Special Report
Highlights Below-Benchmark Duration: Below-benchmark duration positions will continue to pay off in 2019 as the Fed delivers more than the 32 bps of rate hikes that are priced into the curve for the next 12 months. While tighter financial conditions will probably necessitate a pause in the Fed’s gradual rate hike cycle at some point next year, this is already more than discounted in current market prices. A further deterioration in housing starts and new home sales, or a significant uptick in initial jobless claims would call our below-benchmark duration view into question. Neutral Corporate Credit: In an environment where the yield curve is quite flat but still positively sloped, excess returns to corporate bonds also tend to be quite low, but still positive on average. Investors should be looking for low, but positive, excess returns from credit on a 12-month investment horizon. However, credit spreads will probably widen further in the near-term and then tighten once the Fed signals a pause and global growth improves. Overweight Munis and Local Authorities: Tax-exempt municipal bond yields are very attractive relative to corporate bonds and both municipal and Local Authority bonds are relatively insulated from the weakness in global growth that will threaten the corporate profit outlook in the coming quarters. Both of these sectors should perform well in 2019. Overweight TIPS: Long-maturity TIPS breakeven inflation rates have shifted down in recent weeks, but will move higher in 2019, eventually stabilizing in a range between 2.3% and 2.5%. The rebound in oil prices that our commodity strategists expect will help, but TIPS outperformance will largely be driven by investor expectations slowly adapting to the new reality that inflation will remain much closer to the Fed’s target than it has in recent years. Yield Curve Inversion In Late 2019: Below-target TIPS breakeven inflation rates and an inverted yield curve cannot coexist. As such, investors should not worry about a sustained inversion of the yield curve until later in 2019. To profit from this view, investors should position for steepeners at the front-end of the curve. We recommend going long the 2-year bullet and short a duration-matched 1/5 barbell. The belly (5-7 year) part of the curve has become very expensive and should be avoided at all costs. Feature BCA published its 2019 Outlook two weeks ago.1 That report lays out the macroeconomic themes that our strategists think will drive markets next year. In this Special Report, we specify how investors should implement those views in the context of a U.S. bond portfolio. Key Views The main conclusions from the 2019 Outlook are: Overall, we expect the pace of U.S. economic growth to slow from its recent strong level, but it should hold above trend, currently estimated to be around 2%. […] that means capacity pressures will intensify, causing inflation to move higher. With the U.S. unemployment rate at a 48-year low, it will take a significant slowdown for the Fed to stop hiking rates. […] Ultimately, the Fed will deliver more hikes next year than discounted in the markets. This will push up the dollar and keep the upward trend in Treasury yields intact. We expect the 10-year Treasury yield to peak sometime in 2019 or early 2020 in the 3.5%-to-4% range, before the next recession sends yields temporarily lower. In the verbiage of monetary policymakers, the BCA view is that U.S. interest rates remain below the neutral level that is consistent with trend GDP growth and stable inflation. This means that the Fed’s rate hike cycle will continue in 2019, and that monetary policy will not turn restrictive until later in the year. It is this view of U.S. interest rates remaining below neutral until late 2019 that drives our portfolio recommendations. Key Risks Given our main premise, the biggest risk to our recommended portfolio allocation is that interest rates move above neutral sooner than we anticipate. We will be monitoring three main risks in the coming months to help us decide whether our main premise needs to be re-evaluated. Risk #1: Housing Since a large amount of leverage is employed in the acquisition of new homes, there is good reason to believe that housing is the main channel through which interest rates impact the real economy. This is validated by the empirical data which show that residential investment, housing starts and new home sales all provide a good indication of when monetary policy turns restrictive and of when Treasury yields peak for the cycle.2 With that in mind, the housing data have clearly deteriorated during the past 6-9 months. However, with the 12-month moving averages of housing starts and new home sales still trending higher, it is too soon to say that housing has peaked for the cycle (Chart 