The challenges of calculating the cost of social problems
22 September 2011
Riots. Homelessness. Abuse. Underage drinking. The list goes on. Its really hard at times to ignore all that is wrong in today’s society. But the more you are confronted by these issues, the more you understand how many different problems there are out there and how complex their various causes and consequences are. Which is a problem if you work at a place like NPC where we try to analyse social problems and their solutions in an empirical and rational way.
This is why our research for Barclays Wealth was both extremely interesting but extremely challenging. The final aim of this research was to prioritise a list of interventions that, when invested in, will save the highest amount of money for society. The obvious starting point therefore was to find out what society’s costliest problems are. I was tasked with phase one of this research where we started with the thirty biggest social problems in the UK and prioritised what we considered to be the six most serious, based on the numbers of people affected and their economic cost.
How did we to do this? The answer is—with difficulty. In an ideal world all costs would be comparable and it would be easy to see which were the most expensive. But this is not the case—especially when it comes to social issues. This meant that we had to overcome a number of methodological issues at the first stage of research:
- We wanted to look at both individual and government costs—it became apparent that it would not be feasible to find cost figures in such detail.
- It was necessary to prioritise based on the best and most comparable figures we had: scale and larger society costs.
- A lot of costs had been calculated using different methodologies and so true comparison was not possible.
- Costs tended to be skewed to existing government funding priorities.
- Most costs presented were current costs, although some were presented as life-time costs, so comparability mixed.
All this meant that we prioritised issues based both on costs and on how much we trusted the quality of the data. Admittedly, this was an imperfect process, but still allowed us to get an approximate idea of the most expensive problems.
Many of the major challenges we faced in this research were to do with the availability and quality of data, an issue which has come up again and again on this blog. But I think the most interesting lesson from this process is that taking an economic approach to understanding which problems to fund can be done – even with these imperfections.