Implicit aid contracts

I thought this was a very interesting and readable paper about aid effectiveness by François Bourguignon and Mark Sundberg.   They describe an emerging consensus for the aid architecture, based on what they call implicit aid contracts: The use of implicit aid contracts, based on monitorable evidence of improvement in Read more…

The role of prizes in innovation

One of the most important and cost effective ways that rich countries can help poor countries is to invest more in R&D, especially in products that would benefit the poor (such as a malaria vaccine, cheap solar panels, or a cassava plant resistant to mosaic virus).   We do have large research programmes; and I took part yesterday in an interesting discussion about whether we should fund such research by paying the researchers directly, or whether we should create financial incentives for the private sector to invest its own money in looking for solutions.

My conclusion is that we need a judicious combination of "push" funding (e.g. subsidies to research institutes) and "pull" funding (e.g. guaranteed markets, or prizes).  Push funding is especially useful for R&D that produces basic science of general value which cannot be mainly appropriated by a firm through the sale of products; pull mechanisms are more efficient for R&D that is specific to a particular product.

David Wessell has an interesting article in the Wall Street Journal on January 25th the use of prizes to spur innovation.  I thought this was particularly interesting:

One surprise: The further the problem was from a solver's expertise, the more likely he or she was to solve it. It turns out that outsiders look through a completely different lens. Toxicologists were stumped by the significance of pathology observed in a study; within weeks after broadcasting it, a Ph.D. in crystallography offered a solution that hadn't occurred to them.

One of the merits of prizes over government-directed research is that they encourage engagement by a more diverse range of investigators than would be likely to be supported by cautious and risk averse bureaucrats.  This finding suggests that might be rather important.

Full text of WSJ article below the fold.

(more…)

Are Republicans good for the world’s poor?

Many progressives here in the UK have a stereotyped view of US politics (roughly speaking: 'Democrats good, Republicans bad').  These assumptions have been reinforced by negative perceptions of the Bush presidency.  And so there is an assumption that the Democrats are more likely to pursue policies that are good for Read more…

Do we learn good?

 Andrew Jack wrote in the Financial Times on Wednesday about measures to reduce AIDS in Africa.  In a nutshell, he says that there has been too little support for campaigns to change sexual behaviour, even though these efforts appear to have worked in Uganda, Zimbabwe and Kenya.  By contrast, there has been more investment in drugs to treat AIDS and condoms to help prevent it.

Purnima Mane, director of policy evidence and partnerships at UNAids, the United Nations' Aids agency, says: "In the last 10 years, the focus has been highly biomedical. Social scientists have withdrawn a bit. They have been seen as playing second fiddle."  Furthermore, when programmes have been launched, monitoring and evaluation has often been lacking. "We have launched boutique projects rather than scaling things up," she says. "We don't establish what makes them effective. That has been the tragedy, because we have wanted to act quickly."

I drew several conclusions from this:

  • donors do tend to invest more in development programmes for which there is statistically robust evidence of effectiveness, such as clinical trials which demonstrate effectiveness by comparing people who receive the drug with otherwise similar people who don't
  • we need to invest more in rigorous impact evaluations of social programmes (e.g. behaviour change campaigns), preferably using random assignment trials wherever this is ethical, practical and cost-effective
  • the value that we attach to acting quickly, with the minimum of delay and bureaucracy, to get programmes up and running, militates against embedding rigorous statistical evaluation into new programmes
  • the benefits of developing having better information about the effectiveness of social programmes are long term and global, and the costs and delays are mainly borne by the individual donor and the community in which the programme is being conducted; as a result, there are insufficient incentives to do this kind of analysis.

The Center for Global Development report on the Evaluation Gap looks at these issues in more detail.  Their conclusion – with which I agree – is that we need a coordinated international effort to increase statistically rigorous evaluation of social programmes in development.

This all sounds very dry and technical.  But read the article in the FT: if we had better, more rigorous evaluation evidence we might have done better at reducing the AIDS epidemic in Africa.  And that would be quite something.  (more…)

Limiting the risks of government data sharing

The UK Government is going to consult more widely on its proposals for data sharing within government. A national identity register that allows data sharing across government could be the technological underpinning of a huge improvement in the provision of government services.  (It is important that the technology will not Read more…