service or non-profit careers, since their educational costs vary depending on what they do with them.

There's of course some details that would need to be worked out to deal with emigration, etc, but fundamentally this seems like a sound option.

Barring a complete overhaul of the system, a really big thing that I think Canada needs to address is cost certainty. When you start University, your fee schedule for at least 4 years should be set, rather than being left to the vagaries of an annual tuition hike debate. Not knowing if you'll face a 2% annual hike or 8% makes it difficult to make an informed choice until you've already invested a substantial amount of time and money.

Likewise financial aid should be confirmed for the 4-year period. Whenever tuition debates come up, University administrators like to point to how much money is available in scholarships and bursaries to ensure access. However, because you don't find out how much you'll receive in loans and bursaries until the beginning of each year, everyone needs to start their degree with a plan for paying the full sticker price. Better up front information could make a difference in access.

[Jacques René Giguère]

Stephen:"But most of the gains to PSE - monetary and non-monetary are captured by the student" -University graduates are complement to other workers. Even orinary workers benefit from the higher education of others. The way they benefit from a doctor's education

[nemi]

If you put point one and two togheter, you will realise that a fully subsidised toutition = partly subsidised education.

Maybe the "optimal" subsidy would be 70 or 140 % of the toutition cost (i.e., what, 30-60 percent of the education cost?), but when we aren´t sure a 100 % subsidy of the toutition cost seem pretty reasonable to me.

[Bob Smith]

"Even orinary workers benefit from the higher education of others"

Well, it's undeniable that my secretary benefits from my post-secondary education, but that doesn't seem like a particularly compelling reason for my making her pay for it.

"The way they benefit from a doctor's education"

Yes, but they pay for those benefits, though doctors fees/income tax.

[Patrick]

I get what Stephen is saying. I even agree with it. But I find myself asking why wouldn't his arguments apply equally to e.g. elementary or high school? Or conversely, why doesn't the same logic that applies to universal public secondary education not apply to university? I think Quebec tried to grapple with this issue with the CEGEP system. Not sure how successful it was.

If we're saying that secondary education has become insufficient to accomplish whatever goals it was supposed to accomplish, then maybe the solution is to fix it. Universities where not designed to provide education to the masses, and they don't seem to be scaling very well. We already have a system in place to provide mass education. We should make better use of it.

[Stephen Gordon]

A couple of points:
1) Elementary and secondary school are compulsory. The 'public good' argument works more strongly here, because a democracy needs a literate and numerate citizenry. Most of the gains to PSE are private.
2) Sacrificed wages don't enter into it, because parents/guardians are obliged to supply the necessities of life to minor children.

If we made PSE compulsory, extended parental obligations to age 22 and ensured that *all programs in all universities were of equal quality*, then extending the elementary/secondary free-tuition model to PSE would work.

[Bob Smith]

Patrick,

I think you're on the right track. From my perspective, the role of elementary and secondary education is to provide citizens with the neccesary skills to fully participate in a liberal democratic society. That's a basic endowment of skills/knowledge that we owe all citizens. (The rationale for universality is, at least in part, that we want to encourage all citizens to participate in the same system and to ensure that its a reasonable quality, universality is one mechanism for achieving that to some (albeit imperfect) degree).

Is a post-secondary education neccesary to fully participate in a liberal democratic society? No, absolutely not. The role of a post-secondary education is two-fold (i) to impart specialized knowledge/skills on students and (ii) to give them a little piece of paper that says they have succesfully absorbed most of it. The latter is of purely private value and does nothing to aid in civic participation (other than for dorks who feel "I'm a university graduate so listen to me" is a meaningful contribution to civil debate).

Having specialized knowledge can, of course, be helpful to civic debate (although its clearly not essential, since none of us have specialized knowledge about everything and nevertheless feel inclined to participate in civil debates), but it doesn't really require a formal post-secondary education. As the memorable line on the merits (or lack thereof) of post-secondary education from Good Will Hunting goes: "you dropped 150 grand on a fuckin' education you could have got for a dollar fifty in late charges at the public library!".

