Retraction: Dividend Taxes and the Allocation of Capital (American Economic Review (2022) 112:9 (2884-2920) DOI: 10.1257/aer.20210369)

This notice explains the authors’ reasons for retracting the published paper, Boissel and Matray (2022b). The authors are grateful for the data and code policies at the AER that make it possible to be entirely transparent. The code that accompanies the paper (Boissel and Matray 2022a) fully reproduces all of the published results and had been independently replicated under the AER’s policies prior to publication. The errors described in this notice, which are responsible for the retraction, are visible in that code and were brought to the authors’ attention by Bach et al. (2023). Boissel and Matray (2022b) studied the financial and real effects of a dividend tax policy change on firms in France. It contained four main results: first, firms exposed to the tax hike (treated firms) significantly reduced their dividend payouts; second, the increase in internal liquidity led treated firms to significantly increase their capital and labor; third, treated firms grew in other ways including in their average wages and sales; fourth, the investment effects were larger for firms with more profitable opportunities and with higher marginal product of capital ex ante. The authors are retracting the paper because Figure 4 of the published paper, which pertains to the second result on the increase in investment and which plots the event study difference-in-difference coefficients of the within-firm change in year-to-year investment flows by treated firms, relative to control firms, differs from the figure that was conditionally accepted by the handling editor in two ways. First, the firm size control variable in the conditionally accepted manuscript was coded in changes but was incorrectly written in the baseline specification as levels. This error was caught in the replication process when the data were rebuilt from scratch, and the variable was redescribed accordingly in the paper. This change in the baseline specification was not requested by the handling editor, and the authors did not notify the handling editor nor the data editor of the update. The authors used firm capital as a proxy for firm size so that in practice, the specification controlled for the change in capital, i.e., investment, in the pre-period. The authors regret that they did not recognize the significance of the modification at the time. Second, the code that plots Figure 4 in the publicly available code repository includes an alteration of two coefficients by a factor of 1.8, which was incorrect. This alteration affects the rendering of the figure, but not the underlying data. All the results reported in the tables and other figures are unaffected. The error in the rendering of the figure was not intentional. The data and the code for the analysis (Figure Presented) are stored on remote, external servers that were inaccessible to the authors for prolonged periods of time during the publication process so it is unfortunately not possible to recover past versions to isolate the error’s origin. Panels A and B of the figure included here, show Figure 4 from the paper that was conditionally accepted next to the one that was in the final submission to the data editor and published. The authors did not notify the handling editor nor the data editor of the changes to the figure after conditional acceptance. At the time these mistakes were made, Dr. Boissel was no longer involved in the project.