Have you heard? They're middle class! Middle class I say! Let's hang the bastards!
Apparently a lot of students are middle class, and therefore (not that anyone seems to have bothered to check this) that means that a lot of the student rioters are middle class too. I might have more to say on this subject later this evening (capsule: student riots are in general a phenomenon that needs to be taken seriously when they happen; empirically they are correlated with all sorts of interesting stuff), but for the meantime this one goes in the "I mean really" file.
I've said this before (albeit in a context which made it inevitable we'd get sidetracked onto a discussion of woo in the NHS) but I'll say it again because it's important - there is nothing particularly bad and a lot potentially good about recycling some of the middle class's income back to them through the tax, benefit and public service system. Things like universal free university education tie the middle class into the system, and avoid the notorious and obvious public choice consequences of means testing - ie, that "benefits for the poor" become "poor benefits". Oddly enough, what with the economics degree and everything, I am aware that the effect of funding university tuition out of general taxation is probably regressive in its distributive consequences. But this would only be an interesting fact if paying for universities was literally the only thing that government did; people really need to be gently persuaded (in the alternative: beaten) out of trying to do distributive analysis of spending at this sort of disaggregated level.
I am, as I've said before in this space, a very typical middle class British burgher, genus "City boy", species "North London". As such, I'm wholly in favour of the enforcement of the law on common assault and on damage to property, and the prosecution of people who throw punches and break windows, in more or less whatever context. But given that the rioters are in fact going to be punished for whatever common assault and criminal damage they've committed, I'm not sure they need or deserve an additional hiding in the form of special moral and political opprobrium from me. Given that I tend to regard riots as an interesting but basically normal part of the political process when they happen overseas, it would be pretty odd and hypocritical to think of them any other way when they happen here.
I just wanted to let you know that on this and related issues (for instance, government funding of opera), you're posts on the subject changed my views fairly substantially.
ReplyDelete(It's difficult for me to understand the mentality that decides that the way to be a prosperous country is to make higher education as painful as possible.)
I think the decline of support for Citizen Basic Incomes has been one of the most strange things about the govt's spending cuts.
ReplyDeleteOne moment people like Tim Worstall are all in favour, next thing you know (when CB was reduced) they're going on about how benefits should be a safety net, means tested and targeted, and so on.
This is one time I think you've got it wrong. The benefits of a university education are overwhelmingly captured by the individual; the small incremental benefit to the state (higher personal and corporate taxes generated by graduates) aren't enough to make it worthwhile for the state to pay 100% of the costs of tuition. Which means, apart from any other consideration, that in a situation where the state were to be paying 100%, there would be quite a strong incentive for it to demand higher education that was as cheap as possible, with all that that entailed for quality.
ReplyDeleteI'm very happy with a good old-fashioned positive externalities story which says absent some state intervention (in the form of subsidy) supply of higher education will be sub-optimal, but that's quite a long way from 100% support.
But the benefits are *so* much bigger than the costs that there's plenty for all to go round; without getting into externalities or any such, the formal model would be:
ReplyDeleteC = cost of a university education
B = private benefit of a university education
T = tax rate.
As long as Bt>C then the state benefits from paying for people to be educated. Rearranging the algebra gives you the alternative criterion t > C/B - ie, that the tax rate is greater than the reciprocal of the productivity of university education.
So if we say that the marginal tax on a university education is basic rate income tax @20%, and that the fully loaded cost per individual is about £13,000 (the fee for a non-EU student doing physics at Oxford), then ignoring discounting, it still makes sense for the state to pay tuition if the lifetime graduate premium is more than £65k. Actually it's £100k.
In fact, there's enough left in there for the state to provide £7k of maintenance grant and still break even; I would argue that this ought to be the objective of university education policy. The fact is, the university education system subsidises the rest of the fisc here simply through the existence of the graduate premium (ie, the return on the formation of human capital) and the general progressivity of the income tax.
ReplyDeleteAnother way of looking at this is that anyone who says that it's unjust for the taxes of binmen to pay for middle class kids to go to university, must also end up saying that it's unjust for the middle classes to pay higher taxes out of their higher incomes later in life. Which is probably why in actual fact, nine times out of ten the people you see asserting the first are doing so because believe the second.
