On Wednesday, June 2, Olivier Blanchard joined Markus’ Academy for a lecture on rethinking fiscal and monetary policy, post-COVID. Blanchard is a Senior Fellow at the Peterson Institute for International Economics and a Professor of Economics Emeritus at the Massachusetts Institute of Technology.
There are two relevant dimensions of low rates. The first, often called “secular stagnation” occurs when the safe neutral rate (r*) is less than the growth rate (g). The second, called the “zero lower bound”, comes from the fact that nominal interest rates cannot be negative (or only to a very limited extent). It implies that monetary policy cannot achieve a real safe rate lower than minus the inflation rate, and may therefore be limited in its ability to stabilize output .
The long view suggests that low real and nominal interest rates are here to stay. There may be bumps and limited r*>g episodes (for example in the US as a result of the Biden stimulus program), but r*<g is likely to dominate. Our understanding of the determinants of low rates is however limited enough that we should not, in thinking about fiscal policy, fully exclude a reversal. In the same way, unless there is a dramatic increase in inflation targets and associated nominal interest rates, the zero lower bound is likely to limit the ability of monetary policy to stabilize output.
A world in which r*<g is a world in which one must think differently about debt sustainability. It implies in particular that a country can run a primary deficit and still stabilize its debt ratio. Put more strongly, it implies that a country can increase its debt ratio, and never increase taxes. As debt increases at rate r and the economy increases at rate g, the debt ratio will slowly decrease over time, without the need to increase taxes.
One must however take into account the uncertainty associated with r*-g, and with the other determinants of the evolution of debt. If one defines debt sustainability as a high probability that, n years out, the primary balance will be sufficient to cover interest payments and stabilize the debt, then the assessment of sustainability should be based on a stochastic debt sustainability analysis, taking into account the uncertainty about r, g, the evolution of the primary balance, and the implicit liabilities coming from social insurance systems.
This makes it clear that there is no magic number for a dangerous debt threshold, be it 60% or 90%. For example, Japan has net debt that is 177% of GDP and gross debt that is 260% of GDP, and investors still find it sustainable.
Turning to the effects of debt on welfare, the issue is whether the fact that r*<g is an indication that the economy is dynamically inefficient, that the economy has in effect too much capital. If this is the case, then debt is actually good for welfare: While it crowds out capital, this decrease is good, and both the current and the future generations are better off.
Whether this is the case depends in particular on whether the safe rate is the right measure of the risk adjusted marginal product of capital. This may not be the case if there are distortions such as incomplete risk sharing, or financial repression forcing some financial intermediaries to hold more sovereign bonds than they would like.
Empirically, it is reasonable to conclude that the welfare cost of debt is positive, but low. Thus, debt can be used for the right purposes. One such purpose for example is the spending associated with the covid crisis: The benefit of deficits, in both protecting people and maintaining demand, largely exceeded the cost of higher debt. Another is the role for fiscal policy to maintain output at potential if monetary policy is constrained by the zero lower bound. In this case, there is a case for sustaining demand by running deficits and increasing debt.
If however secular stagnation continues and the zero lower bound remains binding, it will be essential to find ways of increasing demand without running large fiscal deficits. Increasing social insurance, for example by offering Medicare for all, should not only be good on its own but decrease precautionary saving and increase r*.