Read Me Documentation "Cyclical Lending Standards: A Structural Analysis??? by K. Chen, P. Higgins, and T. Zha ZIP FILE 1: PublicationData.zip ------------------------------------------------------------------------- 1. The micro SLOOS is confidential data provided by the The Board of Governors of the Federal Reserve System with the agreement that we cannot reveal the code and the data publicly for the following results: Tables 1-3, Figures 2-5 (plotted by Figure2Figure5RED.m---see the description of OtherDataSoftware.zip below). 2. The computer code and the data that is not related to confidential micro SLOOS are provided as follows. - data_logData_all.prn: Ascii time series data used for Figures 1 and 8. Inside this file, we replace actual values for the following columns //10: LSCI (Standardized lending standards for C&I loans) //12: LSCIlevel (Lending standards for C&I loans, black line top panel of Figure B.1) //14: diffLSCIlevel (First difference of LSCIlevel) with the value -999 because our confidentiality agreement with the Federal Reserve Board precludes us from publishing this data. - SRC_pub.zip: C/C++ code for solving the model and for generating results for Tables 4-7 and Figure 9. - PlotIRFsBands.zip: Zipped file for plotting Figure 9 using the simulated data generated by SRC_pub.zip. - CLibraryMaster.zip: Zipped file for C/C++ library to be read and linked by SRC_pub.zip. - Runscript.zip: Zipped script file for running the compiled executable by SRC_pub.zip. 3. The total time for running Runscript.zip (including the posterior draws) takes about 34 hours. ZIP FILE 2: OtherDataSoftware.zip ------------------------------------------------------------------------- Figure1FigureB1RED.m plots data shown in figures 1 and B.1, except for the black solid black loan weighted C&I lending standards line shown in the top panel of figure B.1 that we cannot release to the public as part of our SLOOS microdata agreement with the Federal Reserve Board in order protect the confidentiality of the SLOOS respondents. Figure2Figure5RED.m plots modified versions of figures 2 and 5 using publicly available aggregated SLOOS data. We cannot release the data used to construct the two figures in the manuscript as part of our confidentiality agree with the Board. However, since 2008, the publicly released quarterly SLOOS reports have included aggregated data underlying figures 2 and 5. For example, see the tabulated responses to questions 3Aa and 3Ab and the responses to to questions 3Ba and 3Bb on pages 20-23 of https://www.federalreserve.gov/boarddocs/snloansurvey/201511/fullreport.pdf Figure2Figure5RED.m reads in and plots this publicly available data. Figure4RED.m reads in and plots data for figure 4. The following 3 mat files are read in from this program callData090919.mat: Call report data read in and constructed in ReadCallReportData111920.m GDPCapStockData112020.mat: Macroeconomic data read in and constructed ReadHaverGDPCapStock112020.m QtrlyData071019.mat: Quarterly smoothed population data read in and constructed in SmoothPopulationBigMelinda112020.m --- These mat files are included in the zip file. Note 1: Code for generating figure 1, portions of figure 2 and 5 starting in 2008, figure 8, and portions of figure 1B are included in the programs Figure1FigureB1RED.m, Figure2Figure5RED.m, and Figure4RED.m. These can be run in any order, and input files read by these programs are included in this zip file. To create input files for these programs, user needs to run first 1228 lines of ReadWriteCallOtherData11-23-20.sas, and then subsequently run these three Matlab programs in any order ReadCallReportData111920.m [runtime approximately 5-10 seconds] ReadHaverGDPCapStock112020.m [runtime approximately 5-10 seconds] SmoothPopulationBigMelinda112020.m [runtime approximately 10 minutes] --- These 3 programs are described in more detail below. The code was run on 64 bit Windows operating system with Matlab R2020a and SAS version 9.4 for Windows. Note 2: The processed/aggregated Call Report data are in the files FFIECCallStackSumsSep11th2019.xlsx and