Archive for the 'research' Category

What Works in Poverty Alleviation / A Review of Poor Economics

Much coverage has been given to MIT researchers Esther Duflo and Abhijit Banerjee’s new book, Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty. Banerjee and Duflo are well-known here at MIT for their use of randomized control experiments to test the means of poverty alleviation and their co-starting of the Jameel Poverty Action Lab (J-Pal).

The 273 pages of Poor Economics shares the results of their research – and starts to fill in the gap between economist Jeffrey Sachs’ (aid breaks the poverty trap!) and William Easterly’s (free markets and provide incentives, people will solve their own problems!) theories on development.

Banerjee and Duflo’s work raises oft-debated questions such as what really supports the use of bed nets in malaria infested areas, when is microfinance useful, and why, given the availability of education, are more individuals not receiving quality education? What is actually working?

There are many good intentions in development work (regardless of the location); being able to learn from and distinguish between good intentions and theories and the associated results will help us all push forward the next iterations on our work.

As writers Ramnath and Misra point out in Forbes India, “The approach is not without its critics [one example]. One relates to the danger of generalising [sic] the results (what social researchers call ‘external validity’; it questions whether what worked in one place will work in another). In an earlier interview, Banerjee said it was a serious concern. But some evidence is better than no evidence. Also, many such trials will lead to better policies.”

While we are all not able to commit to (or have the luxury to) conduct randomized control trials, we can learn from Duflo and Banerjee’s work. They offer five key lessons when working on poverty alleviation (summarized here and can be found starting on page 268):

  1. A lack of information often contributes to untrue beliefs.
  2. The poor bear responsibility for too many important decisions (such as whether to spend money on vaccinations).
  3. Markets are not always friendly to the poor.
  4. Many policies meant to help fall short because of the three Is: ignorance, ideology, and inertia.
  5. People live up to their expectations.

Duflo and Banerjee end on a practical note, inviting readers to use the book as an invitation (almost a challenge) to dig deeper. As they say,

If we resist the kind of lazy, formulaic thinking that reduces every problem to the same set of general principles; if we listen to poor people themselves and force ourselves to understand the logic of their choices; if we accept the possibility of error and subject every idea, including the most apparently commonsensical ones, to rigorous empirical testing, then we will be able  not only to construct a toolbox of effective policies but also to better understand why the poor live the way they do. Armed with this patient understanding, we can identify the poverty traps where they really area and know which tools we need to give the poor to help them get out of them. (page 272)

You can find more about the book here: http://www.pooreconomics.com.

Meet IDEAS10 Winners!

Thanks to summer intern Stephen Kaliski for this brisk introduction to some of the people and ideas behind this year’s winning IDEAS teams. Kudos all!

MIT IDEAS Participants from MIT Global Challenge on Vimeo.

Can Information Improve Aid Effectiveness?

In 2009 Oxfam America published a brief on “smart development,” in which the authors advocate for increased transparency and predictability in US overseas assistance (ODA). In defining the challenge and the opportunity, Oxfam America identifies three reforms central to their strategy: provide increased access to information to recipient countries; build capacity and help countries lead their own development; and finally let countries lead by opening control.

In making the case for reform 1 (increasing access to information), Oxfam America gives three examples of the information entanglements that arise from massive aid programs that lack effective information sharing regimes:

For years, Afghans have heard about billions of dollars being promised by foreign  donors, yet they have no way to find out where that money is going. Even their  government does not know how one-third of all aid (some $5 billion) has been  spent since 2001.13 In Uganda, a mapping exercise in 2005 found twice as much  aid being spent than what the government was told.14 In Sierra Leone, the government knows little of the 265 different aid projects that donors are funding.15 And
in Malawi, there was a $119 million difference in what donors reported they were  providing to the government of Malawi and what donors reported to the OECD.16

“How” the authors ask, “can recipient governments use donor aid to plan in such circumstances?”

In reading the report, which is a sleek 40 pages, I realize there is a tremendous innovation opportunity here – which is to develop better tools, standards, and administrative reforms that will lead to the kind of information coherence across ODA actors necessary for effective development planning. In a sense, a GPRA for aid that harnesses the open data reforms of the last decade and the flexibility of information sharing tools available on the web.

A problem with enough technical challenges to be of interest to MIT management and programming talent?

Lessons from America’s Giving Challenge

America's Giving Challenge

Download the America's Giving Challenge report from www.casefoundation.org

In 2007 the Case Foundation sponsored two distinct charity drives across two very different platforms, in part I suspect to change the top-down charity model to one of partnership and engagement and also to stress-test the emerging socialweb as a platform for carrying out its philanthropic interests.

In June they released a report on what they learned, titled America’s Giving Challenge: Assessment and Reflection Report. Written by socialmedia guru Beth Kanter – also one of the Challenge’s winners – and tech expert Allison Fine of Demos.org, its a pretty straightforward head-to-head comparison of the performance of both approaches.

