Title : Monetary Non-Neutrality in a Multisector Economy: The Role of Risk-Sharing Authors : Jae Won Lee and Seunghyeon Lee * MATLAB(.m) and STATA(.do) programs are required to run the code provided here. * For random numbers, we specified 251 as the seed in both in MATLAB (sectoral10_simulate.m) and STATA (bootstrap_run_ns_ffr.do, bootstrap_run_ns_pn.do, bootstrap_run_rr_romer.do, bootstrap_run_rr_full.do) <> * Prior to plotting the figures below, all procedures in the Model(M1-M7) and/or Empiric(E1-E8) sections must be completed. - Figures and Tables in the main text Figure 1, 2 and Table 1 run : sectoral10_irf_figures_f.m Figure 3, 4 and Table 2 run : sectoral12_irf_figures_f.m Figure 5 and Table 3 run : sectoral24_irf_figures_f.m Figure 6 and Table 4 run : sectoral20_irf_figures_f.m Figure 7(a) run : irf_corr_empiric_bootstrap_vs_model_fan_ns_ffr.m Figure 7(b) run : irf_corr_empiric_bootstrap_vs_model_fan_ns_pn.m Figure 8(a), Figure 8(a) run : sectoral10_simulate.m - Figures and Tables in the online appendix Figure A1, A2 and Table A1 run : sectoral12_irf_figures_f.m Figure A3, A4 and Table A1 run : sectoral10_irf_figures_smoothing_f.m Figure A5, A6 and Table A1 run : sectoral10ait_irf_figures_f.m Figure A7(a) run : irf_corr_empiric_bootstrap_vs_model_fan_rr_romer.m Figure A7(b) run : irf_corr_empiric_bootstrap_vs_model_fan_rr_full.m Figure A8(a) run : irf_corr_empiric_bootstrap_vs_model_all_ns_ffr.m Figure A8(b) run : irf_corr_empiric_bootstrap_vs_model_all_ns_pn.m Figure A8(c) run : irf_corr_empiric_bootstrap_vs_model_all_rr_romer.m Figure A8(d) run : irf_corr_empiric_bootstrap_vs_model_all_rr_full.m Figure A9(a) run : irf_gamma_empiric_fan_ns_ffr_vs_model.m Figure A9(b) run : irf_gamma_empiric_fan_ns_pn_vs_model.m <> - Solving models for each specification (350 sectors) indpr = 11 implies that we use data on 350 sectors in the U.S. from Bils and Klenow (2004). (*) : Values not related to the model. (0.001 is recommended to reduce running time) M1. Taylor rule without interest rate smoothing M1.1 Complete market open : sectoral10_rep.m => set indpr = 11, epsil = 0.001(*), smoothing = 0 => run output : sectoral10gen_sig1_theta6_varphi117_eta1_rep_hps11_mean.mat M1.2 Bond economy open : sectoral10_het.m => set indpr = 11, epsil = 0.001, smoothing = 0 => run output : sectoral10gen_sig1_theta6_varphi117_eta1_het_hps11_eps-3_mean.mat M1.3 Autarky open : sectoral10_het.m => set indpr = 11, epsil = 1000, smoothing = 0 => run output : sectoral10gen_sig1_theta6_varphi117_eta1_het_hps11_eps3_mean.mat M2. Taylor rule with interest rate smoothing M2.1 Complete market open : sectoral10_rep.m => set indpr = 11, epsil = 0.001(*), smoothing = 1 => run output : sectoral10gen_sig1_theta6_varphi117_eta1_rep_hps11_smoothing.mat M2.2 Bond economy open : sectoral10_het.m => set indpr = 11, epsil = 0.001, smoothing = 1 => run output : sectoral10gen_sig1_theta6_varphi117_eta1_het_hps11_eps-3_smoothing.mat M2.3 Autarky open : sectoral10_het.m => set indpr = 11, epsil = 1000, smoothing = 1 => run output : sectoral10gen_sig1_theta6_varphi117_eta1_het_hps11_eps3_smoothing.mat M3. Average inflation targeting M3.1 Complete market open : sectoral10ait_rep.m => set indpr = 11, epsil = 0.001(*), smoothing = 0 => run output : sectoral10aitgen_sig1_theta6_varphi117_eta1_rep_hps11.mat M3.2 Bond economy open : sectoral10ait_het.m => set indpr = 11, epsil = 0.001, smoothing = 0 => run output : sectoral10aitgen_sig1_theta6_varphi117_eta1_het_hps11_eps-3.mat M3.3 Autarky open : sectoral10ait_het.m => set indpr = 11, epsil = 1000, smoothing = 0 => run output : sectoral10aitgen_sig1_theta6_varphi117_eta1_het_hps11_eps3.mat M4. nominal output growth rule M4.1 Complete market open : sectoral12_rep.m => set indpr = 11, epsil = 0.001(*), smoothing = 0 => run output : sectoral12gen_sig1_theta6_varphi117_eta1_rep_hps11.mat M4.2 Bond economy open : sectoral12_het.m => set indpr = 11, epsil = 0.001, smoothing = 0 => run output : sectoral12gen_sig1_theta6_varphi117_eta1_het_hps11_eps-3.mat M4.3 Autarky open : sectoral12_het.m => set indpr = 11, epsil = 1000, smoothing = 0 => run output : sectoral12gen_sig1_theta6_varphi117_eta1_het_hps11_eps3.mat M5. Firm-specific