In a bid to understand how much headroom there is for monetary policy, central bankers are increasingly paying attention to the neutral real interest rate, described by former Fed Chair Janet Yellen as “the level that is neither expansionary nor contractionary and keeps the economy operating on an even keel.” More formally, it can be thought of as the rate which balances desired wealth holdings with desired capital holdings. This is the theoretical framework attributable to the Swedish economist Knut Wicksell in which equilibrium in both the goods and financial assets market is simultaneously derived.
The analysis published last week in the BoE’s Inflation Report explained this framework very nicely (see chart) and noted that we can think of the rate as being driven by long-term secular factors (R*) and a short-term component reflecting cyclical issues (s*). John Williams, recently elevated to the role of President of the New York Fed, noted in a speech earlier this year that in his view the real neutral rate (R*) in the US is around 0.5%. The BoE comes to a similar conclusion for the UK, pointing to R* in the range 0%-1% (with a modal estimate of 0.25%).
These estimates are around 200 bps lower than those
prevailing 20 years ago. So what has changed? One of the key secular factors is
demographics. As people live longer they have to save more for retirement with
the resultant increase in savings putting downward pressure on interest rates
(a shift in the red line to the right in the BoE’s chart). Another important
factor is the increased demand for safe assets which has driven down returns on
government bonds relative to those on riskier assets, and which also has the
effect of driving the red curve further to the right. A third factor is the
slowdown in productivity which has reduced business demand for capital, thus
putting additional downward pressure on the interest rate (the blue line shifts to
the left). Finally, a rise in the government debt-to-GDP ratio may depress the
real rate via a crowding out effect since this reduces the quantity of capital
available to finance an expansion of the business capital stock.
As to how these factors will play out in future, there is general agreement that slower population growth in the western world will not reverse the current trend ageing of the demographic profile. Consequently, retirement saving is likely to remain a key driver putting downward pressure on the equilibrium rate. It is less clear what will happen with regard to productivity. It may recover, or it may not, but we cannot say for sure that it will remain as sluggish as it has over the last decade. In any case, as labour force constraints begin to bite, it is possible that demand for capital will rise which will act to raise the neutral rate. But it is unlikely that government debt-to-GDP ratios will decline rapidly any time soon, which argues for continued downward pressure on the equilibrium rate.
However, some doubt has been cast by the BIS on the link between interest rates and the observable proxies that are conventionally used to measure the savings-investment balance. Part of their argument rests on the fact that much of the analysis is based on data only back to the 1980s and that taking the data back to the late nineteenth century suggests a weaker link between them. That said, the BoE’s analysis is based on a long-run of data extending back more than 100 years and they come to much the same conclusion as the rest of the academic literature, which weakens the BIS criticisms to some degree.
However, the BIS does raise another important question: Much of the literature assumes that monetary policy is neutral in the long run and that only real factors influence the real interest rate. But is this necessarily true? For one thing, the expected wealth demand function may be determined by the actions of central banks themselves as interest rate expectations influence portfolio choices. Another objection is that we may underestimate the key channels through which monetary policy exerts a persistent influence over real interest rates (e.g. the inflation process or the interaction between monetary policy and the financial cycle). These are serious criticisms, although the BoE’s framework introduces the short-run variable s* into the framework, and whilst we can estimate R* using conventional measures, the BoE does not try to put a numerical value on s*. However, it does suggest that in the longer-term the s* component will tend towards zero (although it may not be zero at any given time).
What are the takeaways from all this? First off, if we add a 2% inflation rate to estimates of the real neutral rate, we end up with a neutral funds rate in nominal terms of around 2.5%. With the Fed funds target corridor currently set at 1.75%-2.0%, we might only be three 25 bps hikes away from the neutral rate. Similarly, the UK neutral rate is estimated in the range 2%-3% so we do appear to have more headroom. Nonetheless both estimates suggest that interest rates will not get back to the pre-2008 rates of 5%-plus for a long time to come. Welcome to the new normal.
As to how these factors will play out in future, there is general agreement that slower population growth in the western world will not reverse the current trend ageing of the demographic profile. Consequently, retirement saving is likely to remain a key driver putting downward pressure on the equilibrium rate. It is less clear what will happen with regard to productivity. It may recover, or it may not, but we cannot say for sure that it will remain as sluggish as it has over the last decade. In any case, as labour force constraints begin to bite, it is possible that demand for capital will rise which will act to raise the neutral rate. But it is unlikely that government debt-to-GDP ratios will decline rapidly any time soon, which argues for continued downward pressure on the equilibrium rate.
However, some doubt has been cast by the BIS on the link between interest rates and the observable proxies that are conventionally used to measure the savings-investment balance. Part of their argument rests on the fact that much of the analysis is based on data only back to the 1980s and that taking the data back to the late nineteenth century suggests a weaker link between them. That said, the BoE’s analysis is based on a long-run of data extending back more than 100 years and they come to much the same conclusion as the rest of the academic literature, which weakens the BIS criticisms to some degree.
However, the BIS does raise another important question: Much of the literature assumes that monetary policy is neutral in the long run and that only real factors influence the real interest rate. But is this necessarily true? For one thing, the expected wealth demand function may be determined by the actions of central banks themselves as interest rate expectations influence portfolio choices. Another objection is that we may underestimate the key channels through which monetary policy exerts a persistent influence over real interest rates (e.g. the inflation process or the interaction between monetary policy and the financial cycle). These are serious criticisms, although the BoE’s framework introduces the short-run variable s* into the framework, and whilst we can estimate R* using conventional measures, the BoE does not try to put a numerical value on s*. However, it does suggest that in the longer-term the s* component will tend towards zero (although it may not be zero at any given time).
What are the takeaways from all this? First off, if we add a 2% inflation rate to estimates of the real neutral rate, we end up with a neutral funds rate in nominal terms of around 2.5%. With the Fed funds target corridor currently set at 1.75%-2.0%, we might only be three 25 bps hikes away from the neutral rate. Similarly, the UK neutral rate is estimated in the range 2%-3% so we do appear to have more headroom. Nonetheless both estimates suggest that interest rates will not get back to the pre-2008 rates of 5%-plus for a long time to come. Welcome to the new normal.