1). Our sense is that the recent deterioration is a result of the sharp move higher in mortgage rates that occurred earlier this year. Now that rates have moderated, the housing data should improve.3 Chart 1The Housing Market Predicts Recessions A decisive breakdown in the 12-month moving averages of housing starts and new home sales would cause us to question our premise that U.S. interest rates remain below neutral. Risk #2: Jobless Claims With the unemployment rate at 3.7%, the U.S. labor market is in rude health. That being said, a move higher in the unemployment rate would be a clear sign that monetary policy is restrictive and that a recession is right around the corner. In the post-war era, there has never been a case where the 3-month moving average of the unemployment rate has risen by more than one-third of a percentage point without a recession taking place. Often, a turn higher in the unemployment rate is preceded by an increase in initial jobless claims, and the 4-week moving average in claims has increased for four consecutive weeks (Chart 2). So far, that increase is no cause for concern. Historically, the 6-month change in jobless claims needs to reach +75k before a recession occurs (Chart 2, bottom panel). Nevertheless, the recent upturn in claims will bear monitoring in the months ahead. Chart 2Initial Jobless Claims Are Worth Monitoring Risk #3: Weak Foreign Growth & A Strong Dollar It is a bit misleading for us to include weak foreign growth and a strong dollar in the “key risks” section. In fact, our base case outlook involves weak foreign economic growth migrating to the U.S. via a stronger dollar and leading to a mild slowdown in U.S. economic activity during the next few quarters (Chart 3).4 This will probably even cause the Fed to pause its gradual rate hike cycle, but will not bring it to an end. This report also contains our recommendations for how to tactically position for this pause. Chart 3Weak Global Growth Will Drag The U.S. Lower The Awkward Middle Phase When constructing U.S. bond portfolios on a cyclical (6-12 month) investment horizon, we find it useful to split the economic cycle into phases based on the slope of the yield curve. Our economic view informs what phase (or phases) of the cycle will reign during the next 6-12 months, and the phase of the cycle informs our investment posture. We define three phases of the cycle as follows (Chart 4): Chart 4The Three Phases Of The Cycle Phase 1: From the end of the prior recession until the 3-year/10-year Treasury slope flattens to below 50 bps. Phase 2: When the 3-year/10-year Treasury slope is between 0 bps and +50 bps. Phase 3: From when the 3-year/10-year Treasury curve inverts until the start of the next recession.5 Table 1 shows how each U.S. fixed income asset class has performed in each phase. We use excess returns from the Bloomberg Barclays Treasury index versus cash to track the returns earned from taking portfolio duration risk. For other fixed income sectors we display excess returns versus duration-matched Treasuries. We also include the performance of the S&P 500 versus the Bloomberg Barclays Treasury index. Table 1Risk Asset Performance In Different Yield Curve Regimes As should be clear from the macro view discussed above, we believe that we will remain in Phase 2 of the cycle for the bulk of 2019. With the 3/10 Treasury slope a mere 13 bps at present, temporary curve inversions might occur earlier in the year, but they will not be sustained (see Key View #5 below). The first implication of being in Phase 2 is that corporate bond excess returns (both investment grade and high-yield) are likely to be positive on average, but will be very low. The bulk of corporate bond excess returns are earned in Phase 1 of the cycle when the yield curve is very steep. Excess returns don’t turn decisively negative until after the curve inverts and we enter Phase 3. Like corporate credit, Treasury excess returns are also lower in Phase 2 than in Phase 1. This makes Phase 2 an awkward one for portfolio positioning. The expected return from taking an extra unit of credit risk is quite low, as is the expected return from taking an extra unit of duration risk. In fact, cash tends to be one of the best performing asset classes in Phase 2. The excess returns from most other spread products present a similar pattern to those from corporate bonds. Elevated excess returns in Phase 1, much lower – though typically still positive – excess returns in Phase 2, negative excess returns in Phase 3. One exception to this pattern is tax-exempt municipal debt which, outside of the mid-1990s cycle, has performed similarly or better in Phase 2 than it has in Phase 1. Domestic Agency bonds and Supranationals also stick out as being very defensive sectors. They both almost always provide a