The financial crisis showed that there are serious problems with the architecture of all of the main investment types. A new instrument — bonds with payments that vary with GDP — could prove a happier medium.
Consider the weaknesses in the current set-up. The real return of ordinary fixed-interest securities can be savaged by inflation. Besides, these bonds lure borrowers to take on too much leverage in good times, while making default more likely in recessions. Inflation-linked debt, as the name implies, protects investors from increases in the cost of living. But they offer almost no exposure to economic expansion. Equities provide growth, but are more risky.
Nominal GDP forms a far superior base for investment. It captures both inflation and real growth, so brings the advantages of inflation-linked bonds and equities. As a national measure, it is less variable and more predictable than the corporate profits which underlie share prices, but grows roughly as fast as them over time. And because nominal GDP is exactly as cyclical as the economy, it is less likely than any fixed-payment arrangement to make borrowing either too tempting or too burdensome.
Robert Shiller, the economist who has promoted the concept, suggests an instrument he calls the “trill” (crediting the name to co-author Mark Kamstra). A trill would pay an annual coupon equivalent to one-trillionth of the previous year’s GDP (about $14 in 2012, for the United States).
Corporates as well as sovereigns could issue GDP-linked debt. As with any market instrument, investors would decide how much they were willing to pay for the anticipated stream of payments. As with stocks, both the payment and the price could and would change. As with bonds, the failure to pay the dividend would have all the dire consequences of default.
Some “trills” have been issued, but only by countries that have already defaulted. Indeed, there has been some talk of “GDP-kickers” on restructured Greek debt. To succeed, bond investors would have to be willing to give up some nominal certainty for the sake of a real guarantee. Traders and asset managers would have to develop new pricing models too. But GDP-linked bonds could serve issuers and investors well, while showing that the past years of financial crisis can bear good fruit.
