Lower for longer: finding the long-term "natural" rate of interest
For years commentators have wrongly predicted a return to higher rates of interest. Now, as central banks switch back into a rate-cutting mode, there is a wider expectation that rates will remain low for the long-term
March 2019 marked the 10-year anniversary of the financial crisis stock market low. Equities have recovered strongly since then and, in the US in particular, scaled new heights. However, savers and investors who have been waiting for interest rates to return to the levels that prevailed before 2009 have been perpetually disappointed. Why?
Sluggish post-crisis growth is certainly part of the story.
Over the past decade, growth has suffered as financial institutions, corporates and consumers have dealt with the long shadow of debts amassed before the financial crisis. To help them, central banks around the world kept “accommodative” monetary policy in place for much longer than they expected, keeping a lid on global interest rates. The aim of this is to ease the pain of repayments, allowing household and corporate finances to rebuild themselves gradually while avoiding further crises.
But that is only part of the story. It has become very clear that low interest rates are not just a lingering aftermath of 2008. As the chart, right, makes clear, interest rates have been on the decline for the best part of 30 years. The financial crisis, it appears, added a surge of fresh momentum to a trend that was already well established.
The fact that interest rates have been moving lower for so long, and in so many countries, suggests the decline is rooted in persistent, structural forces. The latest thinking is that the long-term decline in interest rates is the result of a fall in what economists call the “natural rate of interest”.
This is the inflation-adjusted interest rate that is expected when the economy is at full employment. A number of recent studies now suggest this rate is approximately 0.5%. With a rate of inflation of 2%, for example – its approximate current level in the US and UK – this would point to a nominal interest rate of 2.5% as the new norm.
The recent experience of the US economy supports this line of thinking. The Federal Reserve spent three years raising interest rates, albeit slowly, against a backdrop of strong employment and solid growth. However, it is very clear that US interest rates are not going back to levels seen in previous decades. Short-term government bond yields peaked late last year at just over 3% – well below the pre-crisis average. Many now expect the next move in US interest rates to be down.
Why the “natural” rate of interest is so important
The natural rate of interest acts as an anchor for short-term interest rates. It has a big impact on monetary policy decisions as well as investment decisions by businesses and investors.
For central banks – such as the Federal Reserve, the European Central Bank or the Bank of England – one consequence of average nominal rates of 2.5% is a reduced ability to respond to recessionary shocks. If the intention is to boost economic activity by cutting the cost of borrowing within the economic system, it is much easier to do so in a meaningful way when interest rates are at 4% than it is when they are at 2.5%.
This may mean that “unconventional” monetary policy – as has been undertaken by central banks in the wake of the crisis – becomes a more normal response. “Quantitative easing”, which involves the injection of funds into the economy through the purchase of securities; and fiscal stimulus, which involves some form of tax-oriented economic incentive, are examples of responses to recessions which until now have been viewed as unconventional.
Three decades of descent
Real interest rates (%) in developed economies*
Source: Bank of England, IMF, Datastream
Ramifications for investors
There are also implications for many investors, including Cazenove Capital, who are focused on delivering longer term returns. If we expect interest rates to remain in a new lower range, investors may consider revising down anticipated long-term returns from all asset classes including government bonds and equity.
There are a number of long-term structural factors which explain the fall in the natural rate of interest. The three key ones, which we explore in more detail below, are demographics, productivity and global savings.
Demographic trends have far-reaching implications for the natural rate of interest. Over the past 30 years we have seen an increase in life expectancy and median population age. As a result, the ratio of children and retirees to those of working age – known as the dependency ratio – is increasing, particularly in developed economies (see the table). These changes mean that people are expecting to spend longer in retirement, which creates greater incentives to save rather than consume during their working lives.
The effect tends to be even more pronounced in countries where individuals fear that the public pension system will not be able to cope with the additional burden of an ageing population. In these circumstances a huge increase in demand for savings exerts significant downward pressure on the natural interest rate.