labor market with imperfect risk-sharing within sectors M5.1 Complete market open : sectoral20_rep.m => set indpr = 11, epsil = 0.001(*), smoothing = 0 => run output : sectoral20gen_sig1_theta6_varphi117_eta1_rep_hps11.mat M5.2 Bond economy open : sectoral20_het.m => set indpr = 11, epsil = 0.001, smoothing = 0 => run output : sectoral20gen_sig1_theta6_varphi117_eta1_het_hps11_eps-3.mat M5.3 Autarky open : sectoral20_het.m => set indpr = 11, epsil = 1000, smoothing = 0 => run output : sectoral20gen_sig1_theta6_varphi117_eta1_het_hps11_eps3.mat M6. Firm-specific labor market with perfect risk-sharing within sectors M6.1 Complete market open : sectoral24_rep.m => set indpr = 11, epsil = 0.001(*), smoothing = 0 => run output : sectoral24gen_sig1_theta6_varphi117_eta1_rep_hps11.mat M6.2 Bond economy open : sectoral24_het.m => set indpr = 11, epsil = 0.001, smoothing = 0 => run output : sectoral24gen_sig1_theta6_varphi117_eta1_het_hps11_eps-3.mat M6.3 Autarky open : sectoral24_het.m => set indpr = 11, epsil = 1000, smoothing = 0 => run output : sectoral24gen_sig1_theta6_varphi117_eta1_het_hps11_eps3.mat - For a comparison between model and the empirical results (23 CPI sectors) M7. Redo M1~M4 by setting indpr = 23 to obtain the following outputs. indpr = 23 implies that we use data on 23 CPI sectors in the U.S. from Cotton and Garga (2022). sectoral10gen_sig1_theta6_varphi117_eta1_rep_hps23_mean.mat sectoral10gen_sig1_theta6_varphi117_eta1_het_hps23_eps-3_mean.mat sectoral10gen_sig1_theta6_varphi117_eta1_het_hps23_eps3_mean.mat sectoral10gen_sig1_theta6_varphi117_eta1_rep_hps23_smoothing.mat sectoral10gen_sig1_theta6_varphi117_eta1_het_hps23_eps-3_smoothing.mat sectoral10gen_sig1_theta6_varphi117_eta1_het_hps23_eps3_smoothing.mat sectoral10aitgen_sig1_theta6_varphi117_eta1_rep_hps23.mat sectoral10aitgen_sig1_theta6_varphi117_eta1_het_hps23_eps-3.mat sectoral10aitgen_sig1_theta6_varphi117_eta1_het_hps23_eps3.mat sectoral12gen_sig1_theta6_varphi117_eta1_rep_hps23.mat sectoral12gen_sig1_theta6_varphi117_eta1_het_hps23_eps-3.mat sectoral12gen_sig1_theta6_varphi117_eta1_het_hps23_eps3.mat <> - Datasets E1. Sectoral wage (ce.data.0.AllCESSeries.txt : Current Employment Statistics from the U.S. Bureau of Labor Statistics) run : ce_data.do output : bls_all.dta E2. Controls (controls.xls) run : controls.do output : controls.dta E3. Monetary policy shocks (mpshock_ns.xlsx, mpshock_rr_monthly.dta : Nakamura and Steinsson (2018), Romer and Romer (2004), Wieland and Yang (2020)) run : mpshock.do output : mpshock_combined.dta E4. Sectoral features (sector_feature.xlsx : Cotton and Garga (2022)) run : sector_feature.do output : sector_feature.dta - Panel local projections E5. Merge datasets run : panel_LP_ready.do output : bls_46_sectors_ready.dta E6. Panal LP E6.1 ns_ffr run : panel_LP_run_ns_ffr.do output : panel_LP_results_ns_ffr_cpi_indpro_cpi.xls E6.2 ns_pn run : panel_LP_run_ns_pn.do output : panel_LP_results_ns_pn_cpi_indpro_cpi.xls E6.3 Simulation for 23 CPI sectors (root folder) run : wkrsimul_gamma.m output : wkrsimul.csv (both in the root folder and in "empiric" folder) E6.4 LP with the simulated dataset from E6.3 run : panel_LP_wkrsimul.do output : panel_LP_results_wkrsimul_taylor.xls - Bootstrap the correlation E7. Merge datasets run : bootstrap_ready.do output : bls_46_sectors_bootstrap_ready.dta E8. Bootstrap E8.1 ns_ffr run : bootstrap_run_ns_ffr.do output : bootstrap_results_ns_ffr_indpro_cpi_100.xls E8.2 ns_pn run : bootstrap_run_ns_pn.do output : bootstrap_results_ns_pn_indpro_cpi_100.xls E8.3 rr_romer run : bootstrap_run_rr_romer.do output : bootstrap_results_rr_romer_indpro_cpi_100.xls E8.4 rr_full run : bootstrap_run_rr_full.do output : bootstrap_results_rr_full_indpro_cpi_100.xls We didn't use any special computing equipment. We used only two laptops: - 12th Gen Intel(R) Core(TM) i7-1260P 2.10GHz, 16.0GB RAM - Apple M3 Pro, 18GB RAM We expect the entire set of codes to take about 2-3 days to run, with most of the time attributed to the four bootstrap processes below (each taking less than 10 hours): - bootstrap_run_ns_ffr.do - bootstrap_run_ns_pn.do - bootstrap_run_rr_romer.do - bootstrap_run_rr_full.do