Climate change and conflict

One of my research streams is on the impacts of climate change. Specifically, I am interested in the impacts of climate change on population dynamics, especially migration (which I will talk about in a future post, but if you are interested the large MBIE-funded research programme that I am involved in is described here).

Last year, I ran across this paper in Science (ungated version here), entitled "Quantifying the Influence of Climate on Human Conflict", by Solomon Hsiang of Princeton University, and Marshall Burke and Edward Miguel both of University of California, Berkeley. Essentially, the authors conduct a meta-analysis to investigate whether climate variables cause conflict. Meta-analysis is a method of combining the results of many previous studies to generate a single (and usually more precise) estimate of the effect size. As a method, it is common in medicine and rapidly gaining currency in quantitative social science.

As an aside, definitively identifying causality is a problem (as I've outlined before). At least in this case, reverse-causality (conflict causing climate change) is a little implausible (having said that, of course, it depends on the nature of the conflict and these would probably have some impact on climate), and contemporaneous causation (where some other variable causes both climate change and conflict) is also fairly unlikely.

The authors found that:

...large deviations from normal precipitation and mild temperature systematically increase the risk of many types of conflict, often substantially, and that this relationship appears to hold over a variety of temporal and spatial scales... The standardized effect of temperature is generally larger than the standardized effect of rainfall, and the effect on intergroup violence (e.g. civil war) is larger than the effect on interpersonal violence (e.g. assaults).

In other words, higher or lower than normal rainfall, and warmer temperatures increase conflict. How much more conflict? From the paper:

Nearly all studies suggest that warmer temperatures, lower or more extreme rainfall, or warmer El Niño-Southern Oscillation (ENSO) conditions lead to a 2 to 40% increase in the conflict outcome per 1σ in the observed climate variable.

This is quite a substantial effect. Which leads us to climate change, where the pattern of rainfall is expected to change (increase in some areas and decrease in others), and where temperature is expected to increase significantly. Again, from the paper:

These large climatological changes, combined with the quantitatively large effect of climate on conflict - particularly intergroup conflict - suggest that amplified rates of human conflict could represent a large and critical impact of anthropogenic climate change.

So, as climate change kicks in we can expect more violent conflict. What does that mean for New Zealand? Matching the effects up with this MFE publication by Brett Mullan of NIWA and co-authors leads to some interesting effects. More violence and intergroup conflict on average across the country (through increases in winter and summer temperatures), and more violence in the west of the North Island (which is expected to have both wetter winters and drier summers). Look out New Plymouth!

Does exposure to markets make us more, or less, rational?

We know that people are not fully rational - they try, but they are subject to predictable cognitive biases. These biases form the basis of behavioural economics.

One of these biases is the endowment effect. The classic experiment that demonstrates this effect was run by Daniel Kahneman, Jack Knetsch, and Richard Thaler. Kahneman describes the results in his book (Thinking, Fast and Slow):

Mugs were distributed randomly to half the participants. The Sellers had their mug in front of them, and the Buyers were invited to look at their neighbor's mug; all indiated the price at which they would trade. The Buyers had to use their own money to acquire a mug. The results were dramatic: the average selling price was about double the average buying price, and the estimated number of trades was less than half of the number predicted by standard theory.

Rational decisionmakers, who are willing to pay up to $10 to gain an item, should also be willing to sell that same item for $10. So, on average we would expect buying and selling prices not to diverge greatly. But they do, as shown in the experiment (and in a great many other experiments besides).

There are several explanations for this, but I will concentrate on one in particular - loss aversion. In general, people are loss averse because we value losses more than we value gains. Gaining $10 makes us happier, but losing $10 makes us unhappier to a greater extent than gaining $10 makes us happier. How does this relate to the endowment effect? Using the experimental example above, the sellers have a coffee mug, which they own. Let's say that they would have been willing to pay $10 to get their mug. Giving up their mug creates a loss (they lose their mug), and because of loss aversion they value that loss at more than $10. So, they would only be willing to sell their mug for some amount that is more than $10.

Other experiments have demonstrated that the endowment effect reduces as people gain more experience at trading in the market. The work of John List (surely a future Nobel prize winner) in the markets for baseball cards demonstrates that endowment effects reduce with trading experience (see here for a nice summary, or the gated article here). Given that these results show that market experience reduces the endowment effect, a commonly held belief among economists is that exposure to markets makes us more rational.

Which brings me to this paper from last year, by Coren Apicella and  Eduardo Azevedo of the University of Pennsylvania, James Fowler of UC San Diego, and Nicholas A. Christakis of Harvard University. The paper is described in lay terms in some detail here, so I will forego a long explanation. Essentially, in this paper the researcher describe an experiment where they tested for the presence of the endowment effect among one of the few remaining hunter-gatherer societies in the world - the Hadza Bushmen of northern Tanzania. They exploited the fact that some Hadza live close to tourist trails and interact with tourists and guides, while others live further away and have fewer contacts with outsiders. Those who have had more interactions were hired as guides, sold bows and arrows, etc. In other words, they engaged in more market transactions. What the researchers found was that the Hadza who lived in the most isolated villages did not display endowment effects, while those who had engaged in more market transactions (because they lived closer to the tourist trails) did exhibit significant endowment effects. Based on these results then, exposure to markets makes us less rational.