CALLstackCoreLoansSumsSep11th2019.xlsx. These files can be created by running the first 1228 lines of the program ReadWriteCallOtherData11-23-20.sas. To run ReadWriteCallOtherData11-23-20.sas [runtime approximately 10 minutes], user needs to have and/or assemble data described below. Call report data from the Chicago Fed: Available at https://www.chicagofed.org/banking/financial-institution-reports/commercial-bank-data User needs to download these files to a folder called C:\ChicagoFedCall2019\ 1983 -2000 Complete Call Reports available at https://www.chicagofed.org/banking/financial-institution-reports/commercial-bank-data-complete-1976-2000 2001 -2010 Complete Call Reports available at https://www.chicagofed.org/banking/financial-institution-reports/commercial-bank-data-complete-2001-2010 2011q1-2019q1 Commercial bank structure files available at https://www.chicagofed.org/banking/financial-institution-reports/commercial-bank-structure-data Note 3: The Chicago Fed only has Call Report bank-level financial data through 2010. Data after 2010 are available from the FFIEC. These data can be downloaded in zip-file format at https://cdr.ffiec.gov/public//PWS/DownloadBulkData.aspx To run the code, the user needs to download a "Call Reports -- Single Period" zip file in tab delimitted format for each quarter between 12/31/2010 and 03/31/2019. This will be a total of 34 zip files. The user should download each zip file to the directory they create called C:\ChicagoFedCall2019\FFIEC\ and unzip each zip file. Each zip file will contain roughly 45 tab delimitted files. A subset of these tab-delimitted files need to be reformatted and saved as an excel file to be read into SAS. FFIEC CDR Call Schedule RCCI ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCCII ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCL ddmmyyyy(1 of 2).txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCL ddmmyyyy(2 of 2).txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RI ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RIA ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RC ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCK ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCN 12312010.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCN ddmmyyyy(1 of 2).txt [ddmmyyyy each end-of-quarter between 03312011 and 03312019] FFIEC CDR Call Schedule RCN ddmmyyyy(2 of 2).txt [ddmmyyyy each end-of-quarter between 03312011 and 03312019] FFIEC CDR Call Schedule RCE ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCR ddmmyyyy(1 of 2).txt [ddmmyyyy each end-of-quarter between 12312010 and 12312013] FFIEC CDR Call Schedule RCR ddmmyyyy(2 of 2).txt [ddmmyyyy each end-of-quarter between 12312010 and 03312019] FFIEC CDR Call Schedule RCRIA ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 03312014 and 12312014] FFIEC CDR Call Schedule RCRIB ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 03312014 and 12312014] FFIEC CDR Call Schedule RCRI ddmmyyyy.txt [ddmmyyyy each end-of-quarter between 03312015 and 03312019] Note 4: Open each tab delimitted file into excel as a tab delimitted file. Row 1 will contain the variable names and row 2 will contain variable labels. Suppose there are N nonblank columns in the spreadsheet. Insert 10 rows between row 2 and row 3. For each of these 10 rows, enter the number -999 in each of the 10 inserted rows for each of the first 10 columns. Then change the tab name of the excel file to "Sheet1" and save the excel file using exactly the same name as the input filename except the .txt extension changes to a .xlsx extension. For example, "FFIEC CDR Call Schedule RCE 12312017.txt" is modified and saved as "FFIEC CDR Call Schedule RCE 12312017.xlsx." We've included "FFIEC CDR Call Schedule RCE 12312017.xlsx" and "FFIEC CDR Call Schedule RCE 12312017.txt" in the zip files as an example. The xlsx files can also be requested from patrick.higgins@atl.frb.org. Note 5: With these datasets, users can run first 1228 lines of ReadWriteCallOtherData11-23-20.sas. After line 1228 of code in ReadWriteCallOtherData11-23-20.sas, the program merges call