Data and Code for: ANATOMY OF THE PHILLIPS CURVE: MICRO EVIDENCE AND MACRO IMPLICATIONS

– Readme – Replication Package Anatomy of the Phillips Curve: Micro Evidence and Macro Implications Luca Gagliardone Mark Gertler Simone Lenzu Joris Tielens Contents 1 Overview 2 2 Code reuse policy 3 3 Data availability and provenance statements 3 3.1 Statementaboutrights ............................ 3 3.2 Summary of data availability . . . . . . . . . . . . . . . . . . . . . . . . . 3 3.2.1 Publicly available data . . . . . . . . . . . . . . . . . . . . . . . . 3 3.2.2 Non-public, confidential data . . . . . . . . . . . . . . . . . . . . 10 3.2.3 Pseudodata.............................. 12 4 Affirmation of support for replication checks 12 1 5 Computational requirements 12 5.1 Hardware.................................... 12 5.2 Software .................................... 13 6 Instructions to replicators 13 6.1 Initial folder structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 6.2 Description raw confidential datasets . . . . . . . . . . . . . . . . . . . . 16 6.3 Replicationsteps................................ 17 6.4 Runtime .................................... 18 6.5 Code objective and interdependencies . . . . . . . . . . . . . . . . . . . . 18 7 List of figures and tables 19 1 Overview This ReadMe file serves as guidance for the replication package for “Anatomy of the Phillips Curve: Micro Evidence and Macro Implications” by Luca Gagliardone, Mark Gertler, Simone Lenzu and Joris Tielens. The replication package contains all the code necessary to replicate the tables and figures included in the paper. Aside from the computer code, the package also contains: 1. hard copies of the non-confidential (public) data collected for the project; 2. a description of the confidential data (the confidential data themselves are not included); 3. pseudo data – “artificial data” – randomly generated to mimic the data structure of the actual confidential data. It serves to test and assist anyone interested in replicating the results in the paper and understanding the data structure. 2 2 Code reuse policy The code included in this replication package may be reused for academic and research purposes. If you use any part of this package in your own work, please cite the original paper. 3 Data availability and provenance statements 3.1 Statement about rights We certify that the authors of the manuscript have legitimate access to and permission to use the data utilized in this manuscript. The replication package draws on public (nonconfidential) data as well as non-public (confidential) data. Confidential data. Access to the non-public datasets (infra) is governed by legal decision documents “Verzoek tot het ter beschikking stellen van individuele bedrijfsgegevens voor onderzoeksproject dienst studieën – 2022-JT-040” and “Verzoek tot het ter beschikking stellen van individuele bedrijfsgegevens voor onderzoeksproject dienst studieën – 2023-JT052”). Public data. We certify that the authors of the manuscript have documented permission to redistribute/publish the public data contained within this replication package. 3.2 Summary of data availability The results in the paper are based on a dataset that merges different sources. This section summarizes the data availability and access protocols. 3.2.1 Publicly available data A copy of the publicly available data is provided as part of this replication package. These data are in the public domain and available free of charge. For researchers willing to refresh and/or extend these datasets, Table 1 also points to the online repository of the data 3 (note, however, that our replication package is not guaranteed to accommodate potential revisions in the structure of these datasets). Table 1: Public data Dataset Access and data manipulation details COMEXT A hard copy of the data is provided as part of the replication package (the free-reuse and dissemination policy of the data is stated here.). The data are in the public domain and can be downloaded free of charge from the Eurostat website in directory “COMEXT_HISTORICAL_DATA" and “COMEXT_DATA". In the replication package, the individual files are named «full199501.zip», «full199502.zip»,...,«full202112.zip». Data reference: Eurostat (1995–2021). Deflators A hard copy of the data is provided as part of the replication package (the free-reuse and dissemination policy of the data is stated here.). The data are in the public domain and can be downloaded free of charge. These steps are as follows: 1. Go to «NBB stat»→«National Accounts» →«Detailed Accounts» →«Major components by branch and sector». This renders the table of interest. 