Even though US rates have already risen substantially above those of other developed markets, I seem to feel increasingly alone in my view that US rates have more room to fall than rise, and that now is a good time to start increasing allocation to medium-term US treasuries.
What the Fed said
The most recent sell-off in US Treasuries has been attributed to the statement that the Federal Reserve would keeping raising rates "beyond neutral". I first found this quoting Chicago Fed President Charles Evans on the first week of September, then repeated by Fed Chair Jerome Powell in early October. Fears of Fed rate hikes can push down bond prices in the short term, but hiking rates "above neutral" should choke off inflation and growth and be good for bond investors in the medium to long-term.
Bond investing can be based on clearly defined (though debatable) reference point against what the long-term "neutral rate" should be. Roughly, investors should buy bonds when yields are above this rate, and sell when yields are below this rate. In practice, many multi-asset class investors do not go all-or-nothing into bonds around one single rate, but rather increase or decrease their percentage allocation to bonds based on how far above or below this "neutral rate" bond yields are.
I should be clear that I understand the "neutral rate" as the long-term average short-term interest rate where the economic growth and inflation are "not too hot, and not too cold". Of these two factors, inflation is by far the most dominant driver of short-term interest rates, as the Fed often seeks to target a modest rate of around 2% to avoid deflation and lubricate growth, while savers are quick to demand enough interest to compensate for rising prices. The below chart shows how closely US short-term interest rates have tracked inflation since 1953, with "real" interest rates hitting their highs around 3% in the 1980s and 1990s, and lows of -2% in the 1960s and 2010s.
The rise in short-term US interest rates over the past two years from near zero to near the long-term inflation target of 2% has also coincided with the first two years since the election of US president Donald Trump, whose policies are widely expected to mean higher growth and more bond issuance to pay for it. More supply of bonds means lower bond prices, just as higher growth and inflation expectations means higher rates, which also means lower bond prices. One benchmark measure of this bear market in bonds is the over 9% decline in the iShares 7-10 Year Treasury Bond ETF (IEF) over these past two years, as seen in the chart below.
Source: Yahoo! Finance
In green on the above chart is the corresponding decline in the iShares TIPS Bond ETF (TIP), which holds Treasury Inflation Protected Securities that inflation-index the interest and principal payments, and so pay a "real" rate of interest. The fact that TIP fell 5% in the time that IEF fell 9% tells me that most of the rise in long-term yields over the past two years was due to a rise in real yields, rather than from any increase in inflation expectations. The chart below shows the rise in the 5-year TIPS yield from negative levels during most of 2011-2016 to the 1% level not seen since 2009.
To see the full curve of inflation expectations and real interest rates, I prefer to look at the Wall Street Journal pages comparing regular US Treasury quotes with quotes on TIPS. These curves imply long-term real interest rates evening out around 1% with a long-term inflation expectation of around 2%.
So unless real interest rates can rise to 2% or above and remain there, bonds will be a good investment under the above "neutral rate" test unless inflation also rises significantly above 2%. Below I explain three factors I believe will keep a lid on inflation: the short-term economic cycle, demographics, and technology.
Factor #1: The Short-term Economic Cycle
Just as "everyone" seems to think rates have no where to go but up, the collective wisdom suggests that the current economic cycle is already long in the tooth, and that we should be prepared for a recession over the next few years. I don't pretend to be able to predict the next crash or recession, but I do believe one main reason to hold liquid government bonds is as a form of recession insurance, providing similar "optionality" to holding cash in a downturn, but with higher carry and a "flight to quality" premium.
As Ray Dalio brilliantly explains the video "How the Economic Machine Works", the short-term economic cycle largely revolves around the short-term debt cycle, and so it is worth watching how stretched overall debt levels are getting as rising interest rates make those debts more expensive for borrowers. While US corporate debt levels continue to hit record highs, credit spreads are reaching 10 year lows. Numbers like these make it less attractive to take risk on lower quality corporate issues like those in the iShares iBoxx $ High Yield Corporate Bond ETF (HYG), but rather wait until spreads widen and raise the compensation for taking high yield credit risk.
In addition to the record levels of treasury and corporate debt, one should also watch out for the record levels of US student loan debt and the reported 20% delinquency rates in the $1.5 trillion pool of student loans. While I'm not 100% sure student loans will trigger the next crisis like housing loans triggered the last one, a higher burden and more late payments in the student loan space will have a very real impact on an important share of consumer spending.
