Lesson of the day: don’t say things to journalists relying on memory alone. Yesterday I spoke to a reporter from The Australian, and during the conversation agreed with a claim from University of Melbourne higher education expert Simon Marginson that the federal government spends more on private schools than on universities. That was printed in this morning’s paper.
I had some time ago looked into this oft-repeated claim by the public education lobby and decided that it was defensible but a factoid (meaning 2). It was defensible because if you count only direct subsidies related to tuition then more is spent on private schools than on universities (approximately $5 billion compared to approximately $3.5 billion in 2005). But I deemed it a factoid because private school funding supports more than twice as many students. It is less than surprising that about 1.1 million school students cost more than about 450,000 university students (full-time equivalent).
But I’d forgotten that the university income number did not include research and other grants to universities, which take the spend up closer to $6 billion, or the significant contribution to university cash flow made by student loans, which add in another $2 billion.
So if we looked at total support for universities it is significantly higher than total support for private schools.
In my defence I did dismiss the value of the comparison (which wasn’t reported), and note that there had been a significant recent increase in university funding (which was reported), and the original purpose of the conversation had been to discuss something else entirely, for which I had the relevant spreadsheets open when I returned the call. But I should have said nothing or at least fact-checked myself afterwards. Someone did question my evidence today, and they were right to do so.
Andrew Norton 
If you are going to make the comparison between universities and schools, then you should probably leave the reasearch funds out anyway. While the comparison does not make any sense either way, it makes even less sense to compare education and research subsidies post secondary with education subsidies at the secondary level to only private schools. Besides which, most of the universities are public anyway. Why would anyone care about a comparison between funding to public universities and funding to private schools? As an aside, you should probably only include the interest subsidy on the loans in the government contribution, though. The rest of the money from the loans really comes from the students.
LikeLike
Damien – I classified the loans as generally ‘support’ without, in the most part as you point out, being subsidy. But arguably the loans give the universities a bigger market than they would have had if students had to rely on gifts, savings, or private-sector loans.
LikeLike
I’m not sure you should be sticking research money into the equation either — a lot of that money is potentially contestable by people not in universities. A fair bit of free work is also done by universities such that money can be distributed (obviously there’s conflict of interest there — but no-one is thinking of a better way). Even the real price of grant reviews and the like would be huge if people started to charge a real rate for it.
LikeLike
Incidentally, looking in today’s paper at your comment (with no context) its not at all clear what you mean by R & D not needing to stand on universities. I assume you mean here that R & D can happily by done in private companies, versus no universities need to do research.
LikeLike
Are Marginson’s numbers in the public domain? They look to be on the low side. According to the ABS in 2005-6 the Feds spent $5.8 billion on universties not $3.5 billion. They spent a further $1.5 billion on TAFE. That compares to $7.8 billion on schools. About 2/3 of commonwealth school funding goes to private schools and 1/3 to public schools. So it looks like the commonwealth spends more on universites than it does on private schools.
Conrad, I think Andrew is suggesting that if the prosperity of the Australian economy relied on university research, then we’d all be in trouble. That, however, is a common misconception. The ARC commissioned research that showed a 50% return to publicly funded research (presumably that’s per annum). I true, we’d invest our entire GDP via the ARC, yet somehow I’m not convinced.
Andrew, you neglected to include corporate tax in your comparisons. Universities don’t pay corporate tax, and I understand are exempt from a raft of other taxes too.
LikeLike
Andrew, why are we looking only at the federal government here? Isn’t this as silly as the hoary old comparison of what the federal government gives to public vs private schools? (I appreciate that the state government contribution to universities and private schools is relatively small, but I still think the comparison is worth getting right.)
Incidentally, had you come out directly critiquing Marginson, presumably you may have had difficulty with one of your hats.
LikeLike
States spent $353 million on universities (and, I think, exempt unis from land tax). They spend $4.5 billion on TAFE.
