This paper examines five decades of Earned Income Tax Credit (EITC) expansions to estimate employment elasticities and inform future policy design. I construct a comprehensive accounting of changes in tax credits, public assistance, and the return to work, and synthesize evidence across prior studies using a unified framework. The analysis demonstrates how alternative modeling choices—such as how the return to work is measured or which employment estimates are used—generate substantially different elasticity estimates. I present a distribution of estimates across specifications and assumptions, summarized in a histogram of implied elasticities. While estimates vary widely--even for a given policy change and/or employment response--the central tendency is clear: employment elasticities for unmarried mothers average around 0.3.

Existing estimates of tax elasticities rest on local, between-income contrasts, while the questions asked of them are broad. This paper proposes a source of variation available throughout the map of own and partner income. Because it is within-income—it stems from shifts to the tax liability function which, under family taxation, depend on partner income—own income can be controlled for without absorbing the identifying variation, and comprehensive heterogeneity analyses become feasible. The universe of French tax returns (2003–2023) reveals elasticities that decline with income, and, for women, participation responses concentrated in households with an unequal income split. Answers to normative and positive questions—where to tax income more/less, the e˙ects of reforms to jointness—are sensitive to this heterogeneity.

This paper studies behavioral responses to kinks in effective marginal tax-benefit schedules, focusing on income taxation, social security contributions and benefits, labor-market frictions, and informal compensation. Using matched employer-employee administrative records covering Brazil’s formal workforce from 2010 to 2018, we document sharp bunching at the pension-benefit ceiling but little systematic bunching at other statutory nonlinearities. Bunching rises with age, falls with firm size, and is stronger where informality is more prevalent. We develop an intertemporal model in which workers value future pension accrual, face adjustment frictions, and may receive informal earnings. The pension ceiling generates a convex kink in the effective tax-benefit budget set because marginal pension accrual disappears above it, despite the lower employee contribution rate. Structural estimates imply that the effective marginal tax-benefit rate declines with age and that formal earnings respond weakly to the net-of-EMTBR rate. Firm-size-related adjustment frictions explain the limited bunching observed in large firms.

Using comprehensive administrative tax returns data from 2017–2024, this study examines how Kenyan taxpayers respond to personal income tax reforms using a bunching approach around the first tax kink. I estimate excess mass of up to 2.9 and an elasticity of taxable income (ETI) of up to 0.42. Unlike typical patterns in the literature, self-employed taxpayers consistently bunch above rather than below the kink, while wage-employed taxpayers bunch more modestly and sometimes spread on both sides. Behavioural responses intensify following reforms that make the kink more salient, alongside substantial regional heterogeneity. Bunching is not persistent at the individual level, as taxpayers who bunch in one year generally do not remain at the kink in subsequent years.I develop a stylised framework attributing the pattern to perceived audit probabilities, finding that self-employed filers reporting just above the kink face significantly lower realised audit rates, while wage-employed filers show no significant difference.