The direction of long term rates in an era of high debt

Real interest rates have been on a secular decline since the mid-1980s but have corrected sharply higher since the Covid pandemic, higher inflation and the end of central banks’ Quantitative Easing (QE). Now, public debt is much higher and there are new demands for funding (e.g., for defence and net zero ambitions), which should argue for structurally higher long-term interest rates. Yet numerous empirical studies find that the secular decline in real rates is due to adverse demographics and declining productivity (at least in the advanced economies), which are expected to continue.

Abstracting from the current bout of sharply rising yields triggered by the sea change in Germany’s fiscal policy, what would constitute a prudent capital market assumption over the long term for asset allocation and the expected returns of the main asset classes?

The structural determinants of the equilibrium for real interest rates (R* in the jargon) provide a good starting point. Ageing populations imply lower growth, which in turn requires less investment, hence a lower demand for funds. Similarly, declining productivity reduces incentives to invest. Economists, in general, expect R* to come back to lower levels in the long term, primarily because demographic trends (which only change over very long periods) are expected to remain adverse. While there is some optimism that artificial intelligence (AI) could increase productivity, it would only stem the secular decline that has been underway over the last few decades. This would suggest a real rate back at around 1% in the US (from the current level of around 1.5%), with current R* estimates moved slightly upwards.  Other advanced countries should also see a similar return to their longer-term norms.

But investors will rightly question such a sanguine view, not least on the grounds that, given the current environment, how can economists be confident about getting to this long-run level? For one, there is little evidence that governments are attempting to reduce public debt – for most advanced economies, it will likely rise for a while. The recent rise in German Bund yields, which has a large real component, is a good example of more debt increasing real rates. Moreover, over the last two years, real interest rates have been rising despite well-anchored, longer-term inflation expectations. And even more importantly, we now have ample evidence that longer-end yields are strongly correlated across advanced economies, which implies that countries may suffer higher real rates through global contagion.But investors will rightly question such a sanguine view, not least on the grounds that, given the current environment, how can economists be confident about getting to this long-run level? For one, there is little evidence that governments are attempting to reduce public debt – for most advanced economies, it will likely rise for a while. The recent rise in German Bund yields, which has a large real component, is a good example of more debt increasing real rates. Moreover, over the last two years, real interest rates have been rising despite well-anchored, longer-term inflation expectations. And even more importantly, we now have ample evidence that longer-end yields are strongly correlated across advanced economies, which implies that countries may suffer higher real rates through global contagion.

High, and possibly rising, public debt will keep real rates high and delay the return to estimates of R*. But for some countries, it will raise debt sustainability concerns, including the US if fiscal policy remains on its current path. And for most countries, the market will implicitly enforce fiscal trade-offs. This, in turn, implies lower medium-term growth and that will be one way real interest rates decline. However, it is also possible that high public debt will weigh on inflation, which will result in higher central bank policy (neutral) rates. We also expect higher interest rate volatility due to more volatile inflation through our cascade model structure.

As central banks have maintained a restrictive policy stance over the last three years, much of the rise in long bond yields has been due to an increase in real yields – inflation has come down gradually, but market expectations of inflation have remained stable. This fuels doubt among investors whether real rates will return to pre-GFC levels (of around 1%). But this also begs the question of what has fundamentally changed such that the US, for example, can be resilient to and sustain real rates moving higher than 1.5% towards potential GDP.  Estimates of potential output have not budged much. Consensus estimates, similar to official estimates, point to 1.8-2% as the long-run potential. Elsewhere – in Europe and the UK, for example – potential growth is arguably slightly lower than before. This is the fundamental reason why we believe real rates will eventually revert to their long-run equilibrium.

A more worrying issue is the correlation between long bond yields across developed markets, with two very strong recent illustrations. First, the rise in US bond yields following Trump’s election, when markets revised up US inflation and growth expectations. This had a very immediate impact on German and UK bond yields, despite no change in any underlying fundamentals in Europe. A second, more recent episode is Germany’s policy announcement to increase defence and infrastructure spending. The impact on global bond yields was immediate, including contagion to the US.

Contagion implies that some risks, prime among these would be the secular rise in US public debt, are global risks. An increase in the US term premium would be transmitted across borders. And this could keep real interest rates high and volatile beyond what is warranted by domestic factors. This would amplify headwinds to growth and financing conditions.
In short, a return to equilibrium R* could take time, until public debt is on a stronger footing. Hence, the medium to long-term horizon could be characterised by higher interest rate volatility and some contagion risk.