Looking ahead as central banks look up
The relationship between economic growth, inflation expectations and government bond yields is likely to be a dominant investment theme throughout 2018. Already this year financial markets have been spooked by the prospect of higher inflation and higher bond yields. The important question now is how far and how fast will yields rise from here?
In trying to find an answer, the first step is to establish the ‘neutral’ or ‘equilibrium’ real interest rate (which adjusts for inflation). The equilibrium real interest rate is the most important factor that determines bond yields. Changes in inflation and cyclical deviations in growth expectations as well as other factors, such as the impact of quantitative easing (QE), cause oscillations around the neutral real rate but it’s the neutral real rate that establishes an anchor around which bond yields are unlikely to stray too far.
One approach is to assume the neutral rate is related to potential GDP growth, which is the rate an economy can absorb without pushing inflation higher. If we estimate potential GDP growth at 2% then some statistical analysis tells us that this implies a neutral real rate of 0%. Adding 2% for inflation expectations (which is the average over the past 20 years) and an average spread of 1.25% for the 10-year Treasury yield over nominal interest rates gives an equilibrium 10-year yield of around 3.25%. The Laubach and Williams model used by Federal Reserve (Fed) economists, which applies more sophisticated techniques to the same simple concept, also estimates the neutral real rate at 0%.
In theory, real rates are dictated by desired net savings — if there is a greater desire to save than there is to invest, interest rates will fall to bring the two back into equilibrium (i.e. making saving less attractive relative to investing), and vice versa. Desired net savings are influenced by more than GDP growth: for example, a 2015 Bank of England paper estimated that as much as two-thirds of the fall in real interest rates over the past three decades has been caused by other structural factors.
One major factor is demographics. Typically, people aged 30 to 65 are savers, while everyone else is a spender. In the past, the ratio of the two groups has had a close relationship with interest rates and bond yields (e.g. a higher ratio of savers to spenders puts downward pressure on interest rates and bond yields). Projecting the relative size of these two age groups over the next decade suggests an ‘equilibrium’ bond yield of between 2.75% and 3%.
Figure 1: Implied real neutral interest rates
Market expectations, which we can derive from interest rate and inflation swaps, suggest a neutral real rate of 0%, and an equilibrium bond yield of around 3.25%.
Source: Datastream and Rathbones.
Lastly, market expectations, which we can derive from interest rate and inflation swaps (a type of derivative contract), also suggest a neutral real rate of 0% (figure 1), and so an equilibrium bond yield of around 3.25%.
The next step is to ask whether bond yields are likely to stray beyond the 2.75% to 3.25% equilibrium rate in the short term. With the 10-year Treasury already around 2.9%, this is a clear risk.
There are upside risks to inflation and inflation expectations from wages, fiscal stimulus and changes in the dollar. However, we believe these factors won’t put an extraordinary amount of pressure on inflation this year.
Leading indicators of US economic activity, such as new orders in the non-manufacturing sector, are reaching new post-financial crisis highs. So the scope for growth expectations to put upward pressure on bond yields is limited, but not out of the question. The huge US Budget splurge could start to feed through. Tax cuts will put more money in the pockets of workers and investors which will encourage them to spend more. However, the Budget will do very little for the economy’s potential supply and, therefore, add marginally to upward inflation pressures too.
The long and short
Another issue to consider is that the US Treasury will have to issue more bonds to pay for the boost in government spending. As a result, the bond term premium (the bonus investors traditionally receive for the added risk of holding bonds for a longer time) could rise due to changes in supply and demand.
Term premia are influenced by other economies’ central banks, and could also be pushed higher if the European Central Bank (ECB) and Bank of Japan (BoJ) end or taper their QE programmes sooner than expected. But given the still weak structural backdrop in Europe and Japan, term premia isn’t likely to increase dramatically via this channel in the near future.
The BoJ’s recent announcements suggest policymakers do not expect inflation to reach 2%. Meanwhile, the ECB’s latest macroeconomic projections suggest eurozone inflation will remain below its 2% target for its three-year forecast period. Given changes in the exchange rate since these were put together, the March release may show even weaker inflation projections. Against such a backdrop, it is difficult to see policymakers withdrawing their stimulus measures more aggressively.
Perhaps the biggest risk comes from the Fed itself. The bank’s closely watched ‘dot plot’ — which traces all voting and non-voting members’ estimates of interest rates for the next few years — suggests that nominal interest rates will rise to around 3% by the end of 2020 and beyond. Adding the average spread of the 10-year bond, it implies a Treasury yield above 4%.
This seems at odds with our calculation of the neutral rate and the model produced by the Fed’s own economists. Notably, it increases the risk of a monetary policy mistake somewhere down the line (such as the Fed increasing rates too far), which may start to put upward pressure on bond yields today.
Figure 2: Leverage in the S&P 500 Index
US firms may have become more sensitive to high interest rates following a dramatic rise in corporate borrowing over the past few years.
Source: Datastream and Rathbones.
Sensitive to higher rates
We are confident that macroeconomic conditions will support equity markets this year. Yet we are also conscious that US firms may have become more sensitive to high interest rates following a dramatic rise in corporate borrowing. The median debt of firms listed in the S&P 500 Index is almost as high as it has ever been over the past 30 years (figure 2). For more than 5% of S&P 500 firms, earnings before interest and taxes have been less than their interest payments for more than two consecutive years.
While 5% may not seem a lot, such a high proportion is unusual during a period of economic expansion. As necessary as it was, one of the unfortunate side effects of the period of extraordinary QE that followed the global financial crisis is the creation of an increasing number of such ‘zombie’ companies reliant on ultra-low interest rates. This may leave less room than usual for the Fed to raise interest rates.
Rising, but slowly
Our conclusion is that, although Treasury yields are likely to continue to climb, we believe they won’t break through 3.5% with a speed that could destabilise financial markets. For now, we are not concerned by rising inflation expectations, especially considering that they are rising from such a low base in the US. And importantly, this is in a context where global growth is still strong and inflation is still low. It’s only above 2% in 30% of the 35 OECD countries — 60% is more the norm.
Across the world’s seven largest advanced economies (popularly known as the G7), unemployment, personal income and industrial production are all still accelerating away from their average long-term levels, which means we are still in the expansion phase of the business cycle. But if growth indicators deteriorate substantially while inflation expectations continue to climb quickly then we will revisit our position.
That’s because higher inflation would flow through to higher US Treasury yields, and these yields are an important part of valuing equities – higher yields would erode the value for future cash flows like dividends or earnings. Our analysis suggests that some combination of better earnings growth and/or a willingness to pay more for those future earnings (say, a change in perceived risk that boosts how many times earnings investors will pay for a stock) will be required to offset a further 0.5% rise in the 10-year Treasury yield over the next 12 months. Neither of those are a certainty and that’s another reason to be cautious about the outlook for equities and other risk assets.