10 March 2011

CAREless Use of Data

CARE is holding a conference down in DC to celebrate International Women's Day and the 65th aniversary of the organization. With speakers like Melinda French Gates and Laura Bush, there has been plenty to share via twitter. However, it was this tweet that caught my attention.

I immediately requested that they show the data. Having seen previous claims made which are blatantly false (ie. $20 = 20 years of clean water for a person), I was immediately skeptical (what's new?). After doing a little digging, I found that the statistic is a favorite for a bunch of other organizations.

Overseas Development Institute
Finally, the cost-effectiveness of hazard warning and prevention is only too clear. The World Bank has estimated that every dollar spent on risk reduction saves $7 in relief and repairs. Prevention now needs greater emphasis.
Relief Web (using UN data)
The United Nations says $1 invested in reducing the risk of disasters in developing countries saves around $7 in losses, and it costs $1 to feed a child compared to the $80 needed to save a starving child's life.
National Institute on Drug Abuse
Every dollar spent on drug and alcohol abuse treatment saves the public $7, largely through reduced crime, according to a study of the cost-effectiveness of California's substance abuse treatment programs. The University of Chicago's National Opinion Research Center, which conducted the study from 1991 to 1992, called it the largest and most scientifically rigorous cost-benefit analysis of drug abuse treatment ever undertaken.
Maybe I missed a study, but it looks like there is some lazy application of data going on here.  CARE seems to be running with a static and applying it to development in general.  At best, it seems like one dollar invested in disaster prevention will prevent large expenditures post-disaster.  To me, that is pretty specific and would not apply to something like maternal care or sanitation.

It is likely true that early interventions are more cost effective than later fixes.  Why not compare it to an investment verses an expenditure.  If you make the investment in advance, it will bring later returns to pay for times in need.  If you choose to wait, you will have to pay and have nothing left over.

OK, that is not a great analogy, anyone have a suggestion?