On the one hand, Case launched America’s Giving Challenge with PARADE Magazine and through parade.com, an example of a mainstream publishing giant adapting to a new Internet reality. The second challenge, the Causes Giving Challenge, was run through Facebook, one of the most popular social networks on the web.

Both Challenges performed incredibly well, though in very different ways. As the report states, “America’s Giving Challenge raised $1,193,024 from 46,044 donors for 2,482 causes. The Causes Giving Challenge raised a total of $571,686 from 25,795 unique donors for 3,936 causes.” Continue reading ‘Lessons from America’s Giving Challenge’

McKinsey Report on Prizes and Innovation

"And the Winner Is..." (McKinsey, March 2009)McKinsey and Company recently put out a significant report on prizes as incentives to innovate which is quite good, and very accessible. The report, And the Winner Is, offers a view of the current landscape of prizes and competitions for innovation and provides good insight in good practice. To follow are some notes.

Prizes to incentivize innovation are going through a renaissance – they’re calling these “philanthropic prizes.” Interestingly, find that, “prizes are a unique and powerful tool that should be in the basic toolkit of many of today’s philanthropists.” Some benefits of the “prize inducement” model are:

  • Identify new levels of excellence
  • Encourage specific innovations
  • Change wider perceptions
  • Improve performance of communities of problem solvers
  • Build the skills of individuals
  • Mobilize new talent and capital

Some promising practices identified in the report include:

  • Philanthropist matching a clear goal with a large group of potential problem solvers who are willing to absorb some risk.
  • Start with a clearly define aspiration for society benefit which can be translated into prize objectives that are specific, motivational, actionable, results-focused and time bound.
  • A good prize will invest significant resources into its design, specifying the competitor pool, rules and award attributes.
  • An effective prize process is at least as important as the prize design, which will attract candidates, manage the competition, celebrate winners, and publicize the effort.
  • A good sponsor will invest significant resources in post-prize activities that convert the awards result into long-term societal benefits.

Continue reading ‘McKinsey Report on Prizes and Innovation’

Technology for Emerging Markets at Microsoft

On 4/3/09 Microsoft Cambridge hosted a seminar on Technology for Emerging Markets at Microsoft Research India (MRI), research.microsoft.com/india. 

Kentaro Tayama, the director of MRI, first gave an overview of the institution and the kind of research pursued. MRI is run like an academic institution in the sense that funding is given without any strings attached. Microsoft gets value out of it
 in at least three ways: MRI establishes useful contacts (in India this includes government contacts), it is good for company morale, and research aids corporate strategizing for long-term global growth. The keywords for MRI are immersion, design, evaluation, implementation.

Indra Medhi then presented her field research on the use of cell phones among low-level literate  people in rural India. She had measured the completion rate of the task of remitting money on cell phones with three different systems: text-based, voice-based, and rich multimedia based user interfaces. Not surprisingly, the text based UI left almost everyone floundering,  whereas about two thirds completed the task with the other two.

The ensuing discussion brought up a number of issues. The design of cell phones have to pay great  attention to sunlight/glare and to audibility in noisy environments. There are cognitive issues relating to the use of softkeys and the concept of layered menues. The are social issues, e.g. that untouchables often think of cell phones as not for them. Most interestingly, imho, were the questions raised about the role of the mediator,  e.g. the village telephone lady. They learn accountancy on the fly and quickly become highly competent.

 Their technical and clerical competence set them apart and give them the opportunity for abuse.  So far customers do not seem to mind but this issue may well come to a head. From a design perspective,  cell phone design might think about the relationship between mediator and customer, aiming to increase the transparency of the mediator’s activities to the customer, in the process both hindering abuse and  increasing the trust.

Inducement Prizes and Innovation

Thanks to Erika W. at X Prize, am reading a recent paper by Liam Brunt (University of Lausanne), Josh Lerner (Harvard) and Tom Nichols (Harvard) which examines a 19th century data set to discern insight into whether prize-awards are a useful mechanism to encourage innovation. The answer seems to be yes.

Some insights:

  • Metric. The number of contestants the prize competition attracts. 
  • Entry fees. Discourage spurrious entrants.
  • Money and medals. Used as substitutes, gold medal having largest entrant effect.
  • Diffusion. Shows drew popular interest, spread of technical knowledge.
  • Patents. Prize winners much more likely to patent after show than non. Interestingly, doubling of monetary award increases patent activity 6-7 percent in the technical area in the year of the show, while a gold medal has a 33 percent effect.
  • Lead time. Providing longer lead times to inventors raised the number of entrants.

The paper concludes that, among other things, that the monetary awards covered only about one third of the costs of implements and machinery exhibited. More potent was the exposure from the show and “society’s mark of approval.”