#865 2000 · Green Bank of Caraga · Rural banking / microfinance
Green Bank sold savers an account they could not touch — the lock was the product
the problem
Clients wanted to save but kept dipping into their own accounts — willpower, not income, was the bottleneck
background
Green Bank of Caraga, a rural bank in Mindanao serving low-income clients, watched a familiar pattern: customers who genuinely wanted to save toward a goal — school fees, a roof, a small investment — kept withdrawing from their own accounts long before they got there, and ended up where they started. The bank's toolkit for this was the industry standard: better interest rates, more encouragement. Neither touched the real problem, which happened in the moment of temptation.
From 2000, the bank worked with researchers Ashraf, Karlan and Yin to turn the product itself into the experiment. They designed SEED — Save, Earn, Enjoy Deposits — then ran a randomised trial through 2003: 1,777 existing or former clients were surveyed, and 710 of them, randomly chosen, were offered the account.
what everyone would do
The standard levers for growing savings balances are price and persuasion: raise the interest rate, run financial-education drives, open more branches. Each costs the bank real money or assumes the customer's problem is ignorance. None of them stops a hand reaching into its own account the moment an expense tempts.
what they saw
The bank saw its struggling savers didn't need a better return — they needed protection from their own hands. It sold the one thing banking never sells: a restriction. Only those who knew they'd be tempted signed up.
the move
SEED was an ordinary savings account paying the same 4% annual interest as the bank's regular product — plus one contractual twist. At opening, the client chose either a goal date or a goal amount, and no withdrawal was allowed until it was reached. Deposits stayed open at any time, including automatic transfers from a regular account, and clients could buy a lockbox for which only the bank held the key. The bank was selling the restriction itself, at no discount and no premium.
why it works
A saver who knows she cannot resist temptation values the lock itself, so the account self-selects exactly the customers who need it without the bank screening anyone. Because the restriction is contractual and visible — the goal written at opening, the optional lockbox keyed only by the bank — giving in means breaking a promise made to oneself plus a trip to the teller, enough friction to outlast most impulses. And because SEED paid the same 4% as an ordinary account, the demand it revealed was for commitment alone: the balance growth came from behaviour, not subsidy. The lock also guarded against the second failure mode the researchers named — money being claimed by others in the household — which is why married women with the least bargaining power gained the most.
the payoff
Balances grew 42% beyond controls within six months and about 80% within a year — off the same 4% interest rate as an ordinary account.
where it breaks
It fails for the majority who don't recognise their own temptation — only 28% of those offered opened an account, and about 10% of the treatment group ended up saving more. It fails when life is genuinely unpredictable: an emergency before the goal date turns the lock into hardship, and SEED offered no early-exit valve. And it only works while the restriction is credible and held by someone else — a lock whose key stays with the saver is just a box.
what came after
Ashraf, Karlan and Yin's write-up — 'Tying Odysseus to the Mast', Quarterly Journal of Economics, 2006 (doi:10.1162/qjec.2006.121.2.635) — became one of the most-cited field demonstrations in behavioural economics and seeded a generation of commitment-savings products in several countries. Its honest footnote is the one the data carries: the product only works for the minority who know they need it — a quarter signed up, and about 10% of those offered ended up saving more.
references
- [1]Commitment Savings Products in the PhilippinesInnovations for Poverty Action, 2014poverty-action.org
- [2]SEED: A commitment savings product in the PhilippinesCenfri, 2020cenfri.org