# I am trying to calculate the "spmfedtaxac" var (federal taxes after taking away tax credits) in the CPS SPM.

#1

smpfedtax is total federal tax liability after all tax credits have been accounted for. no matter what combination of vars I try, however, i cannot get the amounts listed. Below is an example:

. list spmid spmfedtaxbc spmfedtaxac spmeitc spmsttax spmfica actccrd ctccrd if spmid == 10001, noobs

spmid spmfedtaxbc spmfedtaxac spmeitc spmsttax spmfica actccrd ctccrd

10001 0 -1140 777 614 2456 363 1637

the amount of tax credits is (777+363+1637), or -2,777 in tax credits. Tax liabilities (fica) are \$2,456 subtracting tax credits from fica I get federal taxes of -\$321. I tried incoporating state taxes, but the math doesn’t work either. This is not a function of using old ASEC values; the data was downloaded after April 2014. Please advise. Thanks!

0 Likes

#2

For the 2012-2014 ASEC samples, you should use SPMEITC and the sum of ACTCCRD values within an SPMFAMUNIT. Keep in mind that ACTCCRD values of 99999 should be recoded to zero. Subtracting SPMEITC and the sum of ACTCCRD from SPMFEDTAXBC equals SPMFEDTAXAC to within \$1 for all respondents in these years.

For the 2010-2011 ASEC samples, you need to also subtract the sum of MWPVAL within an SPMFAMUNIT. For these years, I am unable to match 2.4% of the sample to within \$1 of SPMFEDTAXAC. The small number of cases that differ are either clustered around \$400/\$800 (maximum values of MWPVAL) or have children but were not assumed to have taken a child tax credit. Since the ACTCCRD and MWPVAL variables are not SPM variables, these discrepancies may be caused by the Census Bureau’s update of SPMFEDTAXAC values.

Hope this helps.

0 Likes