Thursday 13 May 2010

Evaluation and decentralisation

After spending yesterday reading through some of the extensive material on RDA evaluations I suggested that I wasn't much clearer on the impact of RDAs. But it did get me thinking on the relationship between evaluation and decentralisation.

I confess to knowing very little about the history of the RDA evaluation exercise, but I was interested to read from the national audit office that (with the notable exception of EMDA) an assessment in 2006 found: "impact evaluation to be one of the weakest elements of their performance". BIS and the RDAs responded to this by drawing up the framework that I discussed yesterday and RDAs were then asked to use that framework to evaluate properly impacts by 2008. A follow up review in 2006 found that out of 400 projects "only in a very few cases had RDAs used a robust methodology to forecast or measure outputs and outcomes". By December 2007 external "consultants found that only about 40 per cent of [evaluations] were sufficiently compliant with the Framework". With further intervention 70% of projects had been covered by the time of the overall evaluation I discussed yesterday. Subsequently BIS has published further guidance to help make evaluations more uniform in future.

The broader question that all of this raises for the new government is what kind of frameworks for evaluation to put in place when decentralising powers to subnational organisations (LAs, RDAs - or whatever replaces them). The RDA experience would suggest that this is one area where systematising practice across organisations has big payoffs. Despite these obvious benefits, Local organisations are often hostile to this process because it appears to trample on local autonomy. Whatever changes to delivery the new government is planning it will be interesting to see how they try to square this particular circle.