How defaults and commitment devices change financial behavior.

Automatic enrollment in retirement savings dramatically increases participation — and Save More Tomorrow nearly eliminates contribution inadequacy.

Auto-enrollment experiments in 401(k) plans find that participation rates rise from around 40% opt-in to 85–95% with automatic enrollment. Save More Tomorrow — which automatically escalates contributions when employees receive raises — found that participants increased savings rates from 3.5% to 13.6% over 40 months, with almost no opt-outs. Together, these behavioral interventions address the two main failures of voluntary retirement saving: inertia preventing initial enrollment, and present bias preventing contribution increases.

Commitment savings products increase savings for households with self-control problems — but require voluntary uptake.

The SEED commitment savings account in the Philippines randomized offers of a savings account that restricted withdrawals until a goal was met. 28% of those offered opened accounts — primarily those who showed prior evidence of present-biased preferences. Savings balances increased by 81% for account openers. Similar commitment savings experiments in Kenya and Malawi found comparable effects. The mechanism — allowing people to protect future savings from present-self spending — consistently works for those who choose it.

Mobile money access improves financial resilience for low-income households, particularly women.

M-PESA in Kenya — evaluated through a combination of rollout variation and direct experiments — found that access to mobile money enabled households to cope better with economic shocks. Women with access to M-PESA were 22 percentage points more likely to have savings, and households with access showed smaller consumption drops after income shocks. The mechanism is liquidity: mobile money enables rapid transfers from relatives or savings, smoothing consumption over income shocks.

Financial Coaching for Low-Income Adults

United States (multiple cities) · 2015

A randomized evaluation of financial coaching programs by the CFPB found mixed effects: some programs improved financial behaviors for specific subgroups but no program produced consistent significant effects on credit scores, savings balances, or debt reduction across the full sample. Financial coaching is a popular and well-intentioned intervention; the evidence for it is weaker than for structural changes like auto-enrollment.