Noisy ← this week

The world is being quietly rearranged by people who write very long documents.

March 1, 2026
NBER
The title they went with
Cash Transfers and Productive Inclusion: Evidence from Bolsa Familia Noisy translates that to

Giving poor people money helps get them jobs

Turns out the thing stopping people from working was being too sick and too hungry to work.
What happened
A major increase in Brazil's cash-transfer program (Bolsa Familia) raised employment by 5 percent while hospitalizations fell 8 percent and mortality dropped 14 percent, saving roughly 1,000 lives. The finding breaks the assumption that giving money to the poorest people makes them work less — instead, it turns out that when subsistence constraints loosen, people work more and stay healthier.
5% increase in employment
8% decrease in hospitalization
14% decrease in mortality
1,000 lives saved
Why this matters
For decades, antipoverty policy has been built on a trade-off: cash transfers help the poorest survive, but they reduce work incentives. This paper shows the trade-off is false — at least when people are poor enough that basic health and nutrition are binding constraints. The mechanism is simple: healthier, better-fed people are more productive and more able to work. This means the design question for cash-transfer programs in low-income countries shifts from 'how much can we give without killing work incentives' to 'how much do we need to give to unlock productivity gains.' The implication is blunt: antipoverty spending that improves health first may pay for itself through higher employment and tax revenue.
Who wins, who loses
who wins Poor Brazilians enrolled in the transfer program, who got credited retroactively with having rational labor supply responses all along.
who loses Policymakers clinging to their classism
also Anyone designing or defunding a cash transfer program, which is most of the world's development agencies right now.
Jargon decoder
productive inclusion the mechanism by which cash transfers relax immediate survival pressures, allowing people to stay healthier and work more productively
binding subsistence constraint immediate survival needs that consume all time and energy, preventing work productivity
The signal
Why this hasn't landed yet
The finding is a quiet positive result from an academic paper about a foreign program, and it requires updating a prior that has been politically useful for cutting domestic transfer programs, which makes it easier to file than to publish.
What happens next
Development agencies currently using labor-supply disincentive models to means-test or cap transfer programs will face pressure to rerun their welfare calculations, probably within the next grant cycle.
The catch
Governments looking to cut transfer program budgets will note that the employment effect was 5 percent, declare success, and propose replacing the cash with job training.
The longer story
The longer arc
The debate over whether cash transfers reduce work goes back at least to the U.S. negative income tax experiments of the 1970s, which found modest reductions in work hours and were used to slow guaranteed income proposals for the next four decades. This study, using the world's largest such program, finds the opposite direction of effect in a low-income setting, which is where most of the programs actually run.
Part of a pattern
Part of a decade-long accumulation of evidence from GiveDirectly, Bolsa Familia, and various African transfer programs that cash works better than in-kind aid, but this is the first large-scale result that shows employment going up rather than staying flat, which moves the argument from 'cash doesn't hurt' to 'cash helps on the metric critics said mattered most.'
If you insist
Read the original →
The Sendoff
The paper set out to study cash transfers and ended up saving a thousand lives as a side effect. The authors noted this in the limitations section.