small positive excess return versus Treasuries, but never provide a large reward. In the remainder of this report we discuss the five key implications for U.S. bond portfolio positioning that follow from remaining in Phase 2 of the cycle for most of 2019. Key View #1: Below-Benchmark Duration We think below-benchmark portfolio duration positions will continue to pay off in 2019. We have already shown that Phase 2 of the cycle tends to coincide with relatively low excess Treasury returns, but the slope of the yield curve is not the best indicator for Treasury returns versus cash. For that, we turn to our Golden Rule of Bond Investing which says that Treasuries tend to underperform (outperform) cash on a 12-month investment horizon when the Fed delivers more (fewer) rate hikes than what was discounted at the beginning of the 12-month period (Chart 5).6 Chart 5The Golden Rule's Track Record At present, the market is priced for only 32 bps of rate hikes during the next 12 months. More specifically, the market is pricing-in a rate increase this month, followed by one more in 2019 and then rate cuts in 2020 (Chart 6). Chart 6Market's Rate Expectations Are Too Low This extremely depressed market pricing makes us reluctant to increase duration, even tactically. While we do expect U.S. growth to slow during the next few quarters, probably by enough to necessitate a pause in the Fed’s tightening cycle, this pause is already more than reflected in current market prices. Key View #2: Neutral Corporate Credit Cyclical Horizon (6-12 Months) Being in Phase 2 of the cycle warrants a relatively defensive posture toward credit risk. For now, we recommend a neutral allocation to corporate bonds with an up-in-quality bias. We will further reduce exposure to underweight when we transition to Phase 3 of the cycle, likely late in 2019. We also recommend looking at the long-end of the credit curve to increase the average spread of your portfolio.7 Table 2 makes the importance of correctly identifying the phase of the cycle even more apparent. It shows the excess returns to both investment grade and high-yield corporate bonds for different investment horizons directly after the 3/10 Treasury slope flattens into a given range. For example, the median excess return to investment grade corporate bonds in the 12 months after the 3/10 slope breaks below 25 bps is -1.02%. Table 2Corporate Bond Performance Given The Slope Of The Yield Curve (1975-Present) As in Table 1, Table 2 shows that excess returns are much higher when the yield curve is steep and that they tend to turn negative after the curve inverts. But unlike the results in Table 1, the analysis in Table 2 includes recessionary periods and makes no attempt to split the cycle into different phases. It is a purely forward looking rule that calculates excess returns after different “trigger points” are reached. For example, the 12-month median excess return of -1.02% after the 3/10 slope breaks below 25 bps is biased downward because of periods when the slope broke below 25 bps and then continued to flatten until it inverted. An environment where the slope stayed range-bound between 0 bps and +25 bps for an extended period – closer to what we expect in 2019 – will deliver somewhat better excess returns. Tactical Horizon (< 6 Months) The phase of the cycle helps us specify our excess return expectations for the next 12 months, and based on our outlook, we expect excess returns will be positive, but close to zero. However, as we write this report, corporate spreads are widening at a fairly rapid clip. We expect the carnage will continue in the near-term, but are monitoring catalysts to initiate a tactical overweight recommendation on corporate credit.8 As they were in 2015, corporate spreads are widening at the moment due to the toxic combination of slowing global growth and relatively hawkish monetary policy. We expect that sometime in early 2019, Fed policy will ease at the margin and this will coincide with a near-term peak in credit spreads and a period of improved global growth. To determine when spreads peak we are monitoring several indicators of global growth and Fed policy that successfully called the last peak in early-2016. On the global growth side, the key indicators are (Chart 7A): The CRB Raw Industrials Index The BCA Market-Based China Growth Indicator9 The price of global industrial mining stocks On the monetary policy front, the key indicators are (Chart 7B): The 12-month Fed Funds Discounter The gold price The trade-weighted dollar Chart 7AKey Indicators: Global Growth Chart 7BKey Indicators: Monetary Policy All in all, our conviction that we will remain in Phase 2 of the cycle for most of 2019 suggests we should maintain a neutral allocation to corporate bonds on a 6-12 month investment horizon, looking