That isn't to say that our current public education system does a particularly good job of imparting the basic skills of citizenship on youn Canadians, witness the (i) rank illiteracy of recent high school, to say nothing of university, graduates, (ii) the widespread and general innumeracy and ignorance of statistics - a damning failure in a modern democratic state), (iii) low levels of historical literacy and knowledge of our society and its institutions, etc. And of course, there's the massive failure vis-a-vis First Nations Canadians. But the solution to that, as you suggest, is to improve (or perhaps more accurately, fix) our existing public education system, rather than trying to expand something that

[Howard Cohen]

It is nice to talk about tuitions in a theoretical framework, but in the medical field we now have empirical evidence of the effect of high tuitions. In the eighties medical schools were populated by the top fifty percentile by income, today few students come from the bottom eighty percent U of T, with a high tuition,graduates doctors with a median debt of $140,000. While I do not have the numbers form Quebecs four medical schools, I am sure the student body is more representative of the general population. The net effect of excluding student from other than upper class, is less physicians in poorer and rural communities. Physicians who train in a mixed environment are more likely to understand the difficulties and financial barriers to achieve good health. A poorer patient who cannot afford his/her medications will not take them. A single parent at a minimum wage job cannot take time from work to bring their sick child to the doctor, and thus delays care. Physicians from wealthy backgrounds do not even consider these barriers. Maybe other fields of study should be more self supporting, but medicine truly is a public good, and needs to be treated as such

[Patrick]

" extended parental obligations to age 22 "

At least where child support is concerned, parents are on the hook past 18 if their kids are attending PSE. IANAL, but it may apply more generally.

[Andrew_M_Garland]

Students are going into debt and inviting failure because the schools defraud them using statistics. Schools advertise that a college degree is associated with $1 million in additional lifetime earnings. It is a no brainer to borrow $120,000 in order to earn an extra million.

But, that $1 million figure is true on average for degrees 40 years ago, when colleges were much more selective, many fewer people attended, and they came from wealthier families. It is not true for the average student today.

Also, I suspect that the $1 million figure is the simple sum of increased earnings over say 40 years. That would be an extra $25,000 per year. The present value of that benefit at 5% is $428,997. (That is, $428,997 invested at 5% will pay you 40 payments of $25,000 until the money runs out.)

At a 25% tax rate, the present value after tax drops to $321,733. The student loan is not deductible; it cannot be subtracted from the extra income to lower the tax. And, the loan is only part of the cost of getting the degree.

I could use the same bad reasoning to show that owners of expensive cars earn much more than owners of economy cars. So, all we must do is lend money so everyone can buy an expensive car!

College is an expensive IQ test

The law says that a company cannot give an employment test unless it has been shown to be non-discriminatory in effect, that it doesn't screen out people of color at a different rate than people of pallor.

So, employers don't create their own tests or use standardized tests. Companies rely heavily on college degrees to give them some little information about the quality of candidates. Interviewers talk randomly about whatever they want, using personal judgment to decide if the candidate is "a good match". This is supposed to be less discriminatory!

Schools are conveniently exempt from testing restrictions because they are supposedly altruistic and not connected to the filthy pursuit of money. So, they administer tests to determine who gets in and what grades they receive. I think that most of the excellence claimed by the top schools is actually selected up front by taking the students who test best out of high school. Schools offer no magic "excellence" which can accept a poorly testing student and produce a great testing one.

The Career Consequences of Failing versus Forgetting
02/09/12 - EconLog by Bryan Caplan [edited]
=== ===
How would your career have been different if you had failed all the classes you've totally forgotten?

The Human Capital model proposes that schools teach useful stuff. Never learning course material (failing) should have exactly the same career consequences as forgetting the course material. Either way, you lack the skills, and the labor market should treat you accordingly.

