Among the zombie ideas refuted in my book, Zombie Economics, “trickle down” economics is the one that dare not speak its name. Even those who believe, or are paid to say, that favored treatment for the rich will benefit the poor mostly avoid the term “trickle down”, preferring bromides like “a rising tide lift all boats”.
But that didn’t deter Ian Young, Vice-Chancellor of ANU and head of the Group of 8 Universities (basically, those established first, which have, as elsewhere in the world, gained a permanent high-status position as a result). As I predicted not long ago, he wants to raise fees and reduce the number of students at elite universities, including ANU, allowing them to offer a more personalised education.
Young’s argument is that students excluded from the Go8 will “trickle down” to lower-status universities, giving them a chance to both increase numbers and raise standards. But this suggestion doesn’t stand up to the most cursory examination. Both logic and historical evidence suggests that all or most universities will follow the lead of the Go8. In both the UK and Australia, whenever universities have been given option to increase fees or hold them steady, nearly all have gone for the maximum increase.
Think about this from the position of a university in the tiers immediately below the Go8 in the prestige hierarchy, the 1970-vintage unis like Griffith and Macquarie, and the Universities of Technology. Both groups can fill all the places they have, and both, like all Australian universities are straining at the seams in terms of both physical space and overloaded staff. They could not possibly take in more students with their current finances. It makes perfect sense for them to do the same as the Go8, raise fees a lot, and pass on some of the benefits in the form of smaller classes.
There’s a cumulative effect here. Suppose the Go8 institutions reduce their student intakes by 30 per cent. A few of those will give up on uni altogether, deterred by higher fees, but most will try a second-tier uni, displacing other students who would otherwise have been accepted. On top of that, there will be less places in those uni, say another 30 per cent. So, something like 60 per cent of the students formerly admitted to these unis will be excluded.
At the bottom of the status scale, the hard-pressed regional universities and former CAEs probably won’t be able to raise their fees as much as the Go8. But they will still be in a position to raise fees and entry standards at the same time, and, if they choose, to reduce their numbers as well. This isn’t so much trickle down as a cascade effect.
Of course, if you believe the increasingly silly Business Council of Australia, this is all to the good. Its head, Catherine Livingstone (BA, Macquarie) thinks we need less university students. Her members clearly don’t agree, judging by their hiring patterns. The unemployment rate for university graduates is estimated at 3.3 per cent, about half that for non-graduates. Wages and participation rates are also higher.
John Quiggin 
I think you are forgetting that education has other values as well as it’s training for jobs. Surely it is better if we live in a society with advanced technology and administrative complexity etc to educate more/all people to a level beyond what is taught at a usual state secondary school? Since we currently are overproducing/consuming material goods then education is a useful public good not requiring too many resources (although more now than it used to).
@John Quiggin
A 15-year-old arriving in my workplace would have to be shown how to do things, but then the same applies to a university graduate. How many 15-year-olds would be capable of acquiring the necessary competence I don’t know, but then again the same applies to university graduates. Whether my workplace could function as effectively on the basis of recruiting 15-year-olds as on the basis of recruiting university graduates I don’t know. I’m not sure how the question could be settled, since I know we’re not going to try the experiment of recruiting 15-year-olds. (I don’t know, either, to what extent we do specifically recruit university graduates — I’m not in a position to know whether all my co-workers are university graduates.)
Ten storeys up, I run the joint/ You man the firm’s front door/ We’re nicely placed to make a point/ Of economic law/ For, if I choose to urinate/ And if my aim is true/ What I despatch, will very shortly/ Fall, alas, on you/ Which illustrates, if I’m correct/ The so-called ‘trickle-down effect’.
@John Quiggin
Well, at the very least there is cross subsidisation between different units at universities. Big first year units tend to be cheap to teach, and are used to subsidise other units.
@Uncle Milton
At least in Aus, Open Universities is basically an education broker, where they group courses run by universities (who generally run them) and then resell them.