LikeLike
Andrew, as a friend, I won’t try to make you feel better by saying that research funding should not have been included – it should have for the sake of the comparison they made. But as you say, what in the world does the comparison mean? Frankly, I think it’s a good thing that the Government subsidises school choice and limits tertiary education funding (so long as they deregulate the sector of course). I’m also with you that the holiness of R&D generally is overrated. The very fact that there are so many free rider problems with ‘basic research’ is a good reason to not subsidise it at all in my view. Why can’t we just free ride on the basic research of other countries? To the extent there are positive externalities from simply having these basic researchers hanging around university cloisters, are they greater than having more accountants or doctors in the community?
LikeLike
RS. I can’t imagine where universities would actually find employees if they didn’t give time to a reasonable percentage of their full-time teaching staff to do research — already the pay is lower in many professions and conditions presumably worse for your average guy that can speak English (like engineering). It is also rather easy to imagine that without employees that are actually up to date with anything, even more lip-synching textbooks from the 60s is going to going on versus now.
This to some extent is also an answer to RS — basic research forces staff to remain up to date, and means you actually find staff to (at least in some areas). Another advantage is that whenever new areas happen to spring up (like IT), you actually have people that can apply their basic knowledge to understand stuff and train other people. Historically, this how many new developments often occur (Computers, for example. Biotechnology is another — by the time you can buy the stuff to do applied research easily, you are way behind).
Note that I think its a good question as to how much you should free-ride on other countries (its a really a question you could ask of many areas, not just science), but this seems to me a question that most science people don’t have much time for. Thanks to global mobility of science professions, lots of people simply move to the place where the best pay/facilities are (which has often been the US in the last 50 years), and this appears to benefit both applied and basic science — I don’t think they are really perfectly dissociable. I can’t think of anywhere that doesn’t do basic science but is still good at applied science.
LikeLike
sorry, that should say an answer to your question, versus an answer to RS
LikeLike
conrad, thanks for your considered response. In relation to finding staff, I see this as really a derivative matter. If there is, say, no case to subsidise research, the market will determine who and how universities hire. If this is reflected in less smart teachers, that is simply a reflection of students’ willingness to pay.
On your second point, the issue is whether you need researchers to disseminate the work of other researchers. Given that applying knowledge to “understand stuff and train others” does not seem to suffer more from free-riding/market failure than most other activities, I again question whether this needs to be subsidised.
LikeLike
Sinclair – The $3.5 billion is my figure and derived from the Commonwealth Grant Scheme and a few other grants, but not research-only funding from DEST or funding derived from other Commonwealth departments.
Andrew L – Of course only looking at federal funding is an old AEU misinformation tactic, but in this case (unlike the AEU case) it is not material to the point being made as virtually all state funding is for research and capital works while state funding for private schools would widen the tuition income gap. I’m not worried about directly attacking Simon; it is only commentary on the U of M as a corporate entity that I feel conflicted about and therefore say very little.
Conrad and others – The only point I was making was that university R&D makes a minimal contribution to economic growth. The research commercialisation report that came out last week again showed how trivial these activities are, and the ABS business innovation reports confirm that negligible amounts are derived from universities. Even if that ARC 50% return figure was true, which I don’t for a moment believe, we are looking at $1 to $1.5 billion after taking into account opportunity costs, which is about 10% of the economic expansion of the last 12 months. As you suggest, there may be other reasons for spending on research at universities, but economic growth is not in my view a compelling justification.
LikeLike
Rajat,
(a) due to current Stalinesc regulation and perhaps social pressure, you can’t assume that market forces would/would not pay for research. If Melbourne Uni could charge US style fees (which it possibly could), I’m sure they could fund a lot more of their own research. However, if they started doing this, it isn’t hard to imagine the government would legislate against them so they couldn’t (just look at the current level of complaints even for very minor moves into realistic fee paying levels). I’m not sure of the answer to this, but as far as I see it, if you want good universities (or perhaps adequate universities) with such legislation, I don’t see what the alternative to funding research is. If you can get rid of this legislation (and get rid of it for good — i.e., not be under threat of having it re-introduced), then there is clearly a different story.