The Signaling model proposes that schools discover character by making students perform difficult tasks, even if useless. Here, the consequences of failing and forgetting are different. When you fail to learn useless material, you send a bad signal. When you demonstrate mastery of useless material, you send a good signal. The material doesn't matter. Employers naturally snub people who fail, yet smile upon those who merely forget.
=== ===

[Patrick]

Howard Cowen - Considering the NPV of wages a doctors can expect to make over their career, even in socialized Canuckistan, (see here), I see little reason to worry about a $140K down payment. The return on the investment seems pretty good to me.

I suspect that the selection criteria for getting into medical school in the first place very much favors students from wealthy backgrounds. Why give the rich a further break on an already lucrative investment? If you want to increase accessibility to the professional, you'll have to start long before the student is being charged tuition.

[Patrick]

Oops. Typo. Should be "accessibility to the profession".

[Bob Smith]

"While I do not have the numbers form Quebecs four medical schools, I am sure the student body is more representative of the general population."

Howard, don't take this the wrong way, but without any data to back that belief up, what possible weight can it be given? I'd also be curious as to the basis for your data on the backgrounds of doctors in the 80's compared to now. Is there a source for that. I'm not trying to be a jerk about this, but too much of this debate is based on people's subjective beliefs, which beliefs are often not supported by (or are inconsistent with) the empirical evidence.

Now, as it turns out, there has been some excellent research on the impact of higher tuition fees on professional school enrolment by Marc Frenette which concluded that modest tuition fee increases had no significant impact on the socio-economic make-up of the student body (measured by parental education as a proxy for income), while large (sudden) tuition fee increases (in Ontario) resulted in increased enrolment, over all, and increased enrolment from families from low and high socio-economic backrounds, with a drop from students from middling socio-economic backgrounds. (http://publications.gc.ca/collections/Collection/Statcan/11F0019MIE/11F0019MIE2005263.pdf)

That seemingly counter-intuitive result is explained (I think credibly) by the fact that increased tuition is usually associated with (i) increased overall enrolment (i.e., schools can provide more spots when they have more money) and (ii) increased students aid for poor students. Where enrolment among the middle-class groups fell in some case (though not with moderate tuition increases), the suggestion, again wholly credible, is that the combination of a sudden spike in fees (which people couldn't prepare for) coupled with inadequate financial assistance may have priced them out of the market. Of course, that's an argument for gradual (and predictable) tuition increases coupled with properly designed student aid programs, not for low or no tuition.

And to follow-up on Patrick's point, the reality is that tuition could be set at zero and the children of the wealthy will always be over-represented in post-secondary education (indeed, to the extent that zero-tuition polices reduce the overall funding for universities, reducing the number of spots, such a policy might even increase the representation of wealthy kids, as poorer kids are squeezed out), for a whole host of reasons (some of which can be ameliorated by state policy, others simply cannot). I'll give you an obvious example, I'm going to suggest that the under-representation of aboriginal Canadians in post-secondary education has nothing to do with tuition fees (an easy proposition since in many, but not all, cases aboriginal Canadians are entitled to free tuition), and everything to do with the whole host of challenges facing young aboriginal Canadians in getting to a point where university is even an option - it's hard to go to university if you can't graduate from high school, have to overcome an abusive family, suffer from FAS or are battling a drug addiction. Now, that's a clear cut example of a case where tuition fees, frankly, are the the least important barrier to university enrolment, but it applies equally to other disadvantaged groups in Canada

To the extent that the state can improve the educational achievement of disadvantaged groups freezing or lowering tuition is a very expensive and ineffective way of doing it (although one that makes middle-class families who benefit from it feel all warm and fuzzy - hence its popularity). I can think of much better uses of scarce public funds.

[Bill Ellis]

Maybe your recommendations are right for Canada, But I believe the US would benefit greatly from free tuition.

It seems to me your argument boils down to three objections...
1) Free tuition results in society paying for a privilege ( If it's not a right it's a privilege) that would mostly benefit the individuals who receive the privilege while somehow not benefiting the rest of society. ( Neat trick, that. I have my doubts to say the least.)

2) Free tuition is a subsidy to the rich, because it is the wealthy who attend college in the greatest numbers.

