There are endless exotic financial products indexed to almost anything you can think of. While it is easy to bet on politics, earthquakes or the weather, it’s virtually impossible to invest in the most basic thing of all: the economy.
That is starting to change as troubled governments inch toward bonds linked to gross domestic product. This week, Italy became the first developed country to include a GDP link in bonds—sort of. Unfortunately, it is a niche offering sold only to private investors, and the GDP link is merely a final loyalty bonus designed to encourage patriotic buyers.
The real breakthrough would come with a mainstream bond issue where either payments or the final value are tied to GDP, the ubiquitous measure used to estimate total economic activity. The attraction for Wall Street is that all those expensive economists paid to predict the economy would finally have a tradable instrument to put their predictions to work. As this year has again made obvious, the stock market isn’t the economy.
More prosaically, pension funds and others whose future costs are tied to parts of the economy not captured by stocks and bonds, such as salaries, would be better able to link their investments to their liabilities.
Governments should like it, too. It would be easier to justify spending in a recession if the national debt had just dropped. Countries whose bonds are treated as a credit risk—including Italy and the rest of the European periphery, but mainly emerging markets—would benefit by having less pressure on their debt in a crisis. This would allow them to avoid the austerity often demanded by lenders.
Italy’s issue helps to raise the profile, as it is Europe’s biggest debtor. But its GDP link is really a marketing device with the slogan: “Because Italy grows with you.” The GDP-linked bonus is paid only to those who buy and hold to maturity in 10 years, and is limited to 1% to 3%, tied to the average growth of the economy over the decade. Popular savings certificates sold by Portugal have a stronger connection to GDP, but aren’t bonds.
True GDP bonds would be quite different. At the very highest level they would be a form of insurance for governments against a drop in tax receipts. The private-sector buyers act as the insurer. To work, the buyers as a group need to be stronger than the government, and need not to be the taxpayers of the government. Your own taxpayers can’t insure your tax receipts, so the small investors who have piled into Italy’s new bonds are exactly the wrong customers.
Emerging markets that borrow in foreign currencies, and weak eurozone countries that can’t be sure of central-bank support, might want to pay up for the insurance. Big foreign institutions should be happy to provide that insurance, for a price. And speculators would get a pure instrument to bet on economic growth.
Robert Shiller of Yale University, who highlighted the benefits of GDP-linked investments back in 1993, thinks the bonds make sense even for the U.S. to protect against disaster. “The coronavirus is a reminder that catastrophes happen,” he said. “This is a time to think through GDP-linked bonds again.”
The catastrophe in question would have to be truly awful, though. U.S. Treasurys, British gilts and German bunds are typically more in demand in a recession, and the countries have, so far, been able to raise their debt levels without scaring investors.
A frequently raised obstacle to economy-linked bonds is that the GDP number itself is so uncertain. As Stephen Cecchetti of Brandeis University points out, GDP revisions can be big, and tend to be particularly big in crises—exactly when the bonds should be working their best.
In the final quarter of 2008, for example, the first estimate of 2008 fourth-quarter real GDP was a fall of 3.8% thanks to the Lehman crisis, revised to a drop of 6.3% by the “final” estimate in March 2009. Later revisions have concluded that in fact the fall was even bigger, at 8.4%.
“You want to know the truth, not some extremely noisy statistic that’s going to be revised,” said Prof. Cecchetti.
The compromise suggested in a book edited by Prof. Shiller and economists from the Bank of England and International Monetary Fund is to use the estimate for the change in GDP after six months and ignore subsequent revisions.
SHARE YOUR THOUGHTS
What do you think about GDP-linked government bonds? Join the conversation below.
The danger is that it might turn out that the government has paid out far too much on the bonds if GDP is later revised down, or bondholders might feel conned if GDP is later revised up. Inflation-linked bonds solve this problem by the expedient of not revising headline inflation figures—and I suspect bondholders could get used to the idea of ignoring GDP revisions after the cutoff date, as long as they believed the statisticians weren’t biased.
Creating any new bond involves problems of liquidity, matching demand and the inherently conservative nature of finance ministries. But many have done it for green bonds, linked to environmental objectives, which have even bigger measurement and trust issues.
It’s time to experiment with proper GDP-linked bonds both to reduce government default risk and to create a market in the most obvious thing of all to invest in: economic growth.
Write to James Mackintosh at James.Mackintosh@wsj.com
Copyright ©2020 Dow Jones & Company, Inc. All Rights Reserved. 87990cbe856818d5eddac44c7b1cdeb8