Self-reported weather-related home damage has been widely used as a key explanatory variable in empirical models examining the impact of natural disasters on various life outcomes. However, the accuracy of this commonly used measure remains largely unexamined. This gap in the literature is particularly important to address given the growing concerns about the effects of natural disasters and the increasing availability and use of such self-reported data. Moreover, most existing studies treat self-reported weather-related home damage as exogenous to individual behaviour when assessing its effects on health, financial, and well-being outcomes. Whether this assumption holds in practice—and the consequences for estimated relationships if it does not—remains unclear.
Using nationally representative longitudinal data from the Household, Income and Labour Dynamics in Australia (HILDA) Survey, this study documents a previously unreported pattern in self-reported weather-related home damage: in more than half of cases where one member of a household pair reports damage, their co-residing family member does not. This high rate of intra-household inconsistency is striking, particularly given that respondents are asked the same question within a similar timeframe and that prior research typically treats self-reported damage as exogenous to individual behaviour.
Employing a household fixed-effects model, we find that a range of factors—including individual health, life satisfaction, local socio-economic conditions, and exposure to tropical cyclones—are associated not only with the likelihood of reporting home damage but also with the probability of inconsistent reporting within households. The overall pattern suggests that individuals in better health, with higher life satisfaction, or residing in more socio-economically advantaged areas are less likely to report damage—either consistently or inconsistently—relative to their household member. Moreover, our findings indicate that using a more objective measure of home damage substantially attenuates the observed associations between reported damage and individual health and life satisfaction.
Taken together, these results challenge the common assumption of exogeneity in self-reported weather-related damage and underscore the potential for biased inference when endogeneity is not properly addressed.