. 2022 May 9;22:618. doi: 10.1186/s12913-022-07899-2

Table 2.

Estimated price and income semi-elasticities and elasticities

Variable	(1)	(2)	(3)
Variable	All	Non-informal	Middle-income or higher
Premium semi-elasticity, AME ( $p_{i}$ )	-0.0194***	-0.0206***	-0.0346***
Premium semi-elasticity, AME ( $p_{i}$ )	(0.00352)	(0.00413)	(0.00715)
Income semi-elasticity, AME ( $y_{i}$ )	0.0144***	0.0166***	0.0284***
Income semi-elasticity, AME ( $y_{i}$ )	(0.00170)	(0.00170)	(0.00295)
Observations	41,153	14,378	8320
Proportion who spent on VPHI, E[ ${vphi}_{i}$ ]	0.00416	0.00862	0.0149
Calculated price elasticity ( $ε_{p}$ )	-4.674	-2.389	-2.324
p-value	0.000	0.000	0.000
Calculated income elasticity ( $ε_{y}$ )	3.459	1.921	1.905
p-value	0.000	0.000	0.000

Each column presents the marginal effects associated with the parameters $β_{1}$ and $β_{2}$ for Eq. 1, corresponding to the semi-elasticity concerning the premium and household income value. The estimation of the simultaneous equations model using data from the ENPH2017 includes (i) estimation of a Heckman selection model (household size is the instrument), (ii) estimation of $β_{1}$ and $β_{2}$ using the predictions of the premium considering counts the inverse of mills. VPHI users are defined as those who reported any expenditure above 30 USD in the expense article 12,530,101, “Annual payment of prepaid medicine or complementary health plan.” All models include controls for the head of the household, gender dummies, age groups, and educational level. From the household composition, it is considered whether a person over 65 is present and the income level. There are also region-fixed effects. Standard errors in parentheses. Significance: * 90%, ** 95%, ***99%