Economic Geographies

Dean Karlan & Jonathan Zinman have quantified something most field researchers in the developing world probably sensed already – that respondents are often very sensitive to social perceptions around “undesirable” activities such as borrowing.  In South Africa, they found that roughly 47% of known borrowers at a microcredit bank told surveyors they had not borrowed, with women being more likely to lie than men, and especially more likely to lie to male surveyors than female.  (The surveyors had no connection with the bank, and thus no independent knowledge of the correct answer for a given respondent.)

This is the eternal question at the heart of field research – what do you do if self-reported data isn’t accurate, but it is all you have?  And how does one incentivize truth-telling?  An interesting corollary to this experiment would have been to tell a subset of respondents that the surveyors knew they belonged to the bank (without giving the surveyors any other personal information about respondents’ accounts), and measure how the mere mention of the lending institution changed respondents’ rate of truthful reporting.  How would this work in situations where there is legitimately no authority with credible information on respondents, like measuring landholding in situations where land tenure is totally informal?  Would this be too destructive to surveyor impartiality (and general research ethics) to be an option?  The more I delve into these complicated questions of survey design, the more I find them strangely fascinating.

I promised Asif Dowla some time ago that I would blog about some of his work, and after digging myself out from a fair deal of work that accumulated whilst I was focusing on grad school applications, I finally read his interesting 2005 article on how Grameen Bank built social capital among its members. Dowla identifies three components of social capital – trust, norms, and social networks – and what primarily fascinated me from an institutional perspective was how the norms & networks of rural Bangladesh played into Grameen’s search to build vertical trust between bank & clients.  Women’s access to finance seems to be a less contentious issue in Bangladeshi microfinance today, given the country’s relatively high penetration of microfinancial services, but at the outset Grameen had to work carefully around purdah norms and women’s correspondingly limited social networks in order to convince potential clients that it was a serious partner.  As religious restrictions on gender and public space are less pronounced in most of Africa, I had previously thought of trust-building as more a matter of demonstrating reliability and transparency of services, as Portfolios of the Poor observes.  It’s interesting to see the creation of institutional trust explicitly embedded in a local context like this.  (On a related note, I was surprised to learn that Bangladesh has more microfinance borrowers than all of Africa and Latin America combined.  They’ve come a long way.)

Naturally, all of this put me in mind of Tim Harford’s hugely insightful article on the importance of institutional trust for economic growth.   Dowla has a good section on how building non-kin-group networks through group lending was an important effect of Grameen’s work in the Bangladeshi context, but he also hints at the limits of building a financial system on personal trust when he observes that joint liability within groups seemed to be enforced more by staff than by other group members.  This makes sense on the face of it, when one thinks about the conflicting incentives any group member might have to force another member to cover a missed loan payment.  On the one hand, predicating continued access to credit on full and timely group repayment is a powerful incentive to force payment, but on the other hand, clients might want to be given some leeway themselves if they miss a payment in the future – or might not want to act as hostile enforcers towards people they’ll continue to see around the neighborhood.  This is mere speculation, of course, but it does point to the advantages of shifting loan enforcement duties from a situation governed by personal trust to one governed by institutional trust.  I think it would be a good sign if Grameen’s groups had naturally drifted towards this arrangement.

Finally, my favorite phrase?  The poorest of the poor often have “a short radius of trust … because trust [must be] enforceable by the threat of retaliation.”

Another interesting document recently – this one a qualitative study of group liability borrowers in India, which highlighted the role of compulsory savings in both easing loan repayments and increasing borrowers’ confidence to take on larger loan amounts.  I find this fascinating in part because it’s somewhat counterintuitive – if compulsory savings place a greater burden on borrowers to provide some money every week, all other things being equal, I would have expected such a program to be a disincentive for participation, rather than an incentive.  The study suggested that women were more confident in their borrowing because they knew they could draw down their savings if necessary, which almost suggests that compulsory savings functions as a type of emotional capacity building for financial planning.  Cool result, although it needs more quantitative testing.

[NB: Can't find the citation for this one either!  Definitely kicking myself for not including it in the first place now.]

The insight of pay-as-you-go schemes as a response to variable incomes is a great one.  Its applicability to things like mobile phone minutes is obvious, and I’m very curious as to how it could be applied to financial services for the poor.  The big stumbling blocks appear to be the questions of credibility and enforceability.  Or, rather, the fact that enforceability tends to engender credibility – banks believe their clients when they say they’ll pay, as they can easily check up on their payments and enforce them if they fall behind.  A pure pay-as-you-go plan in microcredit – say, a three-month loan that can be repaid in any amount at any time before the three-month period is up – would be unenforceable over the period of payment, as the bank couldn’t demand payment at any given time before the deadline.

