The Effect of HIV Infection Risk Beliefs on Risky Sexual Behaviors: Scared Straight or Scared to Death? - Job Market Paper

Much of the development work is about behavior change, the assumption that with extra resources and knowledge people (also known as beneficiaries) will do things better in the future. Nowadays there is an increasing shift towards the 'at-risk' label (although 'vulnerable' is still going strong) to justify targeting: youth at-risk, women at-risk, children at-risk. However, Kerwin's job market paper "The Effect of HIV Infection Risk Beliefs on Risky Sexual Behaviors: ScaredStraight or Scared to Death?" suggests that we may be wrongly identifying both target groups and responses.

Abstract

"Economists typically assume that risk compensation is uniformly self-protective - that people become more careful as the health risks of their actions increase. However, risk-seeking, or fatalistic, responses can also be rational: increased risks can lead people to take fewer precautions. I extend the typical model of risk compensation to show that fatalism is a rational response to sufficiently high risks if people do not have perfect control over all possible exposures, and if the condition in question is irreversible. This result holds even for people who do not understand how to add up probabilities. I test this model's implications by randomizing the provision of information on HIV transmission risks to people in Malawi, a country with a severe HIV epidemic where there is qualitative evidence of fatalistic responses to the virus.

Average risk responses are self-protective and statistically significant, but small in magnitude: the mean risk elasticity of sexual behavior is roughly -0.6. To test the model of rational fatalism,I develop a method of decomposing 2SLS estimates of the risk elasticity of sexual behavior by baseline risk beliefs. Consistent with the predictions of my theoretical framework, I find that this elasticity varies sharply by baseline risk beliefs: the risk elasticity varies from -2.3 for the lowest initial beliefs to 2.9 for the highest initial beliefs. 13.8% of the population has a positive elasticity, suggesting they are fatalistic."

While the study focuses on HIV, how we handle 'at risk' groups is conceptually similar across the sectors. In most cases, targets are identified according to socioeconomic status, employment, location, age and other indicators of potential risk. The advantage of this method is that is relatively easy to identify people with this criteria. The study adds yet another criteria: fatalism, an indicator extremely relevant and yet very difficult to measure (somehow some behaviors can be used as proxy). A fatalist view will not attempt to mitigate the risks but rather embrace them and assume the worst already, making traditional risk education infective (as it tends to highlight negatives as a preventive methodology).

We see a similar issue regarding suicide bombers, gang membership and other high risk behaviors, where the at-risk profile is wide enough to include thousands (if not millions) and yet the actual number of joining are well below. Maybe what we need is better psychological profiling in order to be more precise in both identifying beneficiaries and tailoring the message.

HT to Development Impact

Give Everybody a Voice! The Power of Voting in a Public Goods Experiment with Externalities - Working Paper

While, in general, I am not keen on social science results coming out of controlled lab experiments, Engel and Rockenbach's "Give Everybody a Voice! The Power of Voting in a Public Goods Experiment with Externalities" offer an interesting though with potential large implications (if proven by field testing) for development practitioners.

 
Abstract:  

"We study the effect of voting when insiders’ public goods provision may affect passive outsiders. Without voting insiders’ contributions do not differ, regardless of whether outsiders are positively or negatively affected or even unaffected. Voting on the recommended contribution level enhances contributions if outsiders are unaffected and internalizes the negative externality by lowering contributions when outsiders are negatively affected. Remarkably, voting does not increase contributions when it would be most desirable, i.e. with a positive externality. Here, participants vote for high contributions, yet compliance is poor. Unfavorable payoff comparisons to the outsiders that gain a windfall profit drive contributions down."

In many development interventions, specially with the popularity of Community Driven Development (CDD), communities are asked to chose priorities and, in some cases, also contribute to them. Those priorities usually reflect public goods (or rather public services that can have a public good component, like schools, medical posts, markets, community centers) that do have externalities on other communities (or even within the community). Unfortunately, in development we tend to treat externalities either as a risk (other communities/actors reacting negatively, spoilers) or as a multiplier-effect (bonus points because there is a positive spill-over beyond the community) but rarely as something integral to the design of the decision-making process and the community contribution (both at initial stage and on running costs).

The study shows that enforcement of decisions is key (nothing new there) but also that voting and the nature of the externality have an effect on contributions: If negative it lowers (acknowledgement of impact on others?), if positive it also lowers!! (why should we pay for those free riders?) and if neutral increases (it is all ours?).

In my view policy implications are multiple: 

1) we need to field test this hypothesis, 

2) we may need to tailor decision-making processes and corresponding contributions according to externalities (assuming we want to maximize community contribution). Of course this is easier said that done, specially in CDD settings where the prioritization outcome is not necessarily known beforehand (although the guessing can be quite narrow),

3) extend voting to all contributors. While this point is not looked into by the study, the fact that there might be contributor not participating in the decision-making (indirectly so, through the non-binding phase), we can assume that shifting them from semi-bystanders to decision-makers/contributors would increase the compliance (a point made in many studies in political science on broadening the voting franchise AND the tax base),

4) in stable systems, both and positive negative externalities are dealt by shifting the decision-making process to a higher level (district, region, nation) that can provide for a compensatory system (insurance, investments, balancing out positive and negative externalities). However, in many development settings a 'bump-up the ladder' is not an efficient option (and a whole different 'governance' debate). Clustering (identifying and including affected parties) for decision-making and contributions (in the case of positive externalities) would make the public good neutral (and raise contributions). For negative externalities it is much harder unless the potential damage has been pre-identified and a compensatory mechanism already established (i.e. another development project in the affected community, mitigation measures).