for small positive excess returns. In the near-term, we expect spreads will continue to widen in the next few weeks, but will peak once the Fed signals a pause in its rate hike cycle and global growth indicators show some improvement. We are monitoring several catalysts that will help us decide when to initiate a tactical overweight position in corporate bonds. Key View #3: Overweight Munis And Local Authorities The analysis in Table 1 showed that tax-exempt municipal bonds often provide strong excess returns in Phase 2 of the cycle. This makes them an attractive place to position in the current environment, especially given the relative attractiveness of muni yields. Table 3 shows that the average yield on the Bloomberg Barclays Municipal Index is 2.75%. If we assume even a 30% effective tax rate, the taxable-equivalent yield becomes 3.93%, well above the average yield offered by the Aa-rated Corporate index. Table 3Municipals Are Attractive Another reason to like munis in the current cycle is that state & local government revenues are relatively insulated from weakness in the global economy. As foreign growth weakens and drives up the dollar, corporate profits will suffer much more than state & local government tax revenues. A similar case can be made for the Local Authority sub-index of the Bloomberg Barclays Aggregate. This index is comprised largely of taxable municipal debt (and some Canadian provincial debt), and while the average yield is lower than for tax-exempt munis, it is still competitive compared to corporate bonds. But most importantly, the sector is relatively insulated from weak foreign growth and a strong dollar. Municipal bonds and the Local Authority sub-index are important overweights in our recommended portfolio as we head into 2019. Key View #4: Overweight TIPS Versus Nominal Treasuries Though long-maturity TIPS breakeven inflation rates have fallen in recent weeks, we continue to recommend an overweight allocation to TIPS versus nominal Treasuries. We believe that both the 10-year and 5-year/5-year forward TIPS breakeven inflation rates will reach our target range of 2.3% to 2.5% in 2019. At present, TIPS breakevens are caught between being pulled down by weakening global growth and pushed up by mounting U.S. inflationary pressures (Chart 8). Most recently, weaker global growth has been winning and breakevens have moved lower alongside the plunge in oil prices. Chart 8TIPS Breakevens Face Opposing Forces Taking a step back, it is very unlikely that global growth and commodity prices will continue to fall at their current rates throughout 2019. At some point, a dovish turn from the Fed will lead to some depreciation of the dollar and global growth will stage a rebound. Our commodity strategists also expect a rebound in the oil price. They target an average of $82/bbl for Brent crude oil in 2019.10 In the meantime, core U.S. inflation will continue to print close to the Fed’s 2% target, and maybe even a bit higher in late 2019. At some point, inflation expectations will need to adapt to the new reality of inflation being near the Fed’s target. Historically, this suggests a range of 2.3% to 2.5% for TIPS breakeven inflation rates. Inflation expectations can be slow to adapt to a changing environment, and after several years of the Fed missing its inflation target from below, many investors remain trapped in a deflationary mindset. To get an idea of how long it takes inflation expectations to adjust to changes in the economy, we use our Adaptive Expectations Model of TIPS breakevens (Chart 9).11 The model is based on three factors: Chart 9The Adaptive Expectations Model Of The 10-Year Breakeven Rate The 12-month rate of change in headline CPI The New York Fed’s Underlying Inflation Gauge The 120-month rate of change in core CPI Of the three factors, the 120-month rate of change in core CPI carries the largest weight in the model. In other words, the catalyst for moving TIPS breakeven rates higher will simply be core U.S. inflation continuing to print near the Fed’s target for a prolonged period of time. All in all, investors should maintain overweight allocations to TIPS versus nominal Treasuries in 2019, targeting a range of 2.3% to 2.5% for both the 10-year and 5-year/5-year forward TIPS breakeven inflation rates. The current slowdown in global growth and commodity prices will not last for the entire year, and U.S. inflationary pressures will continue to mount as the U.S. economy grows at an above-trend pace with a very tight labor market. Key View #5: No Yield Curve Inversion Until Late 2019 The final key view that falls out of our main macro premise, which is that the fed funds rate will remain below neutral for the bulk of 2019, is that the yield curve will not sustainably invert until late 2019. This is also probably