Monday, May 23, 2011

R.I.P. Aid Watch

On May 19, Bill Easterly and Laura Freschi announced that their blog Aid Watch was ending, after running about two years. I was shocked, and (after checking my calendar to make sure it was not April 1), dismayed.

Who else, I thought to myself,was going to call B.S. on the foibles and failings (and sometimes nearly criminal negligence) of certain aid agencies?  Who else was going to relentlessly debunk the egomaniacal schemes of certain self-styled aid messiahs?  Who else was going to have the guts to speak truth to power?  Who else was going to remind us that so many new programs had been tried in the past, with disappointing if not disastrous results?

Who else, I asked myself, was going to demand each day that aid simply "benefit the poor?"  Who else was going to point us toward new ways of providing aid that actually take into account what the poor want?  Who else was going to contrast the failures of closed, top-down aid systems with the successes of open-access systems that provide fast, rich feedback systems?  Who else was going to redefine the debate about aid?

Aid Watch was controversial from the start.  Bill and Laura did not hesitate to take on sensitive issues, and they did not cloak their criticisms in vague bureaucratic language.  Typical evaluations of aid programs are dense and oblique, with shortcomings buried under mounds of data and jargon.  After a couple hundred pages of analysis touting the benefits of a program, official evaluations often tip their hat to failure with a paragraph beginning "But challenges remain..."  And then phase two of the same program begins with only a modest modification to a fundamentally failed design.

Aid Watch specialized in cutting through all of the obfuscation to say bluntly "This does not work.  We should stop it now and do something else if we really care about helping the poor."

In their valedictory post, Bill and Laura promise that the work of Aid Watch will continue, with longer and more in-depth pieces under the guise of Aid Watch's parent - the Development Research Institute (DRI), at New York University, where Bill is Professor of Economics.  I, for one, am very much looking forward to the next chapter of Aid Watch.

Treating Employees as People

  • Over the period 1998-2008, the stock of the companies voted "best to work for" appreciated nearly seven times as much as the stock of the average company.
  • The most admired companies on Fortune Magazine's list had double the market returns of their competitors over a seven year period.
  • Only 13 percent of unhappy employees recommend their company's products, vs 78 percent of happy employees.
Those are just some of the findings reported in Dave Ulrich and Wendy Ulrich's book The Why of Work.  The authors argue that companies that treat people as just another factor of production are increasingly at a competitive disadvantage.  Conversely, companies that engage their employees in a meaningful way have higher market returns.  

Why?  The authors offer the following possible explanations:
  • "Employees are committed, productive, and likely to stay with the company
  • Customers pick up on employee attitudes and are more likely to do business with the company
  • Investors have confidence in the company's future, giving it a higher market value
  • The company's reputation in the community is enhanced."
There are many approaches to creating a meaningful work environment.  One of the simplest (if still rare) is just to treat employees like real people, with valuable insights and instincts for what is best for the business.  Another is for the company to engage with the broader communities where it does business, by providing mentoring and volunteering and/or by financially supporting schools, clinics, and other social initiatives.  In other words, by treating the broader community members as real people.  

What if all businesses treated their employees and customers this way?

Monday, May 16, 2011

Innovation and Parallel Processing

Tim Harford
This idea of allowing several ideas to develop in parallel runs counter to our instincts: We naturally tend to ask, "What is the best option?" and concentrate on that. But given that life is so unpredictable, what seemed initially like an inferior option may turn out to be exactly what we need. It's sensible in many areas of life to leave room for exploring parallel possibilities—if you want to make friends, join several social clubs, not just the one that appears most promising—but it is particularly true in the area of innovation, where a single good idea or new technology can be so valuable. In an uncertain world, we need more than just Plan A; and that means finding safe havens for Plans B, C, D, and beyond.
That is from an excerpt in Slate from Tim Harford's new book Adapt: Why Success Always Starts With Failure.

Tim notes how breakthrough ideas often come from outside the mainstream.  The famed British Spitfire plane is credited by many with having prevented defeat by Germany in WWII.  Yet the plane almost never came to be.  Not only was the inventor unconventional, but so was the source of funding at a time when the project was nearly cancelled:
Rescue came from the most unlikely character: Dame Fanny Houston, born in humble circumstances, had become the richest woman in the country after marrying a shipping millionaire and inheriting his fortune. Lady Houston's eclectic philanthropy knew few bounds: She supported oppressed Christians in Russia, coalminers, and the women's rights movement. And in 1931 she wrote a check to Supermarine that covered the entire development costs of the Spitfire's predecessor, the S6. Lady Houston was furious at the government's lack of support: "My blood boiled in indignation, for I know that every true Briton would rather sell his last shirt than admit that England could not afford to defend herself against all-comers." The S6 flew at an astonishing speed of 407.5 mph less than three decades after the Wright Brothers launched the Wright Flyer. England's pride was intact, and so was the Spitfire project. No wonder the historian A.J.P. Taylor later remarked that "the Battle of Britain was won by Chamberlain, or perhaps by Lady Houston." 
(Note that given her range of interests, no one would have accused Dame Houston of practicing 'strategic philanthropy!')

Hat tip for this excerpt to Michael Woolcock of the World Bank and Harvard.  His own recent blog makes excellent related points concerning aid initiatives. I have been heavily influenced in my own thinking by discussions with Michael over the years.

