Suspicious Discontinuities (2020)
Points and comments are a snapshot, not live.
Sharp policy thresholds create perverse incentives, like losing money to keep healthcare subsidies.
The U.S. ACA has a hard income cutoff at $48,560 for individuals; crossing it can increase health insurance costs by ~$7200/yr, incentivizing people to reduce income via put options. Similar discontinuities exist in TANF, Medicaid, and CHIP. The post surveys other abrupt thresholds: college admissions spiking around Pell Grant eligibility, Russian election turnout spikes at round numbers, used car prices jumping at $10k boundaries, p-value hacking at 0.05, drug charge amounts clustering just above mandatory minimum thresholds, high school exit exam scores inflated just above passing, and youth sports participation heavily skewed by birth month within a yearly cohort.
What commenters are saying
Commenters broadly agree that tax/subsidy cliffs are a real problem, with many citing similar examples from the UK (VAT thresholds stifling business growth, personal allowance taper), Slovenia (kindergarten fees jumping sharply at income lines), and Finland (promotion refusals due to tax bracket steps). One comment notes that passive income and lobbying bribes escape these steep marginal rates while labor income does not. Another observes that these cliffs hurt the living by delaying young adults from moving out or starting careers, and a shared example tells of a widow losing over €100 in benefits because a €2 pension increase pushed her over a threshold. The thread also touches on marathon finish times clustering at round numbers and age-group binning in races as similar discontinuities.