Tim Leunig analyses the new data on how changes in university fees have affected applications.
The data are out. We know how many people applied to each university, and how much that has changed since the previous year.
We also know how much each university is charging, how much they are spending on waivers, and how much they are spending on bursaries. My thanks to Jeevan Vasagar at the Guardian for matching the applications data to those for fees, waiver and bursaries.
We can therefore answer the following questions:
1) Did institutions that charged more see applications fall by more than those who charged less?
2) Did institutions that offered more waivers get more applicants?
3) Did institutions that offered more bursaries get more applicants?
The answers to all of these questions is NO. High fee institutions have not seen applications fall by more than lower fee applications. Universities offering lower waivers or bursaries have not seen applications fall than at more generous universities.
There are two ways to interpret this. The first is that students are confused, and failing to pick universities rationally. The second is that students know that the lifetime effect of picking the right course at the right university is much more important than the relatively small differences in fees or waivers or bursaries. The income contingent loan system gives them the confidence to go to a top fee institution offering a low support package if that is what is best for them.
The finding that bursaries have no effect on student choices fits with other evidence, notably the Corver report for OFFA, which the same thing.
These results come from simple OLS regression analysis. The left hand side variable is the change in applications, the right hand side variables are basic fee, waivers, and bursaries. The co-efficients are 0.00003, 0.000005 and -0.000004 respectively, with t-stats of 1.0, 0.06 and -0.06 respectively. The Adj R2 is -0.01. In short, the regression explains nothing.
Please read our comments policy before posting