I don't argue with the structure of your argument, just the data. That's £13k per year, not for three years. So that's £39k for a physics degree (I'm not clear whether that includes accommodation). And if you're taking the £100k lifetime benefit from Browne, remember he's using a social discount rate to arrive at that amount. I don't begin to see the justification for that; the returns to this investment decision depend on the vagaries of the economy, so one surely ought to be using a higher discount rate.
ReplyDeleteEven if I grant you your £100k, you're still looking at 39 / 100, which is a lot greater than 20%. And I doubt whether the lifetime average effective tax rate is going to be close to 39% either.
Surely the state gets more than 20%, as there's all the other taxes which rise with income too?
ReplyDeleteGood spot on the per year thing, sorry. But the £100k is not only discounted but post-tax, so the relevant number is significantly greater. Also, physics at Oxford for an overseas student is (the more I think about it) a high-end estimate of cost - those students are explicitly meant to be subsidising domestic/EU students, and most places are much less expensive to run than Oxford.
ReplyDeleteSo if I take out my handy set of movable goalposts and use the average "Undergraduate Overseas Band 1" from Publicgoods.co.uk, then I can get the cost down to 30k; then I can gross up and undiscount the Browne graduate premium to 150k and claim a Pyrrhic victory from the jaws of defeat - obviously to the extent that graduates pay higher rate tax the task is made much easier.
It's not as clear cut as I thought though, you're right.
If you gross up the £100k assuming a reasonable average 30% tax rate you'd get to a pre-tax present value of £142k (although that figure would be lower if we were using a higher discount rate, as we ought to be).
ReplyDeleteAnd that £142k is an average of grads who did humanities & social sciences, and those who did STEM.
Your band 1 fees are *only* for humanities & social sciences, so those with the lowest cost of delivery. And humanities & ss grads earn less, on average, than STEM ones. To be fair, you'd need to match it up with the lower PV of benefits from doing a humanities degree.
So it still looks to me as though 100% state funding works out as a bad deal for the tax payer.
In any case (see, I can move goalposts too) do we want to live in a world where higher education policy is determined by a dimwit monopsonist, whose primary focus is cost, not quality? If you think that government on the whole is not very good at procurement (cough, aircraft carriers, cough), why should it be any good at procuring higher education?
In any case (see, I can move goalposts too) do we want to live in a world where higher education policy is determined by a dimwit monopsonist, whose primary focus is cost, not quality? If you think that government on the whole is not very good at procurement (cough, aircraft carriers, cough), why should it be any good at procuring higher education?
ReplyDeleteTrivially false. Not only did this work rather well from 1945 to 1998, the government didn't "procure" higher education in the sense that it procures aircraft carriers. Students procured it. The government paid for it.
This is, after all, one of those voucher/commissioner-payer split deals that neoliberal and conservative types usually love so very very much.
Your band 1 fees are *only* for humanities & social sciences, so those with the lowest cost of delivery
ReplyDeleteI think this is likely to be more than offset by the fact that I'm spotting the government the profit margin built into the overseas student fees, which I think is actually quite substantial; the standard EU fees might actually be closer to the cost.
I also don't think we need to do theoretical arguments about aircraft carriers - the model I'm suggesting is what used to happen in England and still does in Scotland, and it worked OK.
@yorksranter and @dd re your last sentence. How it worked in a world where 5% of school leavers went on to university is not relevant in a world where >40% do.
ReplyDeleteThe cost question needs careful consideration too. The quality universities can offer overseas students, and hence the price they can charge, is constrained by the quality of education that they can afford to offer domestic students. So if government puts a cap on income per domestic student, and hence quality suffers, then price to foreigners must be lower than it would otherwise be.
Anyway, back to the figures. If the PV(pre-tax incremental benefit) = 142 at a 2% discount rate, then at a more sensible 5% you've got PV(pre-tax incremental benefit) = 89 (you can replicate this by playing around with the annuity formula; I assumed the time period was 40 years).
At an average tax rate of 30%, that gives £26k, which is the amount a rational chancellor would be willing to stump up, on average, to contribute to the cost of educating someone at university.
Anything more than that starts to look like a rip-off for the tax-payer.