School Competition and Product Differentiation - Job Market Paper

School competition is widely presented as a model for improving educational outcomes. Ongoing debates on school vouchers, charter schools and the role of private education show an increased focus on the use of market forces in the education sector. Bau's "School Competition and Product Differentiation" takes a slightly different approach, instead of the 'traditional' comparison between public and private schools, she looks into the effect of competition within the private sector.

Abstract

"Policies that encourage school competition in hopes of improving school quality are increasingly popular in both rich and poor countries. In this paper, I establish that schools may respond to increased competition by catering to wealthier students, at the expense of poorer students. I develop a model of school competition in the presence of differential information.

If (1) the match between a school and a student affects learning, and (2) wealthier students' enrollment decisions are more responsive to their match to schools, then schools cater more to the needs of wealthy students. Therefore, the entry of an additional school can reduce the quality of the match for poorer students. I show that the key mechanism of the theoretical model - that wealthier student are more responsive to their predicted test score gains when choosing schools - is consistent with the data by estimating a structural model of school choice in Pakistan. Exploiting the exit and entry of private schools and an instrumental variable that affects the cost of opening a private school, I find that competition increases private schools' targeting of wealthy students and reduces their targeting of poor students. An additional private school in the market increases within-school inequality in yearly test score gains by 0.1 standard deviations."

Many arguments promoting school competition are framed as pro-poor (elitist arguments are bad form these days), however the study shows that there might be diminishing returns and actually reduce learning for already enrolled poor students. Unfortunately the author does not follow up on one stark implication: the increased focus on wealthier students when the competition rises (as indicated in her own model). Therefore the issue does not seem to be informational (poor students choices are a worse match) as stated, but rather the schools own focus. These two issues might need to be disentangled further.

An interesting note, especially for fans of path-dependence, is that the presence of private schools is correlated to girls' government secondary school in the village (promoted mainly in 1980s through the Pakistan Social Action Program). Secondary-school educated young women are a key labor source for private school teaching, bringing on board another variable for education policy and territorial coverage of competition.

HT Development Impact

Interesting links

Interesting links of the week

- Winning elections one knock at a time, the importance of ground work/canvassing.

- Skeptics guide to institutions, 4 parts so far.

- Spain ranks highest in financial access according to this new study

- Chinese military classics on the economics of peace and war

- The myth of Armchair Socialist debunked?

- Development professionals and their (our) behavioral biases

What is Title Guarantee Worth in Land Markets? Evidence from Bengaluru, India - Working Paper

Land reform has been a constant topic not only on development literature but also in economics and politics for many years. Many popular movements (specially agrarian) have been focused on land distribution and ownership, although in later years the theme has been losing visibility (partly due to the diminishing role of agriculture in many economies, partly due to its association with 'marxist' movements vis a vis 'free market'). On the other hand the issue of titling urban settings, specially in relation to squatter/slums in developing countries has garnered more recent attention (probably it most well known leader being Peruvian Hernando De Soto). Venkataraman's "What is Title Guarantee Worth in Land Markets? Evidence from Bengaluru, India" falls in this last group by looking at the premium generated not only by a solid/guaranteed claim but also adequate planning in urban settings.

Abstract:     

"Land reforms require urgent attention in emerging market economies, and there is a vast body of literature that deals with the economic impact land reforms, especially land titling (Acemoglu et al., 2001). India, like other developing economies, has a presumptive titling system which is notoriously inefficient: a law that is in the draft stages of consultation is the Draft Land Titling Bill, 2011, which seeks to replace presumptive titles with conclusive ones. However, there is no study that has looked at the cost of not having an appropriate titling system. In this study, I attempt to quantify the premium paid towards title by using a quasi-natural experiment on differential titles that is available in the Indian market, utilizing land prices in Bangalore, India. The differential title to land comes about as a result of State's activity in using eminent domain to acquire land parcels, conferring superior title on the parcels so acquired. A hedonic model is applied to a data set of 2263 observations of appraised land values to tease out the impact of guaranteed land title on land prices."

The paper finds that the land title premium represents, in this case, a 4.3% increase in price. While this increase might not be significant enough for a private owner, as the costs of securing rights may easily overcome any expected profits, it begins to become significant for larger private developers and governments. The author calculates that for Karnataka real estate alone the welfare loss (fancy word for the amount the % increase is not currently realized) is at least 430 million USD, while the cost to the state to deliver title guarantee is 6.6 million USD (although probably that budget would dramatically increase if it were to cover the whole state, I would imagine). A very large unrealized potential.

To note that, in the study, the solid title is produced through the use of eminent domain, wiping out potential competing claims through compensation and legal standing. A methodology that may be hard to scale up or replicate. Nevertheless, the study shows both the advantage of clear and uncontested land titles, and also the role of the state in providing a registration and guarantee service.