3) This third one is the the crux of your argument--- That there really are NOT significant financial barriers to deserving poor or middle class kids to higher education.

To objection three, I have to ask... Don't you think having to take on debt is a significant barrier ? Here, south of the border, 7 out of 10 graduates had an average of 30 grand in student loan debt last year. ( 10 grand average for graduates in Canada. Still, significant. ) Total student loan debt is around 1.4 trillion dollars. How can this not be considered a significant barrier ?

From Federal Reserve Bank of San Francisco's brief, "Student Debt and Default in the 12th District"....
"These trends of rising college costs, larger debt loads, and shifts to higher 
cost student loans are particularly concerning as they relate to students from 
LMI households.  These students are especially vulnerable as they must 
borrow a larger proportion of funds relative to family income than their 
higher‐income peers in order to finance their education, but are less likely to 
be able to rely on family assistance to repay educational debt.  As seen in Fig. 
7, the relative cost of postsecondary education is more burdensome for lower‐income households.ix  A family at the bottom income quintile would be 
required to pay more than 70 percent of family income to cover college costs, 
after accounting for grant aid. " Bold mine.
http://www.frbsf.org/community-development/files/student-debt-default-bay-area.pdf

Objection two... Is really a moral position. True it would subsidize the wealthy as well as everyone else. And maybe that is a moral flaw. We could means test tuition but since the cost of the whole program would be relatively small, so would the savings.

On the other hand, universality makes the program more secure and stronger. "A program for the poor, is a poor program" as the saying goes. If I am right and there are really are significant class barriers to higher Ed, than the higher moral position is to mitigate the barriers the common person faces, even if it means subsidizing the wealthy too.

Objection one is hard for me to accept. It seems way too narrow even in the realm of the reasonably quantifiable, but worse it ignores the benefits of education that are obvious but difficult to quantify, like effects on quality of life and society.
How do we measure the difference in the quality of a life examined, compared to one left unexamined ? What is it worth ? “Men can starve from a lack of self-realization as much as they can from a lack of bread.” ― Richard Wright, Native Son.
How can we measure the impact of a public that has the broad horizons that allow them to better understand their world, compared to a public that must except the authority of mass media manufactured reality, on our society and government ?

The federal government could provide free tuition for about 40 billion more than we spend on higher education now . That's a rounding error compared to our total budget. We subsidies the oil and gas industry to the tune of about 50 billion.

The status quo is unacceptable. The costs of universality are relatively minor. University is a strong political message. Opposing it in favor of augmentations to the status quo that in the big picture ( USA budget approaching 4 trillion ) makes little real difference is making a mountain out of a molehill. It blurs the issue. And blurring the issue helps those who defend the status quo.

[Patrick]

Bill Ellis:

You're assuming that everyone who agrees with Stephen's point (which is really an empirical matter, so there should not be much argument) doesn't want low income students to go to university or to get any help to do so. That's a wrong assumption. If we're going to spend scarce resources on a policy, it should at least have some hope of actually accomplishing the stated policy goal.

And as Bob and I both pointed out, there are disadvantaged groups for whom this is totally irrelevant. So if we're throwing around billions, perhaps we'd get more bang for our buck in other ways.

[Bob Smith]

Bill, let's consider those points:

"Don't you think having to take on debt is a significant barrier ? Here, south of the border, 7 out of 10 graduates had an average of 30 grand in student loan debt last year. ( 10 grand average for graduates in Canada. Still, significant. ) Total student loan debt is around 1.4 trillion dollars. How can this not be considered a significant barrier ?"

So if the average loan balance of 7 of 10 graduates is $30K, the average loan balance is $21k, right (assuming the other 3 of 10 graduates have a balance of zero)? I make the point because too often claims about "average" student indebtedness intentionally omit those with zero debt in order to scale up the problem.

