Every Monday our authors provide a round-up of some of the most recently published peer reviewed articles from the field. We don’t cover everything, or even what’s most important – just a few papers that have interested the author. Visit our Resources page for links to more journals or follow the HealthEconBot. If you’d like to write one of our weekly journal round-ups, get in touch.
Decision Making Under the Gambler’s Fallacy: Evidence from Asylum Judges, Loan Officers, and Baseball Umpires. [RePEc] Quarterly Journal of Economics. Published August 2016 (Online March 2016).
This paper poses the question of whether the sequence of a set of decisions matters for the outcomes of those decisions. A large literature has arisen on the myriad factors that can influence a decision including, famously, whether or not the decision maker is hungry. In this study, Daniel Chen and colleagues find evidence of negative autocorrelation in sequences of decisions – previous positive decisions in the sequence reduce the probability of subsequent positive decisions. This, they posit, is due to the gamblers fallacy, which is the common mistake of underestimating the probability of a streak occurring in a sequence of events. After controlling for quality they find evidence for this in US judges’ decisions on refugee asylum cases, loan officers deciding on loan applications, and in baseball umpiring decisions. However, this contrasts with a recently published randomised trial by on the effects of the ordering of scans in mammogram screening, which showed no evidence of the effects of case order using a very large sample. One could try to imagine a narrative to reconcile these two studies: mammogram film readers are more highly qualified, mammogram screening is more objective, or something along these lines. But caution should also be exercised in developing ever more elaborate explanations when the simpler explanation that one of these studies arrived at its results by sheer chance may be true. Nevertheless, Chen and colleagues provide compelling evidence of the gamblers fallacy in sequential decision making. Corrections are typically made for fatigue and learning effects and multiple reviewers are typically used for each case note in case note review studies, for example, this study provides further evidence to support such corrections.
Effect of democratic reforms on child mortality: a synthetic control analysis. The Lancet Global Health. [PubMed] Published August 2016.
The synthetic control method is growing in popularity as a method to estimate counterfactuals when you have one treated unit with multiple possible controls. A seminal example was the analysis of the effects of tobacco control reform in California when all the other states were potential control states. This study attempts to analyse the effect of democratic reform on child mortality in a number of countries from around the world. They show some evidence for a reduction in the child mortality rate following on from democratic reform although only in a minority of the countries studied and typically only those with a high rate to begin with. However, despite some explanations in the discussion, it remains unclear how such reforms would reduce child mortality. Reforms are determined by dichotomising the Polity 2 index so that a state is ‘treated’ when the score goes from negative to positive. Such a change could happen when executive power is moderately limited. It would be useful to see whether there is evidence to support the corollary, that a transition to autocracy leads to an increase in the child mortality rate. Nevertheless, there is a substantial lack of political economy and health studies, many health economics conferences have very few, if any, papers on this theme and further investigation should be welcomed.
The Short-run and Long-run Effects of Birth Weight: Evidence from Large Samples of Siblings and Twins in Taiwan. Health Economics. Published July 2016.
Finally this week a return to the theme of infant health that we have covered previously. Birth weight is frequently used as an outcome to proxy infant health at birth since it is widely and simply measured. Previous studies (e.g. Black et al 2005) have also demonstrated that low birth weight is associated with worse health, educational, and labour market outcomes in later life (thus providing a pathway for the intergenerational transmission of health and wealth). This new study uses new, large samples of both twins and siblings from Taiwan to investigate the effect of birth weight on short and long run outcomes. Higher birth weights are shown to be associated with reduced infant mortality, increased probability of attending college, and higher educational attainment. This reproduces previous findings from the literature (a good goal of any paper). The next step in this literature is to explore the mechanisms mediating this relationship; one might hope that policy can prevent a person being consigned to worse outcomes in life due to merely random fluctuations in their birth weight. As a final note, this otherwise great paper provides a good example of where stars for significance are somewhat redundant in light of the sample size (>6million in some cases).