2. Before downloading, adjust the time frame of the dataset. Go to «Customise» →«Selection» →«Time and Frequency» →«Select data range» →«From 1995 – 2022». From this extraction, the four deflators are created as follows: Continued on next page 4 Table 1 – Continued from previous page Dataset Access and data manipulation details a. «ppi_intermediates»: Divide the value and quantity of the variable “Intermediair verbruik (tegen aankoopprijzen) (P.2)", per NACE 2D, to back out a sectoral intermediate input price index. Set 1995 as the base year. b. «ppi_gross_output»: Divide the value and quantity of the variable “Output (tegen basisprijzen) (P.1)", per NACE 2D, to back out a sectoral gross output price index. Set 1995 as the base year. c. «ppi_investments»: Divide the value and quantity of the variable “Investeringen in vaste activa (bruto) (P.51)", per NACE 2D, to back out a sectoral investment price index. Set 1995 as the base year. d. «wage_index»: Divide the nominal wage "Beloning van werknemers (D.1)", per NACE 2D, by employment. The latter is obtained as follow: 1. Go to «NBB stat»→«Population and Labour market» →«Employment» →«Annual detailed data». This renders the table of interest. Continued on next page 5 Table 1 – Continued from previous page Dataset Access and data manipulation details 2. Adjust the time frame of the dataset. Go to «Customise» →«Selection» →«Time and Frequency» →«Select data range » →«From 1995 – 2022». 3. To back out a sectoral wage price index, set 1995 as the base year. In the replication package, the individual files are named «Deflator_Sector_Year.dta» and «Deflator_Year.dta». Data reference: National Bank of Belgium (2022d). Oil shocks A hard copy of the data is provided as part of this replication package (see infra). They are obtained from Känzig (2021). In the replication package, the individual file is named «oilSupplyNewsShocks_2022M06.xlsx». The file is also publicly available and free of charge on GitHub. Data reference: Känzig (2022). Money shocks The Euro Area Monetary Policy Event-Study Database can be downloaded from website of the ECB – the reuse statement can be found here – and are produced following Altavilla et al. (2019). In the replication package, the individual file is named «Dataset_EAMPD.xlsx». Data reference: Altavilla et al. (n.d.). Continued on next page 6 Table 1 – Continued from previous page Dataset Access and data manipulation details Capital depreciation rates A hard copy of the data is provided as part of this replication package. They are obtained from Lenzu and Manaresi (2018a). In the replication package, the individual file is named «depreciation_rates_age_capital.dta» The data can be used free of charge, conditional on citing the original research article. These data can also independently be obtained emailing the author, Francesco Manaresi (francesco.manar[email protected]). Data reference: Lenzu and Manaresi (2018b). Product code conversion tables A hard copy of the data is provided as part of the replication package. In the replication package, these files are called «CN_to_CN.dta» , «CN_to_PRODCOM_1995.dta», · · · , «CN_to_PRODCOM_2021.dta». The data are in the public domain and can be downloaded free of charge from the RAMON website of Eurostat. The free-reuse and dissemination policy of the data is stated here. This website is archived and no longer updated. Future releases are found on European Union publication office. Finally, we also rely on a NACE 2003 – 2008 conversion table, «nace_2003_2008_concordance.dta.», which can be downloaded here. Data reference: Eurostat (n.d.). Exchange rates Steps to the online repository: 1. Go to «IMF data portal». Continued on next page 7 Table 1 – Continued from previous page Dataset Access and data manipulation details 2. Go to «International Financial Statistics» 3. Under «Data Tables» Tab →«Data Tables by Indicator» → «Exchange Rates incl. Effective ex. Rates» 4. In the data tool a) select indicator «National Currrency per U.S. Dollar, period average»; b) select «Quarterly» tab and desired years; c) download as xlsx. 5. Fill in currency values for countries after they convert to the EURO with export country "Euro Area" as follows: a) Fill 1999 and after: "Austria", "Belgium", "BelgiumLuxembourg", "Netherlands, The", "Finland", "France", "Germany", "Ireland", "Italy", "Portugal", "Spain", "Luxembourg", "Guadeloupe", "Guiana, French", "Martinique", "Reunion", "Saint Pierre and Miquelon", "Andorra, Principality of" b) Fill 2001 and after: "Greece" c) Fill 2007 and after: "Slovenia, Rep. of" d) Fill 2008 and after: "Cyprus", "Malta" e) Fill 2009 and after: "Slovak Rep." Continued on next page 8 Table 1 – Continued from previous page Dataset Access and data manipulation details f) Fill 2011 and after: "Estonia, Rep. of" g) Fill 2014 and after: "Latvia" h) Fill 2015 and after: "Lithuania" i) Fill 2023 and after: "Croatia, Rep. of" 6. Convert the US exchange rate to Belgium by dividing the XR to USD as follows: (XR Belgium/XR USD) where XR Belgium is the given quarterly exchange rate in a given year/quarter for the country "Belgium". This will be equivalent to the "Euro Area" exchange rate in 1999 and after. In the replication package, the individual file is named «Exchange_Rates_incl_Effective_Ex_Ra.xlsx». The data free-reuse and dissemination policy is stated here. We also include the file «cty_name.dta», which includes the ISO 3166-1 Alpha-2 Country Codes. Data reference: IMF (1995–2021). Estimates of production functions. Estimates are based on Lenzu et al. (2025a). In the replication package, these files are called «LRTH_CD.dta» and «LRTH_TL.dta». The data can be used free of charge, conditional on citing the original research article. These data can also independently be obtained by emailing the author, David Rivers ([email protected]du). Data reference: Lenzu et al. (2025b). 9 6.2 Description raw confidential datasets Table 3 lists the confidential datasets and offers a brief description. The variables in the underlying files are labeled for further clarification. Table 3: Dataset description Dataset Description imports_1995_NAT_quarterly.zip This dataset contains quarterly import flows for the year 1995. The folder contains similar vintages for each year up to 2021. VAT_quarterly.zip This dataset contains quarterly value added tax returns of the firms in our analysis. VAT_yearly.zip This dataset contains annual value added tax returns of the firms in our analysis. SSEC_quarterly.zip This dataset contains quarterly social security declarations of the firms in our analysis. SSEC_yearly.zip This dataset contains annual social security declarations of the firms in our analysis. annual_accounts.zip This dataset contains the annual accounts of firms. prodcom_monthly.zip This file contains the monthly extraction of the PRODCOM survey submission. prodcom_quarterly.zip This file contains the quarterly extraction of the PRODCOM survey submission (i.e., an aggregation of the monthly). prodcom_yearly.zip This file contains the annual extraction of the PRODCOM survey submission (i.e., an aggregation of the quarterly). Continued on next page 16 Table 3 – Continued from previous page Dataset Description prodcom_monthly_domestic.zip This file contains the monthly extraction of the PRODCOM survey submission, netting out exports. prodcom_quarterly_domestic.zip This file contains the quarterly extraction of the PRODCOM survey submission, netting out exports. prodcom_yearly_domestic.zip This file contains the annual extraction of the PRODCOM survey submission, netting out exports. capacity_utilization_quarterly.dta Capacity utilization rates of individual firms over time. 6.3 Replication steps The replicator should take the following steps to reproduce the tables and figures in the manuscript: 1. Open the Stata file «0_master.do»; (a) Modify the path in the global «main_dir» to the working directory of the project; (b) If you have access to the confidential data, change the value of the global «data_type» to "confidential". Otherwise, keep its default value "pseudo"; (c) Save «0_master.do». 2. Run the do file «0_master.do». It automatically calls all subsequent do files; 3. Results are stored in the «Output» folder, which is created by the Stata scripts. 