Table 5.

Subsample results.

	Formal home care	Informal home care	Informal home care, child/children	Informal home care, daughters	Informal home care, sons
Social Security income (thousands)	0.059** (0.030)	−0.006 (0.046)	−0.087** (0.040)	−0.039 (0.030)	−0.059** (0.029)
SSI* married	−0.054* (0.029)	0.001 (0.045)	0.086** (0.039)	0.039 (0.029)	0.057** (0.029)
Social Security income (thousands)	0.022** (0.010)	−0.007 (0.016)	−0.030** (0.013)	−0.011 (0.011)	−0.021** (0.009)
SSI* nonwhite	−0.015** (0.007)	0.019 (0.017)	0.036*** (0.013)	0.006 (0.010)	0.029*** (0.009)
Social Security income (thousands)	0.056** (0.028)	−0.051 (0.058)	−0.112* (0.063)	−0.024 (0.035)	−0.087** (0.043)
SSI* less than high school	−0.065* (0.037)	0.085 (0.075)	0.157* (0.083)	0.024 (0.049)	0.127** (0.052)
Observations	6836	6836	6836	6836	6836

This table reports results of the estimated effects of Social Security income from three separated estimation equations. The endogenous variables of interest are Social Security income and the interaction terms of Social Security income and the socio-demographic characteristic of interest. The instruments are an indicator variable for respondents in households in which the primary beneficiary was born during 1911–1917 and the interaction terms of a dummy for the beneficiary was born during 1911–1917 and the socio-demographic characteristic of interest. All regressions are weighted using the Final Annual LSOA II weight. The set of covariates in all estimations includes the control variables listed in Table 1. Robust standard errors are listed in parenthesis and are clustered at the birth year of the respondent.

^*Corresponds to statistical significance at the 10% level.

^**Corresponds to statistical significance at the 5%level.

^***Corresponds to statistical significance at the 1% level.