Part of River and Mercantile Group PLC
Where to From Here? Long-Term Interest Rates
Under a New Political Regime
by Ryan McGlothlin & James Walton
One of the biggest stories since the November 8 election is the increase in US interest rates for bonds with maturities of 10 years or more. Investors, especially those with long-term liabilities such as pension plan sponsors, are rightly very interested in where these interest rates are likely to head from here. The conventional wisdom says that long-term interest rates should continue to move higher, supported by a Fed that is signaling higher short-term rates along with potential fiscal policies, such as tax cuts and infrastructure spending, that many believe will be inflationary. The conventional wisdom is that the catalyst of these new government policies will cause long-term interest rates to soon revert back to levels commonly seen before the Great Financial Crisis (GFC) from their current “abnormally low” levels. Significantly higher rates would enable pension plan sponsors in particular to breathe easier as the value of their liabilities will fall, and would also help insurers and endowments that have struggled to meet income requirements in a low interest rate world.
We believe that the conventional wisdom is correct, but only in part: long-term rates can rise from current levels due to fiscal policy shifts that raise inflation expectations. But long-term rates are also unlikely to reach pre-GFC levels anytime soon. Long-term interest rates behave very differently than the short-term interest rates that are controlled by the Federal Reserve. Just because the Fed raises short-term rates, even several times, it does not mean that long-term rates will rise much, or at all, beyond what is already priced into the forward curve.
Long-term interest rates are, ultimately, highly correlated with the percentage change in Nominal Gross Domestic Product (Nominal GDP). Nominal GDP growth is Real GDP growth (which is the number most often reported as “GDP growth”) plus inflation. Unless Real GDP growth, or inflation, rises significantly, then Nominal GDP, and thus long-term interest rates, is also unlikely to rise significantly. We do not believe that the fiscal stimulus proposed by the new Administration is likely to materially increase Real GDP growth. Fiscal stimulus may increase inflation expectations given that US unemployment levels and industrial output rival levels achieved prior to the GFC (see Charts 1 and 2 below). However, there is still seemingly slack in the economy, with one data point being capacity utilization in manufacturing (Chart 3), which means that inflation is unlikely to rise significantly.
This article discusses the fundamentals that underpin long-term interest rates and shows why rates may very well rise some over the next few years, but also why expectations of larger rises should be tempered. Long-term interest rates are unlikely to return to the levels that many investors consider “normal”.
Long-Term Interest Rates In Historical Perspective
Many investors are focused on interest rates reverting back to “normal” from their current “abnormally low” levels. The question is, what is “normal”? The short answer is that there is no “normal” long-term interest rate, except to the extent that long-term interest rates are ultimately related to long-term economic growth rates and inflation. Chart 4 shows the history of US 10-Year interest rates since the early 1960s. A high growth environment is accompanied by abundant opportunities to invest productively in the real economy and investors therefore demand a higher interest rate from markets. Similarly borrowers are willing to pay higher interest to borrow as they can invest in higher yielding projects. The US has also had periods of high inflation, especially from the late 1960s through the 1980s. Fixed income investors demand higher interest rates to protect their earnings against inflation, pushing up bond yields. See Chart 5 for a history of core US inflation.
The relationship is very strong, as one would expect, between longer-term interest rates and inflation. Higher inflation leads to higher interest rates, but with a lag. It took a period of sustained high inflation to push long-term rates higher, and vice versa.
We have established that there is a close relationship between long-term interest rates and inflation. However, until the US Treasury began issuing Treasury Inflation-Protected Securities (TIPS) in the late 1990s, there was no reliable indicator for what investors thought that future inflation would be in the future, as opposed to what inflation has been in the recent past. This forward-looking expected inflation is known as “breakeven” inflation. Chart 6 shows a history of US 10-year breakeven inflation rates.
Notice how closely inflation expectations hover around the 2-2.5% level through most of the period. This is no coincidence, as it is the Fed’s stated goal to maintain inflation in this region. More recently, and well before the November 8, 2016 elections, inflation expectations started to increase due to continued strong economic news, and a recovery in energy prices. Inflation expectations moved up after the election by about 0.30% through March 2017, as investors anticipate that we may see increased fiscal stimulus on top of an economy that is already relatively strong. Subsequently, inflation expectations have receded as investors have grown more cautious on the extent to which US fiscal policy may actually change. This is shown more clearly in the shorter history of breakeven inflation shown in Chart 7.
The current level of long-term expected inflation, about 1.85%, is just below the Federal Reserve target for inflation. Investors clearly do not believe, at this point, that inflation is going to be the driver of significantly higher interest rates over the medium-term.
