Uses and Abuses of Empirical Evidence in the Death Penalty Debate
Bibliographic record
Abstract
Does the death penalty save lives? A surge of recent interest in this question has yielded a series of papers purporting to show robust and precise estimates of a substantial deterrent effect of capital punishment. We assess the various approaches that have been used in this literature, testing the robustness of these inferences. Specifically, we start by assessing the time series evidence, comparing the history of executions and homicides in the United States and Canada, and within the United States, between executing and non-executing states. We analyze the effects of the judicial experiments provided by the Furman and Gregg decisions and assess the relationship between execution and homicide rates in state panel data since 1934. We then revisit the existing instrumental variables approaches and assess two recent state-specific execution morartoria. In each case we find that previous inferences of large deterrent effects based upon specific examples, functional forms, control variables, comparison groups, or IV strategies are extremely fragile and even small changes in the specifications yield dramatically different results. The fundamental difficulty is that the death penalty -at least as it has been implemented in the United States -is applied so rarely that the number of homicides that it can plausibly have caused or deterred cannot be reliably disentangled from the large year-to-year changes in the homicide rate caused by other factors. As such, short samples and particular specifications may yield large but spurious correlations. We conclude that existing estimates appear to reflect a small and unrepresentative sample of the estimates that arise from alternative approaches. Sampling from the broader universe of plausible approaches suggests not just "reasonable doubt" about whether there is any deterrent effect of the death penalty, but profound uncertainty -even about its sign.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.002 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".