There is already a cap on income per domestic student - I am actually just suggesting that the government picks up the bill without all this playing-at-shops guff, but what you seem to be suggesting is that any system other than an unregulated fee market will lead to a decline in quality and I don't see that.
ReplyDeleteYes, I know there's a cap on income per student; that's my point.
ReplyDeleteNo, I certainly don't think an unregulated free market is the way to go. Given the frictions (oligopoly, potential for cartels, ignorant consumers, uncertainty about value, those positive externalities) there's no way an unregulated free market would be anything other than an abject failure.
But economic analysis suggests that 100% funding of university tuition is unaffordable by the state.
On the other hand, I'm not saying there should be *no* state funding.
A quick perusal of the comments on the various newspaper articles reveals a commonly held view that the Poly to University conversion was a mistake and that there are too many graduates without the corresponding graduate jobs. Also for example the entire accounting profession was filled with non-graduates a generation or two ago, but I don't see the fact that they are now graduates is exactly adding value. So if there is a negative marginal benefit from someone, say, doing business studies at a former Poly instead of starting a business, isn't this a valid (albeit strictly economic) basis for discouraging them by getting them to share the cost?
ReplyDelete"The benefits of a university education are overwhelmingly captured by the individual; the small incremental benefit to the state (higher personal and corporate taxes generated by graduates) aren't enough to make it worthwhile for the state to pay 100% of the costs of tuition."
ReplyDeleteOECD say $43k cost to UK govt for male students (including foregone income taxes), and $138k extra income taxes from higher earnings, "social contribution" and reduced unemployment payments. Likely to be an understatement as doesn't capture extra VAT, property taxes etc.
Which suggests that 100% funding of university tuition is eminently affordable by the state.
http://www.oecd.org/document/52/0,3343,en_2649_37455_45925620_1_1_1_1,00.html
Those aren't present values though! You're a better economist than that...
ReplyDeleteDiscount the cash flows at a sensible rate, and then do the comparison. Otherwise you'll end up looking like Megan McArdle. (I bet that's the worst thing anyone's ever said to you on this blog, sorry).
Oh sorry, I though donpaskini was d-squared.
ReplyDeleteI have to say, I'm slightly gob-smacked that the OECD is arguing about investment decisions over a 40 year horizon without discounting the cash flows.
Cash flows are discounted at a 3% interest rate, according to table A8.4
ReplyDeleteHere - http://www.oecd.org/document/52/0,3343,en_2649_39263238_45897844_1_1_1_1,00.html
ReplyDeletethen click on Indicator 8, and it is table 8.4
I was wrong. Apologies to everyone. Hm, now I have to completely reverse myself.
ReplyDeleteIn a boiled-down and much cruder version here's what the OECD study says:
Across all OECD countries, lifetime earnings differential between those with tertiary education and those without is about 1.5x (this is the key assumption by the way).
They use a 3% real discount rate, which I can just about stomach, since it's plausibly the real long-run cost of sovereign debt.
So, in the UK if you assume average non-grad annual earnings of £25k, that gives £37.5k for grads.
Quick and dirty, at a marginal tax rate of 30%, that gives an annual incremental benefit to the exchequer of £3.75k (=37.5 - 25) * 30%, or £150k over 40 years.
At a 3% discount rate that's £87k. Assume £4k per year taxes foregone for three years as opportunity cost to the exchequer, which have a present value of £11k, and you have a gross benefit with a present value of about £75k.
That's before the cost of education. But even at £20k/year tuition + board per year it would be worth the state paying for 100% costs.
The absolutely key sensitivity is the 1.5x differential. If you take that down to 1.4, and hold everything else constant, then the public value before education costs falls to £58k.
I don't have time in my lunch hour to check the OECD methodology on that point. But one would want to be very confident that the econometrics were sound.
As a vague face-saving gesture, I still do wonder--the private benefits to graduates are much greater than the public ones to the exchequer. Isn't that *some* justification for fees of some kind?
Admitting you are wrong? On the internet?? For shame...
ReplyDelete"As a vague face-saving gesture, I still do wonder--the private benefits to graduates are much greater than the public ones to the exchequer. Isn't that *some* justification for fees of some kind?"
Yes indeed:
http://inequalitiesblog.wordpress.com/2010/09/27/why-progressives-in-europe-should-learn-to-love-high-tuition-fees/