Behavior analysts in the war on poverty: A review of the use of financial incentives to promote education and employment

Holtyn and colleagues scanned every large government program that pays people to attend school or get a job. They looked at decades of reports on welfare-to-work checks, college stipends, and tax credits.

The team asked one question: did these programs use basic reinforcement rules like clear targets, immediate pay, and repeated practice? They wrote a plain-language story of what they found.

Most programs produced tiny or zero gains in school or work. The reason: the checks came months late, goals were fuzzy, and nobody tracked small daily steps.

In short, the programs broke every rule we teach clients about timing, size, and contingency.

Johnson et al. (2021) saw the same flaw in non-contingent reinforcement studies. Lab proofs look great, but real-world plans fall apart when staff skip the moment-to-moment delivery.

Machado et al. (2022) echo the warning in police reform. Body-cam checks and bias classes also lack tight contingencies, so results stay weak.

Bachman et al. (1988) blamed group-design drift for stalled obesity work. Holtyn adds poverty programs to the list: when you drop single-subject controls, you lose the power of immediate, visible change.

You already shape client behavior with quick tokens and clear criteria. Use the same rules when you consult shelters, schools, or job-training sites. Ask: when is the pay delivered, for what exact response, and who tracks it? Push for weekly or daily micro-bonuses tied to attendance, homework, or clock-in times. One BCBA shifted a youth program from monthly stipends to same-day $5 cards for each hour worked. Attendance doubled in two weeks. Small, fast, contingent—that’s the fix.