Sunday, October 23, 2011

Performance Funding: Easier said than done

Recently, HEA released consultation document about the implementation of the National Strategy for Higher Education to 2030 (The Hunt Report). It's worth a read, notably Appendix C, is a consultation paper the Strategic Dialogue process that will lead to performance related funding. The fact that we need a consultation paper about a dialogue leading to performance funding tells a lot about what a gnarly problem performance funding is. I think it's worth taking a little time to unpack some of the assumptions implicit in performance related funding models, and how one might go about implementing them successfully and how they might achieve their intended purpose (not the same thing!).

The basic idea of a performance related funding component is that there will be metrics, they will be wise and good, and that some proportion of funding will be contingent on doing well with your metrics.

The first implicit assumption here is that there will be winners and losers. Outfits that do well will be rewarded with more funding and will probably do better, having more money to play with. Institutions that do badly will lose funding, and, in time, fail. The percentage of funding we make performance based determines how many funding cycles the process takes. Set it to 100%, and we are done in one round. At the other extreme, as the proportion of performance linked funding approaches zero, the time taken for it to have an effect approaches infinity. This suggests an obvious path to painless implementation - set the percentage trivially low. Thus the framework can be successfully implemented and victory declared without actually needing to face any of the policy implications of the framework actually working as intended.

The second implicit assumption is that a market based model, where winners and losers are driven by student choice, either cannot work or should not be allowed to work. This is certainly the case in Ireland, where price fixing (at zero, plus registration fee and living costs) and an excess of demand over supply (almost all courses fill) basically neuters any consequential market competition between institutions. Your courses will fill and your monies will come in, really, no matter what you do as an institution.

Implementing a performance component to funding essentially tries to replace market style incentives with KPI driven ones, where how well you do on the metrics will determine how much money you get. We're keeping the buffet where students choose what course they wish with negligible difference in cost to them between a plate of chips and a plate of salad, but the state will pay more if they eat salad.

The third assumption is that there exists a set of wise and useful metrics to which we can all agree. The HEA paper is very sensitive to this point, and the hazards of chasing single KPI, University Rankings and so forth are well understood. Most academics either scoff the whole idea of metrics at all ("My contribution is utterly unquantifiable!") or, over in the sciences, fall to fighting over which metrics to use. The experience of research funding, which is driven by metrics on papers, patents and so forth, is not terribly encouraging. The choice of metrics strikes at the heart of complex issues around the purpose of the University. Utilitarians like me will suggest employment outcomes and graduate earnings compared to a propensity matched non graduate control group, and will be promptly heckled by holistic types who will argue the broader value of education in democratic society, knowledge as an end in itself, and so forth.

This suggests another easy path to implementation - pick so many metrics that everyone's a winner. Cook up a basket of 100 metrics, and let each institution pick three. They improve on them for a few years, win a biscuit, and then move on to another set once the low hanging fruit in that area have been plucked. Everyone's a winner! This might actually improve some things due to the Hawthorne effect, if anything else, and will give everyone something to do as institutions jiggle about winning performance funding year after year for something or other. Ministers can declare great progress is being made, and only the most cynical of observers (i.e. me) might suggest the process is just brownian motion, like a school sports day for junior infants except no one must suffer the indignity of falling over in the sack race.

There is no good solution to the metrics problem. You can avoid the problem by allowing a free market as in the US, or by funding everyone equally and blindly, as now in Ireland. Those approaches bring their own problems which may well be harder than picking smart metrics. You just have to get institutions to come up with metrics that they care enough about to chase, and the funding agency cares enough about to fund, and hope for the best.

If you solve that problem and don't neuter the process by setting a trivial slice of funding to performance, you must put in place a system for eliminating institutions that fail. Not allowing weak institutions to fail makes a nonsense of the process and, as we are seeing in other sectors, comes at a high price in the long run. If you are not prepared to allow your biggest University to fail if it comes to it, and have no plan in place to deal with that eventuality, then performance related funding is not for you.

And yet, perhaps there are smart ways to reduce the risk of institutional failure while maintaining the incentive effect of performance funding. Applying the performance funding to the total grant is a crude mechanism. A neater way would be to apply it to the payroll component only. If your organisation hit's it targets, everyone gets an end of year bonus, from the President to the lowliest postdoc. After all, you incentivise people, not organisations. Give the actual people an incentive to succeed without permanently tipping the playing field by obliging weaker organisations to find structural cutbacks or allowing strong ones to gold plate themselves. Instead of weakening already weak organisations by cutting their global funding, you provide strong internal incentives for them to boot their management team, and back that up by providing failing institutions with support from a tiger team who can roll in and help them find out why they failed, and remedy that.

Nothing around the implementation of performance related funding for higher education is going to be simple or easy. Maybe what's needed is a framework around the consultation paper for the strategic dialogue leading to an implementation strategy for waiting until it's the next governments problem...

No comments:

Post a Comment