report with other data. Original code also merged in processed/transformed SLOOS microdata and ran SLOOS regressions reported in Tables 1-3, but that code has been removed since confidentiality agreements do not allow us to provide this code. We have also removed code that processes/transforms raw SLOOS microdata from the Board since confidentiality agreement does not allow us to provide this code. Code in this SAS program after line 1228 not likely to be helpful to users unless they can get access to SLOOS microdata Note 6: At the time of this writing, the Federal Reserve economist in charge of the SLOOS microdata is Felicia Ionescu https://www.federalreserve.gov/econres/felicia-f-ionescu.htm. To obtain access to this confidential microdata, users would have to contact her. In our case, access to the data was granted conditional on approval of a project proposal, agreement to a number of conditions protecting confidentiality of the microdata, at least one of the project participants being an economist in the Federal Reserve System, and approval from the bank's Director of Research. Datasets not included in this analysis that user needs to have and/or download to rund code after line 1228 in ReadWriteCallOtherData11-23-20.sas. Note 7: Bank holding company data from the Chicago Fed are available at https://www.chicagofed.org/banking/financial-institution-reports/bhc-data https://www.chicagofed.org/applications/bhc/bhc-home Note 8: Users should download each zip file to C:\ChicagoFedCall2019\ for each quarter between 1989q1 and 2019q1 Then run the included program ImportHoldingCompanyData111920.sas. This will create a SAS dataset called BHCstackB.sas7bdat in the same directory that will be used in the subsequent program ReadWriteCallOtherData11-23-20.sas Note 9: This holding company data is only used for the Senior Loan Officer Opinion Survey (SLOOS) lending standards regression. Unless the user is able to obtain access to confidential SLOOS microdata, this data will not be useful for constructing any of the tables or figures in the paper. Note 10: Data on bank mergers included in file CSV_TRANSFORMATIONS.xlsx [should be put in C:\ChicagoFedCall2019\CSV_TRANSFORMATIONS.xlsx]. Original data were downloaded from the zipped transformations table available at https://www.ffiec.gov/npw/FinancialReport/DataDownload. This data on bank mergers is only used for the Senior Loan Officer Opinion Survey (SLOOS) lending standards regressions. Unless the user is able to obtain access to confidential SLOOS microdata, this data will not be useful for constructing any of the tables or figures in the paper. Note 11: Center for Research in Security Prices (CRSP) data (not included): Assumed to be saved in file name and location C:\ChicagoFedCall2019\Crspeom80to18Raw.sas7bdat. This data should range from 1/1/1980 to 12/31/2018, aAvailable with subscription at: https://wrds-www.wharton.upenn.edu/pages/about/data-vendors/center-for-research-in-security-prices-crsp/ This CRSP data is only used for the Senior Loan Officer Opinion Survey (SLOOS) lending standards regressions; unless the user is able to obtain access to confidential SLOOS microdata, this data will not be useful for constructing any of the tables or figures in the paper. Variable Type Label ACCOMP Num Acquiring PERMCO ACPERM Num Acquiring PERMNO ALTPRC Num Price Alternate ALTPRCDT Num Alternate Price Date COMNAM Char Company Name CUSIP Char CUSIP Header DATE Num Names Date NAMEENDT Num Names Ending Date NWPERM Num New PERMNO PERMCO Num PERMCO PERMNO Num PERMNO PRC Num Price or Bid/Ask Average SHRCD Num Share Code SHRCLS Char Share Class SHRENDDT Num Shares Observation End Date SHROUT Num Shares Outstanding TICKER Char Ticker Symbol TSYMBOL Char Trading Symbol VOL Num Volume Note 12: CRSP-FRB Link included in this zip file as crsp_20171231.xlsx. Should be saved in C:\ChicagoFedCall2019\ directory. Also available at https://www.newyorkfed.org/research/banking_research/datasets.html (2017-4 version used for paper) and https://www.newyorkfed.org/medialibrary/media/research/banking_research/data/crsp_20171231.csv?la=en