I also have my doubts about the voracity of that statistic. It is regularly cited, but when you look at the underlying state-by-state date, it's not clear where it comes from (see, for example, this site: http://projectonstudentdebt.org/state_by_state-data.php). Only three states (all relatively small) have more than 70% of their graduates graduating with any student loans, in most states the proportion is in the 50-60% range. Hard to seee how they get the 7 in 10 number from that. Not sure how one gets to 7 out of 10 when in most states the number is 5-6 out of 10. Maybe there's some glitch that I'm not seeing.

Is $30K in debt a "significant" barrier? It's the price of low-end version of the best-selling car in America (a Ford F-series truck), which Americans seem to have little trouble financing (on, usually, less generous terms than Sally Mae will offer). Maybe it means university graduates will have to make do with a used civic for the forseeable future.

"True it would subsidize the wealthy as well as everyone else. And maybe that is a moral flaw. We could means test tuition but since the cost of the whole program would be relatively small, so would the savings".

Why would the savings be small? If people with high incomes will attend university in large numbers no matter what, the cost of subsidizing them will always be significant relative to the cost of subsidizing university as a whole. Surely those resources could be better used to provide greater subsidies for more poor people, no?

Indeed, many US state university systems have very effective and efficient means testing mechanisms. So while their tuition fees may be high, for low-income students, the effective tuition rate is already nil (note, this doesn't mean they have no student debt - people have to eat, pay rent, etc.). California, for example, has a program which apparently pays the tuition for students from families with incomes below $80K, which gives low income families access to what is arguably the best public university system in the world. If you can do that, why on earth do we feel the need to fully subsidize people from wealthier familes? All moving to zero-tuition would do would be to subsdize the people who are currently too wealthy to get subsidies.

"If I am right and there are really are significant class barriers to higher Ed, than the higher moral position is to mitigate the barriers the common person faces, even if it means subsidizing the wealthy too. "

There are undoubtedly significant class barriers to post-secondary education (lack of exposure to people with post-secondary degrees, barriers to proper elementary or secondary education, lack of encouragement, you name it) - indeed the Canadian research suggest that it is those barriers, not tuition fees, that impede access to post-secondary education. Of course, that's what makes free tuition such an offensive policy, as it does nothing to reduce the real barriers to access to education, while subsidizing those who face no such barriers.

Even if you're right that free tuition could be provided for a mere $40B, is that the best use of those resources? Couldn't that money be better used, rather than largely subsidizing the spawn of the current well-to-do, targetting the real barriers to post-secondary education: ensuring that inner-city schools are safe and effective learning environments, ensuring that kids don't go to school hungry, that libraries provide a stimulating enviroment for young minds, by providing poor children with access to role models they can aspire to emulate, providing programs so that poor kids who attend university can catch up with their more advantaged colleagues (one of the great tragedies in the US, especially at elite schools, is the high drop-out rate for disadvantages students who are simply outclassed by their more advantaged colleagues), etc?

The zero-tuition movement is the perfect policy for the typical middle-class revolutionaries who occupy student unions accross North America, it allows them to assuage their social conscience by spouting about their concern for the poor, while promoting policies which simply reinforce their existing privilege. That isn't to say that there aren't valid needs for reform in the way the US finances its post-secondary education system (their law schools, in particular, are a disaster, reflecting the combined greed of their faculty and the gullibility of their students), but zero-tuition fees isn't the answer.

[Patrick]

BTW Bill, in case you don't hang around WCI that much - Bob leans right, and I'm a lefty. So don't chalk this up to ideology.

[Stephen Gordon]

Howard:

Unless you have evidence that Quebec med schools are overwhelmingly populated by students from low-income households, my point stands.

[Mike]

A lot of tuition debates, including this one, seem to revolve around tuition's effects on whoever pays the tuition, be it students, parents, lenders, etc.

That's a good, worthwhile debate. However, another interesting discussion that doesn't seem to occur as often, is tuition's effects on universities and their different departments. Maybe it's too ambiguous to see any connection between tuition change and universities themselves. But if the above discussion proves anything, it's that the effects of tuition changes on the students are as equally ambiguous. What do you guys think about the effects of rising tuition on universities themselves?