How can markets make us simultaneously more rational and less rational? The apparent paradox is troubling, but consistent with the old proverb that a little knowledge is a dangerous thing - a little bit of market experience may reduce rationality, until we become experienced enough in market transactions to overcome the endowment effect. Alternatively, Apicella and friends give some potential explanations for their results, including exposure to "concepts of ownership, selling and purchasing of goods, and payment for labor, all of which could be driving the group’s preference for owned items". I'm left wondering if it's the concept of ownership that leads to the endowment effect. I imagine that hunter-gatherers have quite a different conception of ownership than we do in the western world (the Apicella and friends paper notes that "While the Hadza do own some items, such as knives, bows and arrows, and animal skins, ownership is limited to what can be carried"). Exposing the Hadza to markets may broaden their conception of ownership, and more importantly to the loss of ownership. I wonder if there are any plans to test whether loss aversion is present amongst this group?

Lock up your daughters and make them watch MTV

Economists are often criticised for using quantitative methods that make little sense to the average person (here is a good example). One of the common tools in the economists toolbox, which is barely used in other disciplines, is the instrumental variables (IV) method. To the average human, IV is almost laughable. Essentially, say you are interested in the effect that one variable (call it 'x0') has on some outcome variable (call it 'y'). Now, instead of running a simple analysis where you see if x0 and y are correlated, you replace x0 with some other variable (call it 'x1', the instrument) and run the analysis to see if x1 is correlated with y. There are good reasons to run an analysis in this way (see here), but I am sympathetic to the ordinary person's critique that this method makes little sense.

To run the IV method, you need two things. First, you need a variable x1, that is a good instrument for x0. For this to be the case, the two variables must be closely correlated - close enough that x1 is a good proxy for x0 (if this isn't the case, then you have a 'weak instrument' problem). This condition isn't often a problem. Second, your variable x1 must not be directly related to y, except through the variable x0. This is the 'exclusion restriction', and I believe this is more often a problem.

Now, an example (maybe!). A few weeks back, this NBER Working Paper (gated, ungated version here) by Melissa Kearney of the University of Maryland and Phillip Levine of Wellesley College hit the news (New York Times story here, CNN here). The authors investigated the effect that the reality TV show 16 and Pregnant had on teenage birth rates in the U.S. The authors conclude:

Our results suggest the introduction of the show led young women to search and tweet about birth control and abortion, indicating that it had some influence on them in a way that could potentially change their behavior. We also find that exposure to the 16 and Pregnant shows had a sizable impact on the rate at which teens give birth in the United States, generating a 5.7 percent reduction in teen teen births that would have been conceived between June 2009, when the show began, and the end of 2010. That can account for roughly one-third of the decline over that period.

First, I though this was a really interesting and worthwhile study. We often hear about the negative impact that media has on behaviour (e.g. violence and television, or video games, food advertising and obesity, and many others). It's good to investigate whether there are positive effects as well.

This study uses the IV method. The reason is explained by the authors:

A critical issue in implementing this approach is accounting for the possibility that locations in which the show is more popular are not randomly selected. Perhaps the show is more popular in locations with elevated rates of teen childbearing. If so, OLS estimates of the relationship between ratings and teen childbearing would include a positive bias, incorrectly suggesting that higher ratings lead to more teen births... To address this form of bias, we utilize an instrumental variables (IV) approach. We instrument for the show’s ratings using ratings among those between ages 12 and 24 for all shows that aired on MTV on weekday evenings between 9:00 and 10:00 in the 4 sweeps months preceding the introduction of 16 and Pregnant (July 2008 through May 2009).

In other words, they replace the television ratings for 16 and Pregnant (x0) with the television ratings for MTV in general from before 16 and Pregnant was introduced (x1). The authors are essentially suggesting that their instrument meets the exclusion restriction, because MTV ratings before 16 and Pregnant began couldn't directly affect birth rates after 16 and Pregnant premiered. But, you ask, if teens are watching more MTV doesn't that leave them with less time to initiate a pregnancy (because they're too busy watching television!)? So, MTV ratings (even excluding 16 and Pregnant) would have an effect on birth rates even without 16 and Pregnant, which violates the exclusion restriction.

By using MTV ratings from before 16 and Pregnant premiered (rather than MTV ratings after), then authors are trying to get around this problem (the argument is that watching more TV before 16 and Pregnant premiered can't directly reduce sexual behaviour afterwards). Here is where I struggle - if MTV ratings after 16 and Pregnant premiered are related to MTV ratings before 16 and Pregnant premiered, and MTV ratings after 16 and Pregnant premiered in turn are related to teen birth rates, then doesn't that violate the exclusion restriction? I think it does.

However, even if you don't believe in the IV approach, the authors have a neat way of convincing us. They show a plausible mechanism through which this causal relationship (16 and Pregnant reducing teen birth rates) would work (see the first sentence of their conclusions above). They show that there is more Google search activity and more Twitter activity relating to 'birth control' immediately following the airing of 16 and Pregnant. The implication is that more searches and mentions of birth control lead to more use of birth control and fewer teen pregnancies.

So, the overall conclusion? If you believe in the IV approach, which many economists do, then 16 and Pregnant caused a significant reduction in teen births. If you don't believe in it, then watching more MTV is related to a significant reduction in teen births. Either of these is an interesting result in its own right.

Closer to home, New Zealand's teenage birth rate peaked at 32.85 per 1000 women aged 15-19 in 2008 and has declined since (to 24.89 in 2012; lower than the 29.4 in the U.S. in 2012). I wonder how many girls in New Zealand are watching 16 and Pregnant? I think I'll advocate that my almost-teenage daughter starts watching more MTV.

[HT: Freakonomics blog]