How much is too much?

Add this article to your reading list by clicking this button

THE popularity of austerity policies has waned over the past several years thanks to evidence that it may have been counterproductive. But many are still worried by the fact that, relative to national income, government debt is now larger in many countries than at any point since WWII. Moreover, for most nations, government debt is projected to grow relative to income for years to come. This is why policymakers across the rich world have been scrambling to slow the growth of public spending while simultaneously increasing tax revenues. (America’s budget fights should be understood in this context.) Does their urgency make sense? The sovereign bond markets in America, Japan, Britain, and the euro area’s “core” do not seem to think so. These governments can borrow cheaply for decades at a time. While it is certainly possible that the markets are wrong, policymakers should probably pay more attention to investors and less to the fear-mongers, especially since economists do not know how much government debt is too much. In fact, there is good reason to think that many countries with their own currencies could become far more indebted without risking trouble. One reason is that many private investors do not own enough sovereign bonds.

It is important to remember that there is an absence of evidence that governments with their own currencies are too indebted. Those who argue otherwise point to the work of Carmen Reinhart and Kenneth Rogoff, the celebrated authors of This Time is Different . Their paper “Growth in a Time of Debt” claimed that sovereign debt creates a burden on the rest of the economy. (They summarise their points here .) But, as Robert Shiller and Paul Krugman have pointed out, Ms Reinhart and Mr Rogoff never explain how public indebtedness restrains growth. There may be other forces at work, especially since sovereign debt ratios are usually at their highest after wars and financial crises. In countries with their own currencies, private interest rates are now so low that many investors have been grasping for yield wherever they can find it, such as in the revived CLO market. When he evaluated the evidence, my colleague concluded that “debt matters, but the precise way that it matters isn’t as clear-cut as Reinhart-Rogoff seem to indicate”.

Why would private investors want to buy more sovereign debt? A previous post on the shortage of safe financial assets mentioned how pension plans in many countries need to buy more government bonds to avoid mismatches between their assets and liabilities:

Defined-benefit pension plans usually have liabilities that can be perfectly hedged through purchases of government bonds denominated in local currency. (The exception is if they have to cover healthcare costs.) This makes these funds very easy to manage, in theory. All that would be needed is to have an actuary determine the size of the liability and its duration, which would then be fully funded by buying the appropriate quantity of government bonds. But pension funds almost never do this. Instead, they “underfund” their schemes. To compensate, they buy risky assets, like stocks, that are poor hedges for their liabilities. This has started to change in the wake of the crisis, but most of the world’s pension money is still under-invested in sovereign debt.

According to Towers Watson, a consultancy, pension fund assets are worth more than $27 trillion (not all of these are owned by defined-benefit plans). This is a large pool of savings but it is not large enough to offset these pensions’ liabilities. Many plans were underfunded before the crisis, but the losses of 2008—yet to be recouped—have made things far worse. To close the gap without making risky bets, pension guarantors would need to increase their contributions by a large amount.

Moreover, most pension funds take excessive risks with their asset allocations. As mentioned above, the safest way to manage most pension plans is to invest 100% of the assets in sovereign debt denominated in local currency. The Bank of England, which may run the world’s most conservative pension fund for its employees, spends more than half of its payroll on inflation-indexed gilts. But with the exception of pension funds in Japan and the Netherlands, less than 40% of pension assets are invested in bonds. For example, American pension funds, which hold assets worth about 107% of America’s GDP, allocate just 31% of their portfolios to fixed income. Even this small share includes corporate debt and other relatively risky assets that are not ideal hedges for pension liabilities. It all suggests that many governments could afford to have higher public debt to national income ratios. In fact, governments should probably try to accommodate pension funds as they adjust their portfolios by running larger deficits and increasing their bond issuance.

Individuals may also be under-invested in government fixed income. Most people invest around 60% of their retirement savings in equities, with the rest invested in bonds. But shares are about three times as volatile as government bonds, so the performance of these portfolios is almost entirely determined by the performance of the stock market. This lack of diversification would be desirable if equities had a much higher risk/return ratio than bonds. However, the Sharpe ratios for bonds and stocks have been basically identical over long periods of time, which suggests that most peoples’ portfolios are needlessly biased in favour of shares. To remove this bias, an investor could allocate a much larger slice of her portfolio to government bonds. Sophisticated investors can already purchase “risk parity ” portfolios on the theory that they can significantly improve their risk/return trade-off by applying these ideas. Again, governments could help accommodate these portfolio shifts by increasing their debt issuance.

How much public debt is too much? There is no straightforward answer. However, it seems that many countries may be able to afford to have significantly higher ratios of government debt to national income. Of course, the way in which we calculate these debt/income ratios may also be misleading, which is why a subsequent post will examine some ways to modernise public accounting practices.

Previous

Monetary policy: What’s moving Japanese markets?

Next

Entitlement reform: Cuts on the sly