The other key demographic change has been a fall in birth rates. Over time, this leads to a smaller labour force, which has the potential to limit productivity growth, depress the potential growth rate of an economy and consequently result in a lower natural rate of interest.
Growth in productivity is another key determinant of a country’s long-run growth potential. Here too there is reason to think that long-term trends may be weighing on growth and, as a result, interest rates.
One of the main long-term drivers of productivity growth in the developed world has been higher levels of educational attainment. However, the level of education cannot increase indefinitely. In the US and other developed markets the average time spent in education is starting to plateau. Furthermore, with the increase in the cost of higher education rapidly exceeding household income growth, further gains in educational attainment are being sharply reined in.
There is also a question mark about whether we are starting to exhaust the productivity benefits of new technology. The theory of “innovation waves” suggest that an economy experiences a high level of productivity growth as new technology is integrated into production processes. Once the efficiency gains have been exploited, productivity growth resumes its previous, lower levels.
Over the long term we should expect that further significant innovation (be it through automation or artificial intelligence for example) will lift global productivity and at least partly offset the demographic and resource constraint headwinds to further growth. However, with global productivity growth slowing in recent years, it may be the case that the benefits of the current innovation wave have been integrated into the economy and over the medium term we should not expect further significant productivity gains.
There is another big accumulation of savings putting downward pressure on global interest rates: global currency reserves. Following the Asian financial crisis of 1998, many emerging market countries significantly increased their foreign exchange reserves (primarily US dollars) as a defensive measure.
Investment bank JP Morgan notes that there are now eight economies in Asia alone where foreign exchange reserves are greater than 20% of domestic GDP. To the extent that these savings pools are not matched by higher domestic investment, global savings have increased.
Officials responsible for investing these huge pools of money cannot invest freely; they are required to hold safe assets – US Treasuries in particular. This has been a key driver of the increase in foreign ownership of US government bonds. By the end of the third quarter of 2018, foreign owners accounted for 36% of the market.
These investors have been willing to pay a premium for safety and liquidity, putting further pressure on global interest rates.The factors described here are all long-term secular forces. While we may see increasing rates in the future, given the slow-moving nature of these factors we are not expecting higher natural rates of interest in the near future.
Demographics: Ageing population and a shrinking working-age group is deflationary
Old-age dependency ratio (ratio of population aged 65+ per 100 aged 20-64, %)
Source: United Nations, Maddison Project. The Economist.
Where would rates be if they reverted to their historic averages?
Data held by the Bank of England charts the rises and falls of the UK’s leading Bank Rate – or its equivalent – going back for more than 300 years. What can that tell us about interest rates in the 21st century?
Financial historians, investors and others frequently hunt through historic data for a long-term average or mean, very often in the hope that it will shed light on today’s economic or market circumstances. Are we in unknown territory? Or have we been here before?
London’s long history as a centre of trade and finance means the Bank of Engand is home to rich sources of data including, astonishingly, a history of Bank Rate movements stretching back to 1694.
Bank of England researcher John Lewis mined this data to produce average rates for the entire period since 1694, as well as for shorter periods within that span. He then worked out what would have to happen to today’s ultra-low rates if they were to revert to their historic means.
Lewis chose three periods with which to compare today’s rates. These are the whole of the 20th century; the decade leading up to the financial crisis (1997-2007) and the entire period 1694-2018. As Lewis says: “The choice of benchmark matters a lot...”
Today’s Bank Rate (0.75%) would need to more than treble if it were to revert to the mean Bank Rate for the 20th century – but it would need to rise far higher if it were to revert to levels seen in the 1997-2007 period.
Today’s mortgage rates would broadly double to return to an all-time mean; but to return to 1997-2007 levels they too would need to rise far higher – climbing by approximatey 200%. Today’s bond yields are similarly far below their mean returns for any of the historic benchmark periods. But must rates revert to their long-term means?
John Lewis notes: “Uncovering long-run trends in real rates is tricky. Views differ, and there’s no economic law that says mean reversion has to happen. “But on all these measures, rates are well below historic averages.”