Table 6 reports the results using an alternative measure of retirement village dummy. We add industry dummy variables to replace the industry-adjusted measurements on the industry-sensitive variables with the natural logarithm of one plus their original data (no industry adjustment). Control variables are the same as in Table 3. We estimate the regression with year fixed effects. The standard errors in parentheses are clustered at industry level. Continuous variables are winsorised at the 1st and 99th percentiles. *, ** and *** denote significance at the 10, 5 and 1% levels, respectively