Matthew Schaffer and Nimrod Segev, "The Deposits Channel Revisited", 
Journal of Applied Econometrics, Vol. 37, No. 2, 2022, pp. 450-458.

Main replication code: 

1. narrow1.R - Full replication of DSS table 3, table 5 (panel B),
table 6, table 7, table 8 using DSS data.  (Tables 1-2 in the paper
and B.2 - B.5 in the online appendix)

2. narrow2.R - Replication of DSS table 6 and 7 using original data. 
(Tables B.6-B.7 in the online appendix) 

3. wide.R - Performs the wide replication. Tables 1-6 in the paper and
table B9 in the online appendix.  (for convenience tables 1 and 2 in
the paper which are the narrow replication of Tables 6 and 7 in DSS
are also added).

4. Descriptive.R - Plots Figure 1, and Figure B.1 (online appendix).
Also produces table Table B.9.

To run each code:

1. set working directory to folder location.

2. Make sure all required packages are downloaded.

3. Run code.

4. Results will appear in file "results".

*Note! the running time of replicating tables 3 and 5 in "narrow1.R"
can be very long (between 5 to 36 hours each...).

------

data folder:

Includes code and datasets used to compile the final datasets used
in the narrow and wide replications.

1. DSS folder -- the data sets were kindly provided by
Drechsler, Itamar, Alexi Savov, and Philipp Schnabl (DSS). 
"The deposits channel of monetary policy." The Quarterly Journal of
Economics 132.4 (2017): 1819-1876. ANY USE OF THIS DATA SHOULD ALSO
REFERENCE THEIR WORK!

*The following data sets are the data files from DSS:
avgherfdepcty.dta   
cpi_fred_year.dta  
cra_short.dta 
ff_tar_yearend.dta 
fips_char.dta  
sample_flows.dta  
sample_call_final.dta  
bls_final.dta

2. Lerner folder -- This file includes code used to compute bank level
annual Lerner index which is saved in Lerner_data.csv 

3. Original_Data -- includes original datasets used in narrow2 
replication. The datasets are constructed directly from four sources: 
Summary of Deposits (SOD), Call Reports, Community Reinvestment Act 
(CRA), and Federal Reserve Economic Data (FRED).

4. county_controls.csv -- county level controls from the BLS. 

5. mydata_loan.R/mydata_county.R combine data from DSS folder, Lerner
folder and county controls spreadsheet to construct datasets 
"mydata_loan.csv", "mydata_county.csv", "mydata_county_ecredit",
"mydata_county_1bil" , which are used in the wide replication.  

Authors contact information			

Matthew Schaffer				Nimrod Segev
Department of Economics,                        The Bank of Israel,  
University of North Carolina at Greensboro,     Research Department,
maschaff [AT] uncg.edu			        nimrod.segev [AT] boi.org.il