(b) the trade-off between market forces from private enterprise and the government can lead to different outcomes for the two groups. There may be no real private market for students/specialists in some areas (say, audiology — I doubt the average person even knows what this is until they have the need for one). However, it may be in the governments interest to have research and keep people in these areas (it keeps people off the dole and governement benefits — deaf people don’t make good employees in many areas — spending a little by the government may save it money in the long term). A lot of left-wing people would argue that this idea applies to a lot of the education system. I personally think some areas could be happily self funding even accounting for research costs (business and law, for example), and adequately supply the market with graduates.
(c) Its hard to work out what you should consider subsidies and what you should not. At present, there is a hybrid of tenders and free-money. However, if you consider the free-money case as the government buying ideas, then it isn’t neccesarily a subsidy. For example, if you want a good mental system, you can either get a whole bunch of dull beuracrats to think of what is important, or you basically offer the money to those that think of the ideas (thats how the ARC and NH&MRC work). Its clear to me that it is and has historically been people from the university sector that are thinking up the good ideas (some of the tenders I have seen are so aweful that no-one even takes them. The stuff that the bueracrats in the education deparment come up with sometimes also horrifies me). If you believe this, then the number you are seeing should be made smaller to account for this (I’m not sure how much smaller — obviously some things are subsidies but others are buying ideas).
LikeLike
Are the opportunity costs of the land housing schools not being used for other purposes included in the costings?
LikeLike
(And universities etc.)
conrad, may I ask what work you do and what your background is? I’m curious.
LikeLike
Sacha – There have been attempts to estimate the cost of land for government schools, such as in the AISV report I blogged on a few weeks ago. It was about $600 per student per year. I haven’t seen any equivalent estimates for universities. But since the federal government doesn’t own universities or private schools, I’m not sure that the issue affects this debate.
LikeLike
Hi Sacha,
I work in the university system, and have done so in 3 countries. I do some basic research, as well as work on collobarative applied projects with the French government every now and then (or sometimes just schools in Australia — although I have given up on this, due to the pain and suffering of having to organize it). I look at things mainly to do with language and literacy — e.g., how do you read — including mathematical models of the processes that go on, what can go wrong (develpmental and acquired), and what can we do to help you (including trying to automate some of the processes that currently use a lot of man-power — so you can potentially help more people). At least in my area, I don’t think you can do good applied projects without a good idea of the basic research (my comment to RS). You probably want someone good at statistics too, which many people in the area are not (you can be excused from that!). This is probably why organizations that rely fully on people that do no basic research end up with such aweful results (some of the Dep. of Educations around are a good example — they have curriculums designed on what we knew from the 60s, which is different to now — this just shows they can’t even evaluate basic research, nor keep even slightly up to date with what goes on. Perhaps you ACER people should be telling them off).
I don’t think I’m too biased in terms of my comments — I have a decent backgroung in computing, maths and statistics, so it wouldn’t be too hard for me to work as a data analyst or something like it if I finally get too frustrated with the system here (in fact, I’d probably just move to an OS university, which wouldn’t be very hard for me either — I often get offers when I got conferences — there’s a fair bit of competition for employees in a large number of areas now).
LikeLike
Conrad, in view of your area of expertise, can you suggest what the teachers should be doing to increase the level of literacy among school leavers? Or is it ok and it is just a beat-up that we have functional illiterates in substantial numbers coming out of the high school system?
LikeLike
Rafe,
I guess it depends whether you want to look at social problems, literacy problems (the two are of course well correlated), or reading problems by themselves. I have no really firm opinions about how the first can solved (its probably unsolvable in at least the medium term in some places I have worked).
As for reading problems not caused by language problems, I think the biggest problems occur early on, and if you don’t pick them up at this stage, then your problems just get compounded (i.e., you should be looking at kids less than 7, which is a problem if your average primary school teacher isn’t bright enough to notice). As for language problems in general (which inevitably cause reading problems), you probably want to start even younger. Again, it isn’t always obvious that some kids have problems unless you specifically look for them. If you go to some of the small number of centres that exist for such kids in Australia, you often see kids who are over 9 years olds — so its clear you can go for years with such problems not being noticed.
