IMF AND WORLD BANK: World Bank turns to guarantees

Calamiris: play the comparative advantage

Today it may seem strange, but lending didn't figure as part of the plan when the founders - principally John Maynard Keynes and US assistant secretary of the Treasury, Harry Dexter White - unveiled their brainchild 55 years ago at Bretton Woods.

The idea was for the Bank to use financial guarantees - also known as credit enhancements - for mobilizing private capital in the drive to rebuild after World War II. The Bank did two of those deals in its early days but lending has dominated ever since.

Today, renewed enthusiasm for guarantees is coursing through the Bank's headquarters complex in Washington. The Multilateral Investment Guarantee Agency (Miga) - the World Bank affiliate which provides political risk insurance - just finished its best year ever. And the Bank's project finance unit, meanwhile, expects to use guarantees in connection with a growing number of deals.

The board also approved a pilot project in April to test $2 billion in "policy-based guarantees" (PBGs). The Bank wants to use PBGs to help crisis-stricken countries - like Thailand last year - re-enter international markets. The Bank plans to aim PBGs at "good performers". That is countries which it considers to have sound fundamentals and good economic policies.

"I am searching with my colleagues for ways in which we can use our credit more effectively than by just lending money," says World Bank president James Wolfensohn. "It's not easy," according to Wolfensohn, "but I think that we're making some real progress." When to use guarantees can sometimes be a tough question. And the Bank's current pricing policies are crippling many of its guarantee products compared with mainstream World Bank loans. Guarantees could emerge as important weapons in the Bank's financial arsenal, only if it can put to rest several complex issues.

The Bank has run into a problem, because it lends to so-called middle-income countries at 50 basis points over Libor. It can be tough to supply funds to a country less expensively than that, as Wolfensohn points out, even for the World Bank with its AAA credit rating. Still, there's some room for manoeuvre. "Finance theory tells us that a loan guarantee is the same thing, from a risk standpoint, as a loan," says Professor Charles Calomiris, a banking expert at Columbia University.

"The difference, in practice, is that you don't have to come up with the actual cash if you're guaranteeing a loan." Calomiris points out that it may be a bit cheaper for the Bank to guarantee a loan made by somebody else. But that's not the way they've been doing things at the Bank.

"The problem has always been that the Bank's treasurers interpret its articles to mean that the face value of the guarantees had to be provisioned against in its lending capacity," says Devesh Kapur, a Harvard professor and co-author of The World Bank: Its First Half Centuryfrom the Brookings Institution. "That meant that the borrower might as well go for a direct loan because it has to be cheaper," he says.

Kapur points out that the Bank would not reserve against the face value of the guarantee if it used an option pricing model. And it puzzles him that the Bank hasn't taken that step. "The Bank has always been extremely conservative with regard to the array of lending instruments," he observes, "but this has begun to change especially after Wolfensohn's arrival."

Nemat Shafik, the Bank's vice president for infrastructure and private sector development, responds by pointing out that option pricing models can be very subjective, particularly when it comes to political risk. "The advantage of the guarantee is leverage," she says. "We bring several hundred million dollars of private money to the table every time we provide a guarantee. Five dollars of private money typically comes in for every dollar which the Bank guarantees."

But Kapur's view enjoys support far from the serene world of academia. "Not particularly catalytic or exciting," says Lex Rieffel of the Institute of International Finance (IFF), a Washington-based group which represents large international banks, "the guarantee products of the World Bank are a disappointment."

They aren't user-friendly products either. The IIF has been arguing that there should be a separate budget allocation for the World Bank's guarantee programme. It wants the Bank's project finance department to be able to do its own originations. And that group should be able to provide guarantees independent of World Bank country operations, according to the IIF.

Last resort

Rieffel complains that World Bank guarantees are labour-intensive to an astonishing degree. "Show me the heavy lifting that's being done in the guarantee programme," he demands. "They must have 200 people across the Bank doing five or six deals a year."

However the big picture is more complicated. Peter Woicke, the executive vice president of the IFC and also head of private sector activities at the World Bank, says: "My personal view is that Miga should provide the political risk coverage, whenever it can. An IBRD guarantee, just as with a loan, should be a last resort when the private sector can't do it alone."

That's the reason, as Woicke sees it, that the World Bank guarantee programme hasn't been as successful as some people in the market would like. Woicke considers World Bank guarantees to be attractive when leverage is needed on the country to encourage reforms capable of promoting private sector development.

The Bank's project finance group has, so far, used guarantees in about a dozen deals. The most celebrated among them provided backup for Egat (Electricity Generating Authority of Thailand) bonds, which went to market last autumn when the Asian crisis was at its worst.

Egat's success convinced some folks at the Bank that they should be doing more of that sort of thing. "Guarantees can be extremely useful in transitional periods," says Nina Shapiro who runs the Bank's project finance department.