The Fed will not be able to raise rates much above levels that trigger a significant rise in defaults in corporate and student debt, and I believe there is plenty of evidence that trigger rate is right around 3%.
One other commonly quoted metric forecasting recessions is the spread between 2 year and 10 year US treasury yields, often known as the "2s10s" spread. More cleanly than most economic charts, the below chart of the 2s10s spread shows how the last three recessions (grey bands) were exactly preceded by this 2s10s spread touching just below zero, meaning short-term interest rates rose just above long-term interest rates.
As of this writing, the 2-year yield of 2.9% is 30bp below the 10-year yield of 3.2%, so this trigger depends how quickly short-term bond traders believe the Fed can raise rates to 3%. If this triggers a recession, this will likely turn the Fed to start cutting, and bond returns will likely turn positive.
Factor #2: Demographics
Longer-term, a dominant factor that will likely keep interest rates low is the ageing population, both nationally and globally. The below charts shows one of the most important relationships across global economies today: that older populations mean lower interest rates and vice versa.
Intuitively, this makes sense: older citizens are expected to live off a fixed income from investing their savings, whether directly or through pension products, while younger citizens are more likely to borrow money for ventures and homes, and so the old and the young balance the stereotypical supply and demand for debt capital respectively.
US demographics (and interest rates) remain at a healthy half-way point between those of Germany and Japan and one end and those of Mexico and India on the other, but we are still getting older. Globally, I find it handy to index the global economy through what I call the five major "JUICE" markets of Japan, the US, India, China, and the Eurozone, where the "G2" of these are the two giants US and China. The second demographic chart show the (perhaps scarier) statistic that China is actually getting older faster than Americans are, which is likely to put downward pressure on inflation and rates on both sides of the Pacific.
Source: Business Insider
Factor #3: Technology
While "technology" is often joked to be what economists call leftover factors that can't easily be explained in other ways, technology must be acknowledged as a major factor differentiating the 2018 economy from the conditions in 1968, 1978, or 1988 that allowed rates to rise to far higher levels in those years than we may see again in our lifetimes.
The first impact of technology is productivity growth, which is a major reason few economists are worried about having "too many dollars chasing too few goods". Many of the goods and services that make up today's consumer's "basket" are more cheaply and efficiently delivered than ever, which is why in recent years the Fed has been more worried about the opposite problem of "too few dollars chasing too many goods". The fact that the Fed is already raising rates tells me they are on top of letting this problem flip around as it did right after WWII.
A second technology impacting rates is energy efficiency. An oil crisis was a major driver of US interest rates rising to double digits in the late 1970s and early 1980s, and since then, the US has become far more efficient in both energy use and energy production, including alternative energy technologies.
Third, advances in technology have also lowered capital intensity of many large companies. On the working capital side, inventories in "old economy" industries can now be far lower than in earlier decades due to "just in time" supply chain systems. On the invested capital side, six of the top 10 US listed companies by market cap (Apple, Amazon, Microsoft, Alphabet, Facebook and Alibaba) are primarily software companies requiring far less physical or debt capital than companies like Exxon Mobil, which used to top US stock indexes.
Above are the reasons why I believe "3% is the new 5%", meaning the long-term "neutral rate" of interest rates is now closer to 3% than the 5% number it might have been a decade or more ago. These are among the reasons why I am starting to increase our allocations to US Treasuries yielding above 3%, and will start to increase our allocation to TIP if real yields get much above 1%.
While I earlier focused on the benchmark IEF for 7-10 year US treasury exposure, the iShares 3-7 year Treasury Bond ETF (IEI) is now also approaching 3% for those seeking lower volatility at the shorter end of the curve. On the other extreme, the longer-dated iShares 20+ year Treasury Bond ETF (TLT) offers more upside and volatility for those more aggressive on the long term "Japanification" of the US.
Note that the larger and more diversified iShares Core US Aggregate Bond ETF (AGG) and Vanguard Total Bond Market ETF (BND) are mixes of treasury, corporate, and mortgage bonds across maturities, and do not provide the pure interest rate risk exposure to a point on the treasury curve as IEF, IEI, or TLT. Non-US clients can also find UCITS versions of these ETFs on the iShares UK website.
Disclosure: I am/we are long IEI, IEF, TLT, TIP, BND.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.