Placebo test

This figure displays the result of a placebo test using the sample of corporate commission cases. For each |$ ˜{K}\in \left\{\$ 5,000,\$ 10,000,\$ 15,0000,\$ 20,000,\ldots \$ 100,000\right\},$| I counterfactually assume that |$ ˜{K}$| is a kink of the trustee commission function. I apply the methodology of Section 3.1 to calculate the bunching estimator |$ \mathbb{B}$| at each value of |$ ˜{K}$|⁠. Each blue dot plots the estimator |$ \mathbb{B}$|⁠, on the |$ y-$|axis, corresponding to a value of |$ ˜{K}$|⁠, on the |$ x-$|axis. For each |$ ˜{K}$|⁠, I bootstrap this procedure 500 times to construct standard errors. The bars display 95% confidence intervals (1.96 multiplied by the standard error on either side of the estimate).  Appendix B.1 provides details.