Apart from this, there is some strange belief by many people that online courses are a lot cheaper to run than non-online courses. This is not true for many courses (probably most) if you want them to be of the same standard as non-online ones. Indeed, many online courses would be more expensive to run, and the reason they arn’t is because everyone cuts corners with them (and they will hence get a bad reputation given that many places are piling into them). This is because you still need to create the course, create assessment, have people to run it and so on, and trying to do this for things like activity based learning is expensive, and indeed very difficult (almost impossible) for most universities still trying run things through web technology suitable for 1992 like Blackboard. So you are just stuck with courses where people are supposed to “discuss” things you really need to do to learn and the actual way the material is presented is often very poor.
To various commenters who prefer anecdotes to data when they assert that the gains from education are all credentialism (ie screening and signalling effects), can I suggest you try having a look at just a couple of the many systematic studies by economists of exactly this question in the last four decades? Go looking at the bibliography in John’s linked paper, or for a more ‘big picture’ but cliometric approach try Katz and Goldin’s “The Race Between Education and Technology”.
Yes, commonsense suggests that it’s mostly credentialist screening, and that’s actually what quite a few of the older studies originally set out to prove. But to many people’s surprise the hard data says that in this case commonsense is mostly (not entirely) wrong. And the reason it is wrong is precisely what John said – because work has changed in ways that draw on skills got by education.
Now of course one of the things that has driven those changes in the nature of work is the availability of educated people, but that doesn’t change the fact that productive workplaces mostly now need pre-educated people.
It’s not an untried experiment, it’s a policy with a proven track record of failure (it worked 100 years ago, to be sure, but it was already failing 50 years ago). That’s why employers have universally abandoned it,. The exceptions are the fast food industry and similar, who design work processes specifically to take advantage of the availability of young people willing to work part-time, for low wages, and with no expectation of, or desire for, a career path.
First off, I for one never asserted that, but rather I hypothesised it. That’s where anecdotal evidence most often really is helpful – in raising questions rather than answering them. By and large it should only be used elsewhere – carefully – if no better data is available.
But here’s the thing. Better data is not available. Look again at the hypothesis and at the data sets you just suggested:-
– Hypothesis: that there may come a point beyond which there is no net overall gain from having more graduates around and maybe even from having more people with a secondary education around, so that the gains to people like that are generating even more costs elsewhere, and that we may have reached such a point here in Australia in the last decade or two.
– Data suggested: “the many systematic studies by economists of exactly this question in the last four [emphasis added] decades”.
By the very nature of the beast, many and perhaps even most of those studies cannot test the hypothesis, as they were necessarily studying data to which the hypothesis did not apply – anything three or four decades ago wouldn’t pick up any effect that hit later. Of the rest of the studies, they are necessarily of limited value because – again, by the very nature of the beast – a full test could only show really solid data after enough time to track most of its subjects’ careers, i.e. rather longer than we have had. And that’s before we even ask if those studies really were testing for those overall gains, rather than merely gains to those with those educations which would not pick up any issues of externalities and the like (e.g., educations that were free at point of sale threw their costs on to the rest of the tax base).
But that comes back to my point about that very situation itself possibly being a distortion (according to the outworkings of the hypothesis above), which in turn means that if the hypothesis were true and corrective action were taken, not only would there be a downwards shift in supply of people with that sort of education but also there would be a downwards shift in demand for them, in such a way that once everything settled out there would be an overall improvement. We can’t simply pray in aid one aspect of the (possibly) distorted situation as a reason for perpetuating that very situation. Of course, that raises the questions of trajectories away from the current situation that would have to avoid dislocations, dislocations that kept some of it in a way that would act as a poison pill, and of how to redress the inequities of having some educated people getting disproportionately more than other people who weren’t as educated but who might as well have been – but those are follow on questions, and do not relate to whether the hypothesis is correct or not. That is indeed a question to be settled by data, so on the one hand we should seek it as we don’t have it, and on the other hand we may have to try out policies even with imperfect information – and we cannot avoid that by pursuing even current levels of education, let alone more, precisely because we do not have solid data after all.