You should receive a 2-line csv file in reply. If not, check the existence
and permissions (Linux and OSX) on the executable file.
For taxsim.wasm the command would be:
cat txpydata.raw | node -e "requires(taxsim.js)"
Format of submissions
taxsim35 expects to read a csv file with taxpayer data. The first row will
be the variable names as documented at
href="http://taxsim.nber.org/taxsim35 and each of the remaining rows
corresponds to a tax return. In addition to the taxpayer characteristic
variables two control variables are available.
Additional Controls
If the default marginal tax rate calculation or level of detail is not
to your liking, it is possible to change either with additiona variables.
These values of mtr affect the marginal tax calculation:
- 11 - Wages
- 70 - Long term capital gains
- 85 - Primary wage earner
- 86 - Secondary wage earner
These values of idtl affect the output level:
- 0 - standard
- 2 - full
- 5 - with text description
Here is an example asking for the full (with many intermediate calculations) with several
different marginal tax rates:
taxsimid,mstat,year,ltcg,mtr,idtl
1,2,1970,100000,11,2
1,2,1970,100000,70,2
1,2,1970,100000,85,2
1,2,1970,100000,86,2
idtl can't vary across observations.
Problems
The most likely problem is:
taxsim35: not found
or something similar. Other problems would be non-decimal characters in
the data (such as missing value codes) or non-ascii data. Check your data
in notepad or similar. If there are data problems, an explicit error
message will be output to the standard output and execution will cease.
Source Code
Unlike the programs of professional programmers, taxsim35.for has no
dependencies, uses no non-standard libraries and will compile on any
computer with any fortran compiler with a single command such as:
gfortran taxsim35.for -o taxsim35.exe
If you need source code I can supply that. It is a single f90 file.
Example using Taxsim with SAS
data txpydata;
input taxsimid mstat year ltcg;
datalines;
1 2 1970 100000
;
run;
filename txpydata "txpydata.csv";
filename results "results.csv" ;
x "rm results.csv"
proc export outfile="txpydata.csv" dbms=csv replace;
run;
x taxsim.exe results.csv;
proc import datafile="results.csv" out=results dbms=csv;
run;
proc print data=results; run;
You will need to replace "rm" with "del" on a Windows system.
Last modified 13 June 2024 by feenberg@nber.org