Results. Two-step GMM estimates of simultaneous equations. Dependent variable: Income.

<p>From left to right, the number of control variables, which are listed on the left, increases in a stepwise fashion. The IVs for disease are variables that are in the disease equation (listed in the corresponding columns in <a href="http://www.plosbiology.org/article/info:doi/10.1371/journal.pbio.1001456#pbio-1001456-t005" target="_blank">Table 5</a>) but not in the income equation here. Robust standard errors are presented in parentheses next to their corresponding coefficient estimates. First-stage <i>F</i>-test indicates strength of IVs if there is only one endogenous variable (disease). If there are multiple endogenous variables (disease, institutions, and spatially lagged income), Shea's partial <i>R</i>² indicates strength of IVs <a href="http://www.plosbiology.org/article/info:doi/10.1371/journal.pbio.1001456#pbio.1001456-Shea1" target="_blank">[75]</a>. Bold indicates significance at the 10% level. <i>n</i> = 139.</p>IV<p>Variable is instrumented.</p>J<p>Based on Hansen's J statistic.</p>a<p>Units×10⁻².</p>*<p>p≤0.10.</p>**<p>p≤0.05.</p>***<p>p≤0.01.