Who is poor and who is not? Toward an empirical basis for identifying the poor

In "Poverty: An Ordinal Approach to Measurement" (1976), Amartya Sen defines the problem of identifying the poor as one of two that have to be addressed when we operationalize poverty.In this dissertation, I propose an innovative empirical approach to responding to this problem.My dissertation unfolds in four stages.First, I argue that an empirical approach to the identification problem is both desirable and possible.Second, I propose my own approach.Drawing on work in the basic needs tradition, I argue that experiencing poverty should result in the life trajectories of the poor looking different from those of the non-poor.Specifically, they should be uniquely vulnerable to suffering a cumulative negative dynamic in functioning over time, which I elect to call "deprivation."Third, I model this dynamic and propose a method for using it to evaluate poverty indicators.Fourth, I estimate the dynamic in the United States and Canada using panel data on individual earnings and household income and dynamic panel data estimators.Then, I use these results to assess the relative poverty indicator most widely used in rich countries, known in Canada as the Low-Income Measure (LIM).I find that a household in the United States needs to command about 61% of median adjusted after-tax income and in Canada, 43%, to be able to protect themselves against deprivation.Broadly speaking, if we accept the theoretical arguments offered in this dissertation, then the LIM-which is defined at 50%-strikes a rough balance between these two empirical standards.Dans La pauvret: une approche ordinale de la mesure (1976), Amartya Sen dfinit le problme de l'identification des pauvres comme l'un des deux problmes rsoudre lorsque nous oprationnalisons la pauvret.Dans cette thse, je propose une approche empirique innovante pour rpondre ce problme.Ma thse se droule en quatre tapes.Premirement, je soutiens qu'une approche empirique du problme d'identification est la fois souhaitable et possible.Deuximement, je propose ma propre approche.Bas sur le travail de la tradition des besoins essentiels, je soutiens que faire l'exprience de la pauvret devrait donner aux trajectoires de vie Part IV.Operationalizing Deprivation and Poverty Chapter 8. Data Sources and Variables Leading longitudinal data in the US and Canada Operationalizing functioning using earnings Operationalizing resources using household income Considering alternative measures of resources Controls Variables Comment on multicollinearity Administrative data and the future Chapter 9. Dynamic Panel Data Estimation Time series and panel data methods Dynamic panel data estimators Dynamic panel data estimators in practice Ensuring robustness Additional considerations Chapter 10.Deprivation and Indicating the Poor in the US and Canada Summary of descriptive results Cumulative dynamics in earnings in the US Empirical poverty thresholds in the US Considering an AR2 deprivation process Cumulative dynamics in earnings in Canada Empirical poverty thresholds in Canada Verifying six-year results using US data Different contexts, different poverty thresholds Summary of findings Conclusion Limits of the indirect approach A new way forward Poverty in different populations and subpopulations Policy Implications Additional insights A final word on existing poverty measures Sen-Shorrocks-Thon (SST) index are less sensitive but need more expertise to understand 1 This dissertation is about how thinking about poverty longitudinally can help us advance a solution to the identification problem but it is not about "longitudinal poverty" as such (e.g.vulnerability (Dutta, Foster, and Mishra 2011) , and chronic poverty (Hoy, Thompson, and Zheng 2011) ).Questions relating to longitudinal poverty address how we aggregate poverty over time (Calvo and Dercon 2007; Christiaensen and Shorrocks 2012) and beginby assuming that we already know who should be counted as poor.2 Partly as a result of the primacy of efforts dealing with the second problem, the term "measure" has come to be strongly associated with tools that aggregate poverty within a population.In this dissertation, I will try to refer to tools that identify individuals who are poor in a population as "indicators."3 I follow Kuklys' (2005) lead and use "welfare" as a generic term to refer to the quality of a person's being.It can relate to everything from utility to "standard of living, quality of life, or subjective well-being" (1) among other things.Defining quality is no small matter and various traditions define it differently.In later chapters, when I wish to refer to a particular type of quality or aspect of welfare I will use more specific terms.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.