Supplemental Appendix for “breaking the ‘iron Rice Bowl:’ Evidence of Precautionary Savings from Chinese State-owned Enterprises Reform” (not Intended for Publication )

This appendix provides some detailed estimation results not shown in the paper “Breaking the ‘Iron Rice Bowl:’ Evidence of Precautionary Savings from Chinese State-Owned Enterprises Reform.” Table A1 shows the full estimation results in the models that control for PIH effects (Table 6 in the paper shows only a few key parameters). Table A2 presents the estimation results for several alternative specifications to control PIH effects. Column (i) shows the results in the model that includes an “income decline” dummy an additional control variable. Column (ii) shows the results of from the benchmark model estimated using the subsample of workers who expected their income to decline in the next five years. Columns (iii) shows the results in the model that includes pension contributions as an additional control. Column (iv) shows the results in the model that includes both pension contributions and interactions between pension contributions and the SOE dummy. In all regressions here, we focus on the sample with government assigned jobs to control for self-selection biases. Table A3 provides the estimation results from the model that controls for firm size effects An abbreviated version is presented as Table 7 Panel A in the text. Table A4 provides the estimation results from the model that controls for life-cycle effects. An abbreviated version is presented as Table 7 Panel B in the text. Table A5 presents the estimation results from the model that controls for other demographic factors (female, less educated, or less skilled). An abbreviated version is presented as Table 7 Panel C in the text. Tables A6–A7 show the detailed estimation results corresponding to Panels A and B of Table 8 in the text. Tables A8–A11 show the detailed estimation results corresponding to Panels B–G of Table 9 in the text. In particular, Table A8 controls for spouse effects. Table A9 controls for housing effects. Table A10 uses an alternative risk measure. Table A11 deals with eliminating zero wealth and alternative wealth measures.