Discrepancy between economic and subjective poverty in Hong Kong : the influence of comparison, household consumption, and neighbourhood services

Abstract

Although the economic approach to measuring poverty via monetary indicators has dominated the poverty research, a subjective approach that asks people to assess their own poverty status has been increasingly used to get a more complete picture of poverty. A systematic review found that existing studies failed to explain the variation in subjective poverty among those with the same economic poverty status. The question why people feel poor (non-poor) even if they are economically non-poor (poor) is not fully understood. Understanding the discrepancy between economic and subjective poverty not only enriches knowledge of the income- wellbeing paradox, but also points out directions for social policies to improve people’s subjective wellbeing. This thesis investigates the factors associated with the discrepancy between economic and subjective poverty in Hong Kong, with a focus on the three important, yet understudied, aspects: comparison, household consumption, and neighbourhood services. A multidimensional resource framework, with three kinds of resources—social status, money, and service, is proposed to explain how the comparison, allocation and compensation (depletion) effect of resources may influence the discrepancy in poverty status. In the individual dimension, people compare their social status with other people. In the household dimension, households allocate money into different consumption categories, leading to varying level of utility. In the neighbourhood dimension, different access to neighbourhood services compensates for (depletes) the scarcity (abundance) of resources at the individual-or-household levels. Using data drawn from the first wave (2015) of the Hong Kong Panel Survey for Poverty Alleviation (HKPSPA), sub-study I employed logistic regression to examine the influence of comparison with one’s parents and friends on the discrepancy in poverty status. The results showed those who have experienced downward mobility compared with their parents were more likely to be subjectively poor even if they were economically non-poor. No significant association was found for the comparison with friends. Sub-study II utilised data from HKPSPA to investigate the influence of household consumption. For type of consumption, results of logistic regression suggested greater expenditure on leisure increased (reduced) the probability of feeling non-poor (poor) among the economically poor (non-poor). Multiple mediation analysis further revealed this association was mediated by self-perceived social status but not social connectedness. For consumption patterns, results of cluster analysis and logistic regression showed a balanced consumption pattern increased the probability of feeling non-poor among the economically poor, and a mortgage-high pattern increased the probability of feeling poor among the economically non-poor. Sub-study III combined the individual-level data from HKPSPA and neighbourhood-level data from Hong Kong Census and Statistics Department to investigate the influence of access to neighbourhood services. The results of multilevel logistic regression suggested access to public spaces, such as self-study rooms and rest garden/parks, were associated with the discrepancy in poverty status. Based on the findings, it is recommended that policymakers to adopt subjective poverty as an indicator to measure citizens’ subjective wellbeing. Social policies that promoting youth mobility, enhancing equal access to cultural, sports and recreational activities, providing affordable housing, and promoting accessibility and usability of public space is important to improve people’s subjective wellbeing.