Inflation Protected

Highlights Chart 12020 Returns After a tumultuous start to the year, corporate bonds rallied in 2020 H2, managing to eke out small annual gains versus Treasuries. Specifically, investment grade corporates outperformed duration-equivalent Treasuries by 4 basis points in 2020 and high-yield outperformed by 185 bps (Chart 1). Treasuries, for their part, bested cash by 7% on the year but returns have been trending down since August. As we look forward to 2021, the economic cycle is in what we call a sweet spot for spread product returns. Economic growth is above trend, but inflation is low and monetary conditions are highly accommodative. This macro back-drop will lead to positive spread product returns versus Treasuries and a moderate bear-steepening of the Treasury curve in 2021. However, stretched valuations for investment grade corporates mean that investors must be selective within spread product. We think the Ba credit tier offers the best risk-adjusted opportunity in the corporate bond space, and also recommend favoring tax-exempt municipal bonds over equivalent-quality investment grade corporates. Feature Investment Grade: Neutral Chart 2Investment Grade Market Overview Investment grade corporate bonds outperformed the duration-matched Treasury index by 79 basis points in December and by 4 bps in 2020. The investment grade corporate index eked out a small gain relative to the duration-matched Treasury index in 2020. Corporates underperformed Treasuries by 18% from the beginning of the year until March 23, the day that the Fed stopped the bleeding in credit markets by unveiling its suite of emergency lending facilities. With the Fed’s backstops in place, the corporate index went on to outperform Treasuries by 22% between March 23 and the end of the year (Appendix A). As we noted in our 2021 Key Views Special Report, the corporate bond index option-adjusted spread is not quite back to its pre-COVID low.1 However, valuation is close to all-time expensive after adjusting for changes in the index’s average credit rating and duration. The 12-month breakeven spread for the Bloomberg Barclays Corporate Index (adjusted to keep the average credit rating constant) has only been tighter 4% of the time since 1995 (Chart 2). The same figure for the Baa-rated credit tier is 5%. As noted, the macro environment of above-trend growth and accommodative Fed policy is very positive for spread product returns. However, better value exists outside of the investment grade corporate space. In particular, we advise investors to look at Ba-rated high-yield corporates and tax-exempt municipal bonds. High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 190 basis points in December and by 185 basis points in 2020. Ba-rated junk bonds outperformed duration-matched Treasuries by 431 bps in 2020, while B-rated and Caa-rated bonds lagged by 13 bps and 238 bps, respectively. Since the March 23 peak in spreads, Ba-rated bonds outperformed Treasuries by 33%, B-rated bonds outperformed by 30% and Caa-rated bonds outperformed by 36% (Appendix A). We view Ba-rated junk bonds as the sweet spot within the corporate credit space. The sector is relatively insulated from default risk and yet still offers a sizeable spread pick-up over investment grade corporates (Chart 3). We noted in our 2021 Key Views Special Report that the additional spread earned from moving down in quality below Ba is in line with historical averages.2   Assuming a 25% recovery rate on defaulted debt and a minimum required risk premium of 150 bps, we calculate that the junk index is priced for a default rate of 2.8% for the next 12 months (panel 3). This represents a steep drop from the 8.4% default rate observed during the most recent 12-month period. However, only seven defaults occurred in November, down from a peak of 22 in July. Job cut announcements, an excellent indicator of the default rate, are also falling rapidly (bottom panel). Overall, we see room for spread compression across all junk credit tiers in 2021 but believe that Ba-rated bonds offer the best opportunity in risk-adjusted terms.    Table 3ACorporate Sector Relative Valuation And Recommended Allocation*Table 3BCorporate Sector Risk Vs. Reward*  MBS: Underweight Chart 4MBS Market Overview Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 22 basis points in December but underperformed by 17 bps in 2020. The conventional 30-year MBS index option-adjusted spread (OAS) tightened 10 bps on the month to reach 61 bps (Chart 4). This is higher than the 58 bps offered by Aa-rated corporate bonds, the 49 bps offered by Agency CMBS and the 24 bps offered by Aaa-rated consumer ABS. Despite the relatively attractive OAS, we continue to view the elevated primary mortgage spread as a material risk for MBS investors. The elevated spread suggests that mortgage rates need not rise alongside Treasury yields in the near-term, meaning that mortgage refinancings can continue at their current rapid pace (panel 3). Our view is that expected prepayment losses embedded in MBS spreads (aka the option cost) are too low relative to this pace of refinancing. Last year’s spike in the mortgage delinquency rate was driven by households that were granted forbearance by the federal government’s CARES act (panel 4). The risk for MBS holders is that these households will not be able to resume their regular mortgage payments when the forbearance period ends this spring. While the situation bears close monitoring, our sense is that excess savings built up during the past nine months will be sufficient to prevent a surge of bankruptcies when the forbearance period ends. The recent stimulus package provides households with even more assistance. Government-Related: Underweight Chart 5Government-Related Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 62 basis points in December but underperformed by 161 bps in 2020. Sovereign debt outperformed duration-equivalent Treasuries by 176 bps in December but underperformed by 98 bps in 2020. Foreign Agencies outperformed the Treasury benchmark by 7 bps in December but underperformed by 640 bps in 2020. Local Authority debt outperformed Treasuries by 146 bps in December but underperformed by 86 bps in 2020. Domestic Agency bonds outperformed by 14 bps in December but underperformed