arXiv 12 Dec 2022 · Econometrics · publishedThe Quarterly Journal of Economics (2023) · 856 citations (OpenAlex)
arXiv:2212.06080 · PDF · DOI · OpenAlex · Extracted main text
When studying an outcome $Y$ that is weakly-positive but can equal zero (e.g. earnings), researchers frequently estimate an average treatment effect (ATE) for a "log-like" transformation that behaves like $\log(Y)$ for large $Y$ but is defined at zero (e.g. $\log(1+Y)$, $arcsinh(Y)$). We argue that ATEs for log-like transformations should not be interpreted as approximating percentage effects, since unlike a percentage, they depend on the units of the outcome. In fact, we show that if the treatment affects the extensive margin, one can obtain a treatment effect of any magnitude simply by re-scaling the units of $Y$ before taking the log-like transformation. This arbitrary unit-dependence arises because an individual-level percentage effect is not well-defined for individuals whose outcome changes from zero to non-zero when receiving treatment, and the units of the outcome implicitly determine how much weight the ATE for a log-like transformation places on the extensive margin. We further establish a trilemma: when the outcome can equal zero, there is no treatment effect parameter that is an average of individual-level treatment effects, unit-invariant, and point-identified. We discuss several alternative approaches that may be sensible in settings with an intensive and extensive margin, including (i) expressing the ATE in levels as a percentage (e.g. using Poisson regression), (ii) explicitly calibrating the value placed on the intensive and extensive margins, and (iii) estimating separate effects for the two margins (e.g. using Lee bounds). We illustrate these approaches in three empirical applications.
appendix boundary found by appendix_command · 68% of the source is main text. Read the extracted text to check this.
The works this paper leans on most, across its whole bibliography — not restricted to papers in our corpus. Ranked by composite intensity, which combines how often a work is mentioned, how many sections mention it, and how much of that falls in the main text rather than the appendix.
| Reference | Intensity | Mentions | Sections | Main text | |
|---|---|---|---|---|---|
| 1 | Carranza, Garlick, Orkin and Rankin (2022) Job Search and Hiring with Limited Information about Workseekers' Skills | 1.000 | 14 | 4 | 100% |
| 2 | Sequeira (2016) Corruption, Trade Costs, and Gains from Tariff Liberalization: Evidence from Southern Africa | 1.000 | 13 | 3 | 100% |
| 3 | Berkouwer and Dean (2022) Credit, attention, and externalities in the adoption of energy efficient technologies by low-income households | 0.928 | 10 | 4 | 80% |
| 4 | Lee (2009) Training, Wages, and Sample Selection: Estimating Sharp Bounds on Treatment Effects | 0.928 | 10 | 4 | 80% |
| 5 | Angrist (2001) Estimation of Limited Dependent Variable Models With Dummy Endogenous Regressors | 0.737 | 4 | 4 | 50% |
| 6 | Mullahy and Norton (2023) Why Transform Y? The Pitfalls of Transformed Regressions with a Mass at Zero | 0.737 | 4 | 3 | 50% |
| 7 | Thakral and Tô (2023) When Are Estimates Independent of Measurement Units? | 0.737 | 4 | 3 | 50% |
| 8 | Santos Silva and Tenreyro (2006) The Log of Gravity | 0.737 | 3 | 3 | 67% |
| 9 | Bellemare and Wichman (2020) Elasticities and the Inverse Hyperbolic Sine Transformation | 0.721 | 8 | 3 | 38% |
| 10 | Abadie (2002) Bootstrap Tests for Distributional Treatment Effects in Instrumental Variable Models | 0.644 | 3 | 2 | 67% |
Showing the top 10 of 62 scored citations.
arXiv econ.EM papers that cite this one, ranked by how heavily they lean on it.