The bank which gave the poor the chance to look after each other

The big American lad from the Peace Corps stood up and cleared his throat. "Today," he announced to the meeting, "we are going to fund four people with loans of up to $500. Do you all understand the obligations of the group?" His words were translated into Armenian; there was a mutter of assent. Then the four borrowers lined up at the table to sign their pledges and receive their cash.

First came a woman, who was going to use the money to sew nightgowns. Then came a man with a plan to sell a sort of pate made out of sheeps' heads - much like the Scottish delicacy called "potted heid". Another woman was going to sell biscuits and cakes baked in her own oven. The fourth applicant, a man, didn't make it to the table; there were some questions about the financing of his business which he had not yet properly answered.

I was in the town of Gumri, visiting a "microcredit" project run by Oxfam and the Peace Corps. Gumri was briefly the focus of world compassion in 1988, under the name of Leninakan, when it was the city worst devastated by the great Armenian earthquake. Cash poured in. Then it was forgotten again. Today, tens of thousands of its citizens are still living in unheated containers for want of new housing. Margaret Thatcher Street, running past the splendid new British School, is a mass of potholes.

Microcredit - schemes for getting money into the pockets of those who have nothing - is now taking the world by storm. Next weekend in Washington brings the opening of the first Microcredit Summit, with Hillary Clinton and the Queen of Spain on the platform. The summit has the backing of the World Bank and a long list of official and non-official sponsors. Its aim is "to launch a global movement to reach 100 million of the world's poorest families, especially the women of those families, with organisations for self-employment by the year 2005".

So microcredit has arrived at last. It is beginning to provide answers to a conundrum that has defeated philanthropists for over a century: how do you lend money to people who have no security - no savings or property to serve as collateral against a loan? The charity hand-out has been one answer in the past. The loan shark has been another. In Latin America, the poor peasant who could not repay an advance of money, seed or tools fell into the condition of "peonage": hereditary debt bondage, which transformed the debtor and his heirs into the slaves of the lender's family for ever.

Nearly 10 years ago in Cairo, I met a quiet and modest young man from Bangladesh called Mohammed Yunus. When he explained to me what he was doing, I realised that I had stumbled on one of the great innovators of the 20th century, somebody who was going to leave the world a measurably better and happier place than he found it. There is no universal panacea for poverty, and the Yunus idea cannot be applied everywhere to everyone. But essentially he had cracked the conundrum. He had found - and applied - a method for advancing credit without collateral.

The Grameen Bank, which he invented and established, had been running for a decade when I met Mohammed Yunus. It already had 600,000 borrowers in Bangladesh. Today it has provided credit to more than 2 million, and the Grameen formula - in countless variants, adapted to local conditions - is being imitated all over the world.

This is how it works, in the original Grameen form. Poor and landless people were invited to form groups of five - if possible, five women. The members of this mutual support group were then each lent money, at 5 per cent interest. But if one member defaulted on repayment, then credit to the other four was cut off as well.

This is called "peer support" or "social collateral". Social pressure is another way to describe the mechanism. And it works; the results are a repayment rate to Grameen of between 96 and 98 per cent. Many conventional banks might envy that.

Grameen went on to develop housing loans on the same system. The Ganges delta is tormented by floods and cyclones which leave hundreds of thousands homeless; the bank offered to credit groups four cement columns, a lavatory pedestal and sheets of corrugated iron for roofing. But its main business has remained small cash loans, almost all of them to groups of women. This emphasis on women has two grounds. One is idealistic: to empower women in societies that undervalue them. The other is practical: women pay up. They stay in the village, they worry about what neighbours think, they are patient and systematic. Men are more erratic, liable to vanish. They are looking for work or escaping their families or using any of the countless other excuses for male bunk.

An iron Grameen rule is that all borrowers have to turn up to weekly meetings of the group. It was the same with the group I saw in Armenia. At Gumri, they have to bring the weekly interest with them, which could be a lot of money; the biscuit-baking woman would get a two-week repayment holiday, but then was committed to return $25 a week on a projected profit of $150 a month. In Armenia, where a refugee has to survive on a dole of $7 a month, these are alarming sums for poor people.

The original Yunus system works best in a peasant village where the borrowers use the money to produce a limited range of traditional goods. But Gumri is the wreck of what was once a modern town surrounded by factories. Its men and women are mostly unemployed, but they are literate industrial workers with a multitude of crafts. What suits Bangladesh is not automatically going to suit Armenia, as I soon found out.

"I kinda hate to say this," remarked a young woman from the Peace Corps, as we watched the ceremony, "but I guess Armenians don't trust one another a lot." She meant that too much emphasis on peer pressure could be counter- productive. Gumri people were nervous about putting their own financial future at the mercy of three or four other individuals, and the sanctions had to be scaled down to get them to join in. Blocking the whole group's credit if one borrower defaulted had proved too brutal; instead, the group would lose part of its initial deposit unless the repayment was made good.

This is why Grameen principles do not transplant easily into developed societies. I talked to Glen Saunders, managing director of Triodos Bank UK, which invests in microcredit schemes all over the world. "It's not really suitable for run-down communities in the West, " he said. "The sort of social bonds you find in Bangladesh do not exist here. Rehousing and general dislocation mean that the old traditional communities are no longer in place." Instead, there was the growing Credit Union movement, which (as in the Ladywood district of Birmingham) encourages small savings schemes and uses them to finance loans to the scheme's members. Social pressure to repay was only one motive among others.

As the Microcredit Summit approaches, Oxfam has warned against over-enthusiasm. Poverty is about a basic lack of power, which isn't to be cured by cash alone; there is a danger that the summit will divert funds into microcredit schemes and away from support for basic health and education. Both warnings are fair and well-founded. I do not think Mohammed Yunus would disagree.

I remember what he said in Cairo. "Every human being has the capacity to take care of himself, but society erects barriers to keep him where he is and make his conditions worse. We want to remove one of those barriers - credit. We have demonstrated that the poor not only pay back, but pay back better than anyone else in the world."