In terms of what should be done, you can certainly get great overall gains by having a really good curriculum, which should get around some of the problems caused by bad teachers (but obviously won’t pick up kids who still don’t learn well with a good curriculum). Unfortunately, it is not the case that such things have been generally well developed in Australia.
Given this view, I think that a lot of the debate has been mainly political and idiotic (we don’t like book X interpreted in terms of communist theory). If a 16 year old can write an essay on some strange topic, I’m sure they’re fine. If a 16 year old is still illiterate, then the problem obviously occured before hand.
The guy you want to believe when it gets onto the television from time to time is Max Coltheart from Macquarie U and the education professor they have there (i’m a bit biased here — I worked with him some years ago). When he calls some of the other guys who are responsible for the education curriculum and education teaching in Australia a bunch of anti-science idiots, he’s correct — and its coming from an introverted and extremely polite guy. Its not just an off the wall comment designed to score points.
LikeLike
Conrad, you should be a politician, you have skirted around the indelicate issue of the number or proportion of kids who emerge from the school system as functional illiterates. I think it is absurd that we can spend so much on school ed and have any significant degree of illiteracy at the end of it. But I guess you just about got to the point of saying that the lunatics are in charge of the asylum!
LikeLike
I guess a better way to look at it is to estimate what degree of illiteracy you should have based on the amount we spend, rather than wonder why there is any significant degree — you are always going to get some degree of illiteracy due to real dyslexia, selective language impairment due to neurological problems etc. . When looked at this way, its still depressing unfortunately (in my books anyway). A good curriculum across the school years should cost the same as what a bad curriculum does now. Its also worth noting that the way things are currently taught, its hard to teach people to improve at a higher level. I think about half my second year students don’t know what a verb is (thats true incidentally, it isn’t a joke — and they are social science students!) — so trying to explain how they should write some things is harder than it should be. I think none that have not learnt a second language properly (which is most of them) could tell me the real definition (versus _doing_ words).
Its also worthwhile noting that some of the changes were made simply to appease some groups, rather than for any decent reason — and not just for stuff to do with literacy. I currently work with someone who has been in the school system for probably longer than I have existed (or close anyway). I was wondering why a decent proportion of our 2nd years are not able to interpret really simple graphs (true!) — we have to painstakingly explain them for their assignments. Her response was that things to do with graphs were taken out of the curriculum in some areas because little girls had a lot of trouble understanding them — thus things were taken out for political correctness (little girls don’t do this well, and we couldn’t have little boys doing better than them on this), which is most probably one of the reasons boys now do worse on every subject. Its crazy. Why not just ask for social problems. Some of these things were accomodated simply by making the tests easier. The grammar tests now, for example, are much simpler than they used to be. (This was partially done to appease second language speakers and partially done because the overall standard declined.) You can ask Sacha as to how his ACER 2007 test he is creating differs to the ones from the 70s — a lot of the complex grammatical structures were simply eliminated.
LikeLike
I don’t know the answer to conrad’s question except to point out that the client has to agree to every question put in the tests.
LikeLike
That should be: “…a question appears in a test in a particular form if and only if the client agrees to that question appearing in the test in that form.”
LikeLike
As someone trained in mathematics, I found the mathematics school curriculum for one of the states to be very confusing. (Yes, I am not a teacher and don’t have that prism to interpret a maths curriculum, but it should at least be comprehensible.)
It was unclear and imprecise, e.g. one of the areas that wasto be taught at one level was “properties of polygons”. Which properties might these be? Certain particular properties of polygons were specified at higher levels but nothing much was specified at this lower level.
LikeLike
There was a competition in The New Statesman to design games for todays kids – I liked the new version of “I Spy” where you say the name of something you can see, and the others have to try and say what letter it begins with.
About illiteracy – I think parental attitude to education must be a factor.
LikeLike
We’re asssuming here that both languages are taught properly. In my experience the second language (Afrikaans) had both grammar and literature, while the first language (English) only had literature.
LikeLike