Tax Audits as Scarecrows: Evidence from a Large-Scale Field Experiment
Bibliographic record
Abstract
The canonical model of Allingham and Sandmo (1972) predicts that firms evade taxes by optimally trading off the costs and benefits of evasion.However, there is no direct evidence that firms react to audits in this way.We conducted a large-scale field experiment in collaboration with Uruguay's tax authority to address this question.We sent letters to 20,440 small and medium-sized firms that collectively paid more than two hundred million U.S. dollars in taxes per year.Our letters provided exogenous yet nondeceptive signals on key inputs for their evasion decisions such as audit probabilities and penalty rates.Using survey data, we measured the effect of these signals on firms' subsequent perceptions of the auditing process.Using administrative data, we measured their effect on actual taxes paid.We find that providing information on audits had a significant effect on tax compliance, but in a manner inconsistent with Allingham and Sandmo (1972).Our findings are consistent with an alternative model of risk-as-feeling, in which messages about audits generate fear and induce probability neglect.According to this model, audits may deter tax evasion in the same way scarecrows scare birds away.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.018 | 0.054 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.003 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".