17 6.4 Runtime The code was run on a machine with the specifications described in Section 5 and took approximately 11 hours to run. Code Approximate runtime (seconds) 1_firm_data.do. 1,199 2_price_indices.do. 35,398 3_final_datasets.do. 695 4_tables_figures.do. 2,447 Total 39,739 Table 4: Stata do files: runtime 6.5 Code objective and interdependencies Table 5: Summary and objective of the code Code Objective «0_master.do» This code sets directories, globals and launches the sequence of do files to produce the datasets required for the analysis. «1_firm_data.do» This code creates an unbalanced panel of firm variables at the quarterly frequency. It contains i.a. firm turnover, costs (wages, intermediates), capital stock (using the perpetual inventory method) and firm metadata (e.g., sector of activity, age, etc.). Continued on next page 18 Table 5 – Continued from previous page Code Objective «2_price_indices.do» This code creates the firm-level price index, the firm-industry price index, the firm-industry competitors’ price index, and price indices using COMEXT data. «3_final_datasets.do» This code creates the instrument for marginal costs and the instrument for competitors’ prices. Together with the results from the previous do files and a set of filters, it generates the final datasets on which the analysis is based. «4_tables_figures.do» This code creates the tables and figures in the draft. 7 List of figures and tables All tables and figures are produced by the do file «4_tables_figures.do». In the following table, we highlight the approximate location in the Stata do file. Table 6: List of tables and figures Table/Figure Caption Location Table 1 Summary statistics Line 408 Table 2 Estimation results Line 560 Table 3 Robustness exercises and assessment of identification threats Line 990 Figure 1 Aggregate inflation dynamics Line 1576 Table 4 Estimates of the output-based slope Line 1423 Figure 2 Dynamic effects of oil shocks on real marginal costs and prices Line 1657 Continued on next page 19 Table 6 – Continued from previous page Table/Figure Caption Location Table A.1 List of manufacturing sectors N.A. Table A.2 Estimates of output elasticities and returns to scale Line 1794 Figure A.1 Dynamic effects of oil and money shocks Line 1850 References Altavilla, C., L. Brugnolini, R. S. Gürkaynak, R. Motto, and G. Ragusa (2019): “Measuring euro area monetary policy,” Journal of Monetary Economics, 108, 162–179. ——— (n.d.): “Euro Area Monetary Policy event study Database (EA-MPD),” https:// www.ecb.europa.eu/pub/pdf/annex/Dataset_EA-MPD.xlsx, European Central Bank (distributor), Accessed 2023-10-05. Eurostat (1995–2021): “Comext,” https://ec.europa.eu/eurostat/ databrowser/bulk?lang=en&selectedTab=fileComext, Accessed 2023-03-11. ——— (n.d.): “Ramon Classification,” https://ec.europa.eu/eurostat/ web/metadata/classifications, Accessed 2023-11-11. FPS Economy (2022): “Kruispuntbank van Ondernemingen (Crossroad Bank for Enterprises),” National Bank of Belgium (distributor), Accessed 2022-11-27. FPS Finance (2022): “Value Added Tax Returns,” National Bank of Belgium (distributor), Accessed 2022-11-15. FPS Finance: Customs and Excise (2022): “Extrastat,” National Bank of Belgium (distributor), Accessed 2022-11-15. IMF (1995–2021): “International Financial Statistics,” https://www.imf.org/ en/Data, Accessed 2023-09-26. 20 Känzig, D. R. (2021): “The macroeconomic effects of oil supply news: Evidence from OPEC announcements,” American Economic Review, 111, 1092–1125. ——— (2022): “Oil supply surprises and news shocks,” Github. https://github. com/dkaenzig/oilsupplynews, Accessed 2023-10-05. Lenzu, S. and F. Manaresi (2018a): “Sources and implications of resource misallocation: New evidence from firm-Level marginal products and user costs,” Working paper. ——— (2018b): “Sources and implications of resource misallocation: New evidence from firm-Level marginal products and user costs: Data on capital depreciation rates,” Unpublished data, Accessed 2022-11-15. Lenzu, S., D. Rivers, J. Tielens, and H. Shi (2025a): “Financial Shocks, Productivity, and Prices,” Working paper. ——— (2025b): “Financial Shocks, Productivity, and Prices: Data on production function estimates,” Unpublished data, Accessed 2022-11-15. National Bank of Belgium (2022a): “Balanscentrale (Balance sheet central),” National Bank of Belgium (distributor), Accessed 2022-11-15. ——— (2022b): “Conjunctuurenquête (Business Survey),” National Bank of Belgium (distributor), Accessed 2022-11-15. ——— (2022c): “Intrastat,” National Bank of Belgium (distributor), Accessed 2022-11-15. ——— (2022d): “Quarterly National Accounts,” https://stat.nbb.be/?lang= en, National Bank of Belgium (publisher), Accessed 2022-11-15. National Social Security Office (2022): “National Social Security Declarations,” National Bank of Belgium (distributor), Accessed 2022-11-16. Statbel (2022): “List of PRODucts of the European COMmunity,” National Bank of Belgium (distributor), Accessed 2022-11-15. 21