Rising DFID Spending hasn't Crowded Out Private Giving

Last week I was poking around the ESRC’s 'Administrative Data Research Network’ and discovered the Charity Commission data download website - containing every annual financial return made by every individual charity in England and Wales since 2007. The data comes in a slightly weird file format that I’d never heard of, but thankfully the NCVO have a very helpful guide and Python code for converting the data into .csv format (which was easy enough to use that I managed to figure out how to run without ever having really used Python). 

One obvious question you could ask with this data is whether the private income of international charities has dropped as DFID spending has gone up (more than doubled over the same period) - it is conceivable that people might decide that they could give less to international charity as more of their tax money is being distributed by DFID.

That does not seem to be the case at all. There are two ways of identifying international charities - by their stated area of operation, or by their stated objective category. I’ve coded charities that have no UK activities as “International”, and also picked out the charities that ticked the box for "Overseas Aid/Famine Relief” as their activity category. These two categories do overlap but far from perfectly. 

Charities have multiple categories of income - I focus here on the ‘voluntary’ category which basically means all donations, whether large or small. 

Charities with exclusively international activities, and those focused on 'overseas aid' did appear to take more of a hit than domestic charities from the 2008 global financial crisis and recession, but since then growth has tracked the income of other charities (and is 40-50% higher in 2015 than in 2007 (not adjusting for inflation)). 

You can download the Stata code here, csv files (large) here, and variable descriptions here.

UK teacher unions as legal insurance

Richard Murphy of the University of Texas confirms something that a teacher told me in person just last year - teachers in the UK only join unions because it provides legal insurance in the event of getting sued. 

This paper identifies the threat of accusation as a new source of demand for union representation and how this has increased union density in specific labour markets. Society has become increasingly litigious and this may have many repercussions on labour markets, especially those where employees have unsupervised interactions with vulnerable groups. A rational response to such changes would be an increase in demand for insurance against these risks. I model union membership as a form of private legal insurance, where the decision to join is partly determined by the perceived threat of having an allegation made against the agent. This is examined by estimating the demand for union membership amongst UK teachers, which has been increasing over the last twenty years. I use media coverage of allegations relating to local teachers as an exogenous shock to the perceived threat. I find that unionisation rates increase with media coverage of relevant litigation at the regional and national levels. Ten relevant news stories in a region increases the probability of union membership by 5 percentage points. Additionally, the size of the effect is dependent on the relevance of the story to the teacher. This paper provides a reason why the demand for union membership in this and related sectors has increased, despite the possibility of freeriding as pay and working conditions are set centrally.

Are insurance companies missing an opportunity here? Or would they face adverse selection issues?

The results agenda is yet to take hold in the UK

DFID Annual Budget: £10 billion

Current (domestic) UK Government "Major projects expenditure" with no plans to evaluate impact or value for money: £49 billion (NAO 2013: Evaluation in Government)