Actually, that’s a separate (and interesting) question – how different types of enforcement may affect the repayment of loans.  Perhaps you could look at peer pressure in lending groups, vs. occasional visits from a loan officer, vs. frequent visits from a loan officer, vs. financial incentives for timely payment, or something like that.  That might shed some more light on designing systems of incentives for ultimate on-time payment of a pay-as-you-go loan.

When I was researching microinsurance and maternal mortality last year, I was struck by some of the observations that other researchers felt it necessary to include in their results.  One of them was the finding that distance to a health center affects people’s access to care.  In other news, water quenches thirst!  I had to wonder if this was a relic of the general lack of forethought that must be put into procuring transport in the global North, where it’s more or less equally simple to reach a doctor one kilometer from one’s home as thirty kilometers.  I otherwise fail to see how it’s notable that people who live farther from a clinic may use it less often.

This does highlight the fact that there are fundamental issues of healthcare access that aren’t purely microeconomic in nature.  Distance is one, but the challenge of retaining skilled doctors in a low-wage environment is a second, and difficulties in obtaining and maintaining quality equipment and medication stocks (non-counterfeit medications!) are a third.  The attitudes of healthcare workers also appeared extremely important to low-income patients, who seemed understandably sensitive about their social status, and hesitant to use centers where they would be treated disrespectfully because of their poverty.

The other thing I’ve been thinking of, however, was a little-discussed (at least in the papers that I read) corollary to the observations that microinsurance increases healthcare access, and health centers are favorably inclined towards patients who can actually pay for their care. My immediate concern upon reading these statements was, if access to microinsurance is still uneven, isn’t there a real possibility that patients who are even slightly better off will crowd out those who are too poor to afford $2-a-year insurance at all? If the resource base of health centers is fixed (and it may not be – I don’t have info on that), dramatic increases in patients covered by microinsurance could very well make the poorest of the poor even more vulnerable. I wonder how you’d best be able to test that.  I imagine you’d have to look at the effects of a growing resource base (if the increased payments are used at the local level) or the improved quality of care referenced in the last post, and sort out what effects those have on the healthcare uptake rates of the poorest.  Perhaps the question actually is, does extending microinsurance to some harm the uninsured by crowding them out, or does it improve their situation by letting them get a bit of a free ride on some improvements brought about by the insurance payments?  Interesting.

Microinsurance is one of the most fascinating and tricky types of pro-poor financial services that I’ve encountered.  The basic concept of insurance (be it for health, crops or life) is an incredibly powerful one in the face of exogenously varying incomes and the strong need for income smoothing and predictability, but it seems to suffer from the same problem as savings vs. microfinance – an existing emergency is a stronger incentive to ex-post savings than a predicted or potential emergency is to ex-ante savings.  Thus there seems to be a very real moral hazard of people joining insurance programs shortly before a planned medical expenditure or crop failure, in order to receive all of the benefits of membership without the burdensome advance saving that long-term membership would require.  Given that insurance, savings accounts, and microloans all represents systems of savings & disbursement, it seems possible to incentivize all of them similarly to promote ex-ante savings.  I wonder if there are any microinsurance agencies that pay their clients to sign up, which has been effective for encouraging regular savings accounts.

Income variability is hugely important.  The salient factor about the incomes of poor people isn’t just that they’re low; it’s that they vary seasonally and daily, which makes regular payments for anything difficult.  Commitment savings devices are even better worth pursuing in this circumstance, because the combination of poverty and a variable income can be tragic when it comes to large welfare-related expenses, such as school fees or healthcare.  Pursuing “opt-in” savings strategies (such as paying people to open accounts) is a good option.   So is designing loan repayment systems with A) realistic assumptions of the timeframe of income generation, and B) flexible payment options.

In a sense, “emergency” microloans are just a more expensive form of saving for the poor.  Rather than saving in advance and earning interest, they have to save after the fact and pay interest.  The incentives of the poor (not reducing current consumption for future expenditures) and the banks (taking in interest instead of paying it out) are currently aligned on this issue, but it results in a suboptimal equilibrium.  Or, actually, I wonder how these incentives might vary between formal banks, and microfinance organizations (often non-profits) which may acquire their capital from different sources.  At the microfinance institution I worked for in Ghana, the interest paid by lendees was just enough to cover their local operating costs.  It’s not totally clear to me how they could have paid interest on savings, even had they legally been able to accept deposits (which they weren’t, because they were registered as an NGO and not a bank).