the most contentious of our key views, given recent market moves. The main reason why we think the slope of the yield curve will remain quite flat, but positive, for most of 2019 is that sustainable yield curve inversion cannot coexist with below-target TIPS breakeven inflation rates. An inverted yield curve is a signal that the market views monetary policy as overly restrictive. It means that investors expect U.S. growth and inflation to fall in the future, necessitating rate cuts. However, long-maturity TIPS breakeven inflation rates below the 2.3% - 2.5% range that has historically been consistent with well-anchored inflation expectations signal that the market believes that inflation will not sustainably return to the Fed’s target. In other words, for an inverted yield curve and below-target TIPS breakeven inflation rates to coexist, we would have to believe that the Fed would tighten monetary policy into restrictive territory without sufficient inflationary pressures to meet its target. It is difficult to envision the Fed committing such an egregious policy error. In the event that the yield curve does invert while TIPS breakevens are below target, it is much more likely that either the Fed will adopt a more dovish policy stance, leading to a bull-steepening of the curve; or, inflation will rise leading to higher TIPS breakevens and causing the curve to bear-steepen. In either scenario, it is hard to see how yield curve inversion will last very long without significantly higher TIPS breakevens. We will call an end to Phase 2 of the cycle only when the yield curve is inverted and long-maturity TIPS breakeven inflation rates are above 2.3%. Curve Positioning As for how to position on the yield curve in 2019, the biggest change since the end of last year is that the belly (5-7 year) of the curve is now very expensive (Chart 10). In fact, the 2/5 slope is slightly inverted as we go to press, meaning there is actually negative rolldown in the 5-year note. Chart 10Par Coupon Treasury Curve By far, the best place to position on the curve is the 2-year maturity point.12 Our model of the 1/2/5 butterfly spread (2-year bullet over duration-matched 1/5 barbell) shows that the 2-year is cheap relative to the 1/5 slope. Conversely, our model of the 2/5/10 butterfly spread shows that the 5-year bullet has become expensive relative to the 2/10 slope (Chart 11). Chart 11Favor The 2-Year Bullet Butterfly trades where you favor the bullet maturity versus the barbell perform well when the curve steepens. For example, the 2-year tends to outperform the 1/5 barbell when the 1/5 slope steepens. At present, the cheapness of the 2-year suggests that the butterfly spread is priced for significant 1/5 flattening in the coming months. Even stability in the 1/5 slope will cause the 2-year to outperform, and our key yield curve recommendation at the moment is to go long the 2-year bullet and short a duration-matched 1/5 barbell. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com Footnotes 1 Please see The Bank Credit Analyst, "OUTLOOK 2019: Late-Cycle Turbulence”, dated November 27, 2018, available at bca.bcaresearch.com 2 Please see U.S. Bond Strategy Weekly Report, “More Than One Reason To Own Steepeners”, dated September 25, 2018, available at usbs.bcaresearch.com 3 Please see U.S. Bond Strategy Weekly Report, “A Checklist For Peak Credit Spreads”, dated November 27, 2018, available at usbs.bcaresearch.com 4 Please see U.S. Bond Strategy Weekly Report, “An Oasis Of Prosperity?”, dated August 21, 2018, available at usbs.bcaresearch.com 5 We use the 3-year/10-year Treasury slope because it closely approximates the 2-year/10-year slope, but with more back-data. 6 Please see U.S. Bond Strategy Special Report, “The Golden Rule Of Bond Investing”, dated July 24, 2018, available at usbs.bcaresearch.com 7 Please see U.S. Bond Strategy Weekly Report, “What Kind Of Correction Is This?”, dated October 30, 2018, available at usbs.bcaresearch.com 8 Please see U.S. Bond Strategy Weekly Report, “A Checklist For Peak Credit Spreads”, dated November 27, 2018, available at usbs.bcaresearch.com 9 A combination of 17 different financial market variables that are highly levered to Chinese growth. Please see China Investment Strategy Weekly Report, “Trade Is Not China’s Only Problem”, dated November 21, 2018, available at cis.bcaresearch.com 10 Please see Commodity & Energy Strategy Weekly Report, “The Third Man At OPEC 2.0’s Meeting”, dated November 29, 2018, available at ces.bcaresearch.com 11 Please see U.S. Bond Strategy Weekly Report, “Adaptive Expectations In The TIPS Market”, dated November 20, 2018, available at usbs.bcaresearch.com 12 Please see U.S. Bond Strategy Weekly Report, “The Sweet Spot On The Yield Curve”, dated November 13, 2018, available at usbs.bcaresearch.com
