Cash transfer programs have become one of development's most studied interventions. The evidence that giving poor households money improves their lives is now overwhelming. But a persistent debate divides practitioners: should we attach conditions to these transfers?

The intuition behind conditions seems reasonable. Require parents to send children to school, attend health clinics, or participate in nutrition programs—and you might get benefits beyond what cash alone achieves. Governments from Mexico to Indonesia have built massive conditional cash transfer programs on this premise.

Yet the evidence tells a more complicated story. Sometimes conditions boost outcomes significantly. Sometimes they make no measurable difference. And sometimes they actively harm the poorest recipients the programs aim to serve. Understanding when strings help and when they hurt requires examining what conditions actually do—and what they cost.

Proponents of conditional transfers offer several theories for why requirements might improve outcomes. The most straightforward is that conditions change behavior directly—families who wouldn't otherwise prioritize education or healthcare are nudged toward different choices by the threat of losing benefits.

A second theory involves information. Conditions signal what behaviors the government considers valuable. In contexts where parents underestimate returns to education or preventive healthcare, this signaling function might shift beliefs and priorities. The condition becomes a form of social guidance.

A third rationale is political. Conditions make transfers more palatable to taxpayers and policymakers skeptical of unconditional giving. Even if conditions don't change recipient behavior, they may enable larger programs or longer political support. The strings aren't for recipients—they're for voters.

Finally, conditions can address household bargaining failures. In families where decision-makers don't prioritize children's welfare, external requirements may shift power toward those advocating for education and health investments. The condition becomes leverage within family negotiations.

Takeaway

Conditions can work through multiple channels—direct behavioral change, information signaling, political sustainability, or shifting household power dynamics. Understanding which mechanism matters most determines whether conditions are worth their costs.

The gold standard evidence comes from studies that directly compare conditional and unconditional transfers to the same population. These head-to-head comparisons reveal a nuanced picture that defies simple conclusions.

In Malawi, a landmark study found that conditions requiring school attendance significantly increased enrollment and reduced dropout rates—but only among girls already in school. For girls who had already dropped out, unconditional transfers performed equally well. The condition helped prevent departure but didn't enable return.

Studies in Morocco and Burkina Faso found negligible differences between conditional and unconditional transfers on education outcomes. Families spent the money on schooling regardless of requirements. The conditions added administrative complexity without measurable benefit.

Health outcomes show similar variation. In some contexts, requiring clinic visits for pregnant women increases prenatal care. In others, women attend clinics at similar rates with or without conditions. The difference often depends on existing barriers: where distance, cost, or knowledge limit healthcare use, conditions add little. Where cultural factors or competing priorities matter more, conditions may shift behavior.

Takeaway

Evidence doesn't support a universal answer. Conditions improve outcomes when they address genuine behavioral constraints—but when families already prioritize the targeted behaviors, conditions add cost without benefit.

The most troubling evidence concerns who conditions exclude. Verification systems require documentation, regular check-ins, and bureaucratic compliance. The families least able to navigate these requirements are often the poorest and most marginalized.

In practice, conditions exclude through multiple mechanisms. Families in remote areas struggle to access required health facilities. Parents working informal jobs can't take time off for mandatory meetings. Households lacking birth certificates or identity documents fail enrollment verification. The administrative burden falls hardest on those with fewest resources to manage it.

A study in Indonesia found that the poorest quintile of eligible households was least likely to receive conditional transfers—not because they failed to meet conditions, but because they couldn't complete enrollment and verification processes. The targeting mechanism meant to ensure resources reached the poor actually screened them out.

Beyond exclusion, conditions impose real costs on recipients. Time spent traveling to verification appointments, money spent on transportation, stress from navigating bureaucratic requirements—these burdens reduce the effective value of transfers. For a family receiving $50 monthly, losing two workdays to compliance requirements represents a significant hidden tax.

Takeaway

Every verification requirement creates a barrier. Program designers must weigh potential behavioral benefits against certain exclusion costs—and the most vulnerable families often pay the highest price for administrative rigor.

The evidence points toward a pragmatic framework rather than ideological commitment to either approach. Conditions work best when they address genuine behavioral constraints, when verification systems are simple and accessible, and when the targeted behaviors aren't already prioritized by recipients.

Unconditional transfers perform equally well in many contexts—and avoid the exclusion costs that conditions inevitably create. They're particularly appropriate where existing barriers are primarily financial rather than behavioral.

The most effective programs increasingly combine both approaches: broad unconditional support to ensure coverage, with lightweight conditions only where evidence suggests genuine behavioral gaps. The goal isn't ideological purity but maximum impact per dollar spent—and sometimes the best string is no string at all.