42/replication_package/data_replication/scf/code/tab3_panelF.do

clear all
set more off
set maxvar 10000
set mem 500m

*Set path
***

*cd ---- specify path to \data_replication\scf\proc_data\
local ourprocdata  "./our_proc"
local output "./output"

use taxsim2001.dta, clear
*use our processed data
*use "`ourprocdata'/taxsim2001.dta", clear


****************************************************************************************
****************************************************************************************

*Rent
rename X708 rent
rename X709 per_rent
replace rent=rent*12 if per_rent==4 //monthly
replace rent=rent*12*4.34 if per_rent==2 //weekly
replace rent=rent*12*2.17 if per_rent==3 //biweekly
replace rent=rent*4 if per_rent==5 //quarterly
replace rent=rent*2 if per_rent==11 //twice per year
replace rent=rent*6 if per_rent==12 //every two months
replace rent=rent*12*2 if per_rent==31 //twice a month
replace rent=. if per_rent==-1 //none
replace rent=. if per_rent==-7 //other

gen HELS=cond(X1005-X1004>0,0,max(X1005,0)) // some people report more debt than they took initially

*Clean
drop J*
drop X*



********************************************************************************
*Create wealth variables
********************************************************************************
*Value of all houses
replace houses=. if houses<0
replace oresre=. if oresre<0
gen house_val=houses
*Homeownership status
gen ownership=housecl

*liquid assets
gen l_assets=max(othfin,0)+max(checking,0)+max(saving,0)+max(mma,0)+max(cds,0)+max(nmmf,0)+max(savbnd,0)+max(stocks,0)+max(bond,0)+oresre+nnresre

*mortgage debt
gen m_debt=max(NH_MORT,0)-max(HELS,0)
drop m_debt
gen m_debt=max(NH_MORT,0)+max(heloc,0)

*HELOC
gen helocs=max(heloc,0)+max(HELS,0)

*liquid debt
gen l_debt=max(ccbal,0)+max(resdbt,0)

*Wealth (net worth)
gen wealth=house_val+l_assets-l_debt-m_debt
gen hnworth=house_val-m_debt

*Age of head
gen age_head=age

*Generate positive house values
gen h1=house_val if house_val>=0

******
replace inc=inc/equiv
replace h1=h1/equiv
******

******************************
*Drop observation with wealth above 80th percentile
******************************
pctile pct = wealth [w=wgt], nq(100) genp(percent)
gen pk1=pct if percent==80
egen mean_pk1=mean(pk1)
drop if wealth>mean_pk1

*Drop households that are too young or too old
drop if age_head<25
drop if age_head>85

********************************************************************************************************************************************8

gen t_debt_paym=max(tpay,0)/equiv
gen rev_debt_paym=max(revpay,0)/equiv
gen mort_debt_paym=max(mortpay,0)/equiv
gen cons_debt_paym=max(conspay,0)/equiv


********************************************************************************
*Compute correlation between mtg payments and all other payments
******************************
gen non_mort=cons_debt_paym+rev_debt_paym
correlate non_mort mort_debt_paym [weight=wgt]
********************************************************************************

********************************************************************************

*10 pct PTI ratio, borrowers
preserve
gen ratio1=mort_debt_paym*12/inc
collapse (p10) ratio1 [weight=wgt] if mort_debt_paym>0
gen res=ratio1*0.6
replace res=round(res, 0.01)
gen title="10 pct PTI ratio, borrowers"
save mpti1.dta, replace
restore

*25 pct PTI ratio, borrowers
preserve
gen ratio1=mort_debt_paym*12/inc
collapse (p25) ratio1 [weight=wgt] if mort_debt_paym>0
gen res=round(ratio1, 0.01)*0.6
replace res=round(res, 0.01)
gen title="25 pct PTI ratio, borrowers"
save mpti2.dta, replace
restore

*Median PTI ratio, borrowers
preserve
gen ratio1=mort_debt_paym*12/inc
collapse (median) ratio1 [weight=wgt] if mort_debt_paym>0
gen res=round(ratio1, 0.01)*0.6
replace res=round(res, 0.01)
gen title="Median PTI ratio, borrowers"
save mpti3.dta, replace
restore

*75 pct PTI ratio, borrowers
preserve
gen ratio1=mort_debt_paym*12/inc
collapse (p75) ratio1 [weight=wgt] if mort_debt_paym>0
gen res=round(ratio1, 0.01)*0.6
replace res=round(res, 0.01)
gen title="75 pct PTI ratio, borrowers"
save mpti4.dta, replace
restore

*90 pct PTI ratio, borrowers
preserve
gen ratio1=mort_debt_paym*12/inc
collapse (p90) ratio1 [weight=wgt] if mort_debt_paym>0
gen res=round(ratio1, 0.01)*0.6
replace res=round(res, 0.01)
gen title="90 pct PTI ratio, borrowers"
save mpti5.dta, replace
restore


forval x=1(1)5{
use mpti`x'.dta, clear
keep res title
duplicates drop
save mpti`x'.dta, replace
}

use mpti1.dta, clear
forval x=2(1)5{
append using mpti`x'.dta
}

order title res

***
cd ..


outsheet using "`output'/tab3_panelF.csv", replace c