Row A of the table, which presents the coefficient on the log SEHI tax price, shows the effect of the change in the price of health insurance while self-employed on the probability of being self-employed for the states with light regulatory regimes, while Rows B and C show the interaction of the log SEHI tax price with dummy variables for the taxpayer's state having a moderate or heavy regulatory regime, respectively.
In the first column, the coefficient on the log SEHI tax price is--0.
In both of these columns, the coefficient on the uninteracted log SEHI tax price is significant and of a similar magnitude to the base specification.
The coefficient on the log SEHI tax price in this column is estimated to be a highly significant--0.
The coefficient for log SEHI tax price in this column is--0.
When the results in this table are taken together, it appears that the decrease in the SEHI tax price led to a smaller increase in heavily regulated states compared to lightly regulated states in taxpayers reporting some self-employment income, but may have led to a larger increase in heavily regulated states compared to lightly regulated states in taxpayers being exclusively self-employed.
In the some self-employment income specification, it appeared that the states of Maine, New York, and Washington were driving the result that the decrease in the SEHI tax price had less of an effect in heavily than in the lightly regulated states.
TABLE 5 Instrumental Variables Regression Results for the Probability of Having Any Self-Employed Earnings by Age Some SE Exclusively Income SE Income Dependent Age Age Age Age Age 40-54 Age Variable 25-39 40-54 55-64 25-39 55-64 Log SEHI tax -0.
In the some self-employment income specification, the estimated coefficients on the log SEHI tax price alone (which represents the effect for lightly regulated states) are correctly signed and significant among all three age groups.
In the exclusively self-employed specification, the coefficient on the log SEHI tax price alone is insignificant for all age groups.
Taken together, our results suggest that in states with community rating and guaranteed issue regulations, higher insurance prices for the youngest group may have diminished the effect of the decline in the SEHI tax price on the propensity to earn some self-employment income, while relatively lower insurance prices for the oldest group may have enhanced the effect of the decline in the SEHI tax price on the propensity to be exclusively self-employed.
On the other hand, comparing heavily to lightly regulated states we find suggestive evidence of larger effects of the decrease in the SEHI tax price on the probability of being exclusively self-employed.