Monday, May 02, 2011

Fatal Errors vs. Real-Time Feedback

Fatal error: Unable to open file

That was the response when I clicked on the Email Us link at the US Postal Service website this morning.  Earlier, I had stood in a long line at our neighborhood post office while the one postal clerk on duty valiantly tried to process all the customers.  While waiting, I looked around and noted once that the run-down facility looks like something from the Soviet Union in the late 1980s.  (At least in the Soviet Union, there were more clerks on duty, even if they were not always friendly.)

The clerk in our neighborhood station has always been nice to me.  She labors under horrendous conditions, but if I smile and am nice to her, she always reciprocates.  Though the lines are predictably very long at the station during certain periods, and USPS never provides more clerks to assist, I try never to complain to the lone woman on duty, because I know she has no power to change things.

While standing in line this time, I decided I would give some feedback to the US Postal Service itself.  I looked around at the walls in vain for a sign saying "Questions? Comments?  Call us at xxx or email us at yyy."  Then I realized I have a USPS app on my iPhone.  Great! I thought, and fired it up.  Alas, the app provided lots of information about zipcodes and rates but no way to give feedback.   Darn, but at least they would have a twitter account, I figured, and ran a search on my phone.  Someone had in fact claimed the @usps name, but had made no tweets and followed zero people.

So I gave up and was finally waited on by the nice lady, who was relieved my particular parcel was easy to process.  Then I went home, and sat down at my computer and thought I would give USPS one last try.  Success!  There it was on - a link saying Questions? Comments?  Click here to email us.  I mentally apologized to USPS before clicking, and then my screen went blank and said: "Fatal error: Unable to open file."

How many agencies or companies do you know that make it nearly impossible for users to give feedback?  They are usually the ones whose leaders say "We hear loud and clear from our customers that..."  But the truth is that they don't hear loud and clear at all.  They only hear sporadically and indirectly, through surveys and analyzing data.

What if each supervisor at USPS had a dashboard that had real-time feedback from customers from email, twitter, text message, and the USPS website?  And what if the supervisor's boss could see that feedback in real time?  And what if, maybe even more important, the public could see the same feedback in real time?  How would that change incentives and behavior?

What if this were on the desktop of each USPS supervisor,
on the wall of each post office, and viewable by
everyone on the web or smart phone?

My own guess is that my neighborhood post office would be a lot cleaner, there would be plenty of clerks during peak periods, and they would rarely run out of supplies.

(Extra credit:  Which company or agency do *you* think would benefit most from such a dashboard?)

Sunday, May 01, 2011

More Tales of Two Tails

The following is cross-posted from AidWatch.
An eloquent 3 year-old would have been better asking "What the dickens are you talking about? Who is defining success? Who says failure is bad, anyway?" - Joe
Earlier I blogged about aid cheerleaders and critics. Each camp argues about the mean outcome of aid rather than the distribution of impact among projects. Both camps agree that some projects have positive results and others negative. So why not try to figure out which projects work and focus our resources on them?

I got some great and insightful comments and a few nice aid distribution graphs from readers. Here are some key themes:
  1. The mean *does* matter if the distribution is random. In other words, if we can't predict in advance what types of projects will succeed, we should only spend more resources if the mean outcome is positive.
  2. Many people believe that on average the biggest positive returns come from investment in health projects.
  3. We should also look at the distribution of impact even within successful projects, because even projects that are successful on average can have negative impacts on poorer or more vulnerable people.
  4. Given the difficulty in predicting ex-ante what will work, a lot of experimentation is necessary. But do we believe that existing evaluation systems provide the feedback loops necessary to shift aid resources toward successful initiatives?
  5. "Joe," the commenter above, argues that in any case traditional evaluators (aid experts) are not in the best position to decide what works and what doesn't.
From reader Daniel Kyba: "Those which do a good job are the ones with defined and observable measures - profit/loss; live/die and so on. These measures provide a form of a feedback mechanism at the project level to which the aid provider can respond. As you move towards the world of fuzzy concepts and measures that is where the ineffectiveness occurs, due to the lack of feedback mechanisms and because there is less definition of success/failure."

From reader Steve White: "Here is my graph based on two stylized facts about aid projects: 1) most projects have very marginal impacts (agricultural tools to villages, microcredit, school construction, textbooks, scholarships, deworming...) and 2) some health projects have HUGE impacts (vaccinations, DDT, bednets)." The two bars represent impacts between -1 and 0, and between 0 and 1"

Petr Jansky sent a paper he is working on with colleagues at Oxford about cocoa farmers in Ghana. The local trade association was upset that they could not get pervasive adoption of a new package of fertilizer and other inputs designed to increase yields. According to their models, the benefits to farmers should be very high. The study found that - on average - that was true, but that the package of inputs has negative returns to farmers with certain types of soil or other constraints. Farmers with zero or negative returns were simply opting out.

At first glance, these findings seem obvious and trivial. But they are profound, in at least two ways. First, retention rates are an implicit and easily observable proxy for net returns to farmers. We don't need expensive outside evaluations to tell us whether the overall project is working or not. And second, permitting farmers to decide acknowledges differential impacts on different people even within a single project.

What other ways could we design aid projects to allow the beneficiaries themselves to evaluate the impact and opt in or out depending on the impact for them personally? And how would it change the life of aid workers if their projects were evaluated not by outside experts and formal analyses but by beneficiaries themselves speaking through the proxy of adoption?