Nominal GDP and Long-Term Interest Rates
Going back to our earlier point that long-term interest rates are highly correlated with Nominal GDP (Real GDP + Inflation) Growth rates, we consider the Chart 8 below…
Chart 8: Nominal GDP and 10-Year Treasury Yields
If investors are not pricing in much increase in inflation, then higher rates will need to come from higher Real GDP, assuming that the long-term relationship between Nominal GDP and long-term rates holds. We could see higher-than-expected inflation but this would take the economy overheating and the Fed falling behind in raising short-term interest rates to slow things down. Right now, the market does not see much risk of that happening.
10-Year Treasury Yields have also been about 1% below Nominal GDP rates since about 2010, and so there is room for them to revert towards Nominal GDP levels. This kind of “mean reversion” to a 3-3.5% 10 Year interest rate from the current 2.5% +/- 0.20% level is reasonably likely to occur at some point, but not necessarily over a horizon desired by many investors.
Long term structural influences on growth and interest rates
Real GDP growth rates are historically quite volatile, as is shown in Chart 9.
However, Real GDP growth rates decade by decade have also trended down in the US and, as seen below, in other developed countries, as shown in Chart 10.
According to a 2016 Federal Reserve paper, “The Effects of Demographic Change on GDP Growth in OECD Countries”, the biggest reason for this slowdown is demographics. As people within countries age, economic growth slows. The biggest driver of this is the increase in people aged 65+ versus those aged 15-39 and 40-64. As the share of 65+ year olds has increased, growth has slowed everywhere. According to the paper, a 1% shift in the share of the 40-64 year-old age group to the 65+ age group slows GDP growth by about 0.5% per year. There are various reasons provided for this slowdown, but the most compelling is simply that people over age 65 tend to consume less than younger people, and this reduced consumption reduces overall GDP growth.
Using this relationship, the Federal Reserve has done calculations on the drag that demographics is likely to have on future GDP growth, as demographic shifts are highly predictable. Charts 11 and 12 show this impact.
Charts 11 and 12 – Population Share of American Age Cohorts and Expected Impact on GDP Growth
Based on the data above, the US should now be through the worst of the demographic headwind, with Real GDP growth reduced by over 1% per year from 2010 through 2015. Real US GDP growth has averaged 1.6% since 2010. If, all else equal, Real GDP growth would have been 1-1.25% higher during this period without the demographic effect, then we would have seen Real GDP growth of 2.5%+. Looking ahead, we can’t know what Real GDP growth will be, but we do know that the impact of demographics should drop to a ~0.75%/year headwind from a >1% headwind. This extra ~0.25-0.5% of Real GDP growth could be supportive of modestly higher interest rates. However, demographics are still expected to have a suppressive impact on Real and Nominal GDP growth, and therefore, long-term interest rates, over the next couple of decades
Long-term interest rates are low in large part because inflation and Real GDP growth (together, Nominal GDP growth) are low. There is some room for interest rates to more closely align with Nominal GDP rates as they are currently about 1% below Nominal GDP. And, there is some support for higher rates from modestly higher inflation expectations due to the potential for stimulative fiscal policy as well as a reduction in the demographic headwind that has contributed to reducing Real GDP growth. Ongoing high demand for long-term income producing assets from pension plans, insurance companies and savers more generally could serve as a counterbalance to this as the population ages and matching assets with liabilities becomes more desirable. We can certainly see long-term interest rates moving modestly higher from current levels (10Y rates from approximately 2.3% to 3 to 3.5%), especially if we see more fiscally stimulative policies. However, it remains to be seen whether these polices will actually increase the pace of Real GDP growth rather than simply raising future inflation expectations. Either can lift long-term rates, but higher inflation is likely to have a negative impact on other asset values while higher Real GDP growth is more generally positive. For most defined benefit pension plan sponsors, the kinds of modest long-term rate increases that we see as possible will be insufficient to close funding deficits without additional contributions and/or very significant increases in non-fixed income asset values.
Industry Since: 1997
Bio: Click Here
Ryan is responsible for developing client investment and risk management strategies. He is the co-head of River and Mercantile’s US business. Prior to joining River and Mercantile in 2007, Ryan spent several years working for Barclays Capital in London where he focused on financial risk management for institutional investors. He began his finance career as an investment banker in Houston advising companies on capital raising and M&A transactions. Ryan holds an M.B.A. from the University of Chicago Booth School of Business and a B.A. (Honors) from the University of Texas at Austin.
Industry Since: 2004
Bio: Click Here
James works with clients on financial risk management, develops investment strategies and new investment solutions. James joined River and Mercantile from Legal and General, a UK-based insurance and fund management group, where he performed a number of investment roles on both sides of the Atlantic. This included ALM, credit strategy, illiquid assets and risk functions relating to retirement business and, most recently, James led the development of the investment strategy backing US Pension Risk Transfer business which began in 2015. James started his career at Aon as an ALM consultant to defined benefit pension plans where he also developed Aon’s LDI, strategic asset allocation processes and analytics. James is a qualified actuary and holds a first-class Masters degree in Physics from The University of Oxford.