by 9 bps in 2020. Supranationals outperformed by 2 bps in December and by 3 bps in 2020. US dollar weakness is usually a boon for Emerging Market (EM) Sovereign and Foreign Agency returns. However, 2020’s dollar weakness was mostly relative to other Developed Market currencies (Chart 5). Value has improved somewhat for EM Sovereigns during the past few weeks, but the index continues to offer less spread than the Baa-rated US Credit index (panel 4). At the country level, Turkey, Colombia, Mexico, Russia, South Africa and Indonesia are the only countries that offer a spread pick-up relative to duration and quality-matched US corporates. Of those, only Mexico looks attractive on a risk/reward basis. Municipal Bonds: Overweight Chart 6Municipal Market Overview Municipal bonds outperformed the duration-equivalent Treasury index by 56 basis points in December but underperformed by 286 bps in 2020 (before adjusting for the tax advantage). We upgraded municipal bonds to “maximum overweight” in our recent 2021 Key Views Special Report.3 Attractive valuations are the main reason for this move. First, spreads between Aaa-rated municipal bonds and equivalent-maturity Treasuries are elevated compared to history across the entire yield curve (Chart 6). Second, municipal bonds look even more attractive relative to duration and quality-matched credit. The Bloomberg Barclays Revenue Bond index offers a greater yield than the quality-matched Credit index across the entire maturity spectrum (before adjusting for the tax advantage). The same is true for the Bloomberg Barclays General Obligation index beyond the 12-year maturity point (panel 3). While the failure to include state & local government aid in the recent relief bill is a big blow to municipal budgets that are already stretched, we think municipal bond spreads offer more-than-adequate compensation for default/downgrade risk. State & local governments are already engaging in austerity measures that will help protect bondholders (bottom panel) and State Rainy Day Fund balances were at all-time highs heading into the COVID downturn. Both of these things should help stave off a wave of municipal downgrades. Treasury Curve: Buy 5-Year Bullet Versus 2/10 Barbell Chart 7Treasury Yield Curve Overview The Treasury curve bear-steepened in December. The 2/10 Treasury slope steepened 13 bps to 81 bps. The 5/30 Treasury slope steepened 7 bps to 129 bps. Our expectation is that continued economic recovery will cause investors to price-in eventual monetary tightening at the long-end of the Treasury curve. With the Fed maintaining a firm grip on the front end, this will lead to Treasury curve bear steepening. A timely vaccine roll-out and the recently passed fiscal relief bill will serve to speed this process along. We recommend positioning for a steeper curve by owning the 5-year Treasury note and shorting a duration-matched barbell consisting of the 2-year and 10-year notes. This position is designed to profit from 2/10 curve steepening. Valuation is a concern with our recommended steepener, as the 5-year yield is below the yield on the duration-matched 2/10 barbell (Chart 7). However, the 5-year looked much more expensive during the last zero-lower-bound period between 2010 and 2013 (bottom 2 panels). We anticipate a return to similar levels.        TIPS: Overweight Chart 8TIPS Market Overview TIPS outperformed the duration-equivalent nominal Treasury index by 141 basis points in December and by 117 bps in 2020. The 10-year and 5-year/5-year forward TIPS breakeven inflation rates rose 22 bps and 18 bps on the month. They currently sit at 2.01% and 2.07%, respectively. Core CPI rose 0.22% in November, pushing the year-over-year rate from 1.63% to 1.65%. Meanwhile, 12-month trimmed mean CPI fell from 2.22% to 2.09%, narrowing the gap between trimmed mean and core (Chart 8). We anticipate further narrowing in 2021 Q1 and therefore expect core CPI to print relatively hot. For this reason, we recommend maintaining an overweight allocation to TIPS versus nominal Treasuries for the time being, even though the 10-year TIPS breakeven inflation rate looks somewhat elevated on our Adaptive Expectations Model (panel 2).4 Inflation pressures may moderate once the core and trimmed mean inflation measures converge, and this could give us an opportunity to tactically reduce TIPS exposure in the first half of this year. We also recommend holding real yield curve steepeners and inflation curve flatteners. With the Fed now officially targeting an overshoot of its 2% inflation goal, we expect the cost of 2-year inflation protection to rise above the cost of 10-year inflation protection (panel 4). With the Fed also exerting more control over short-dated nominal yields than over long-term ones, we expect short-maturity real yields to come under downward pressure relative to the long end (bottom panel).   ABS: Overweight Chart 9ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 15 basis points in December and by 98 bps in 2020. Aaa-rated ABS outperformed the Treasury benchmark by 12 bps in December and by 81 bps in 2020. Non-Aaa ABS outperformed by 33 bps in December and by 207 bps in 2020 (Chart 9). On paper, the Treasury department’s decision to let the Term Asset-Backed Loan Facility (TALF) expire at the end of 2020 is quite negative for ABS. However, as we explained in a recent report, we don’t anticipate a material impact on spreads.5 For one thing, Aaa ABS spreads are already well below the borrowing cost offered by TALF. But more importantly, consumer credit quality is strong. As we first explained last June, the stimulus received from the CARES act led to a significant increase in disposable income and a jump in the savings rate (panel 4).6 Faced with an income boost and few spending opportunities, many households paid down consumer debt. Given the recently passed additional fiscal support and the substantial savings that have already accrued, we see household balance sheets as being in a good place. As such, we advise moving down-in-quality to pick up extra spread in non-Aaa ABS.   Non-Agency CMBS: Neutral Chart 10CMBS Market Overview Non-Agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 113 basis points in December but underperformed by 57 bps in 2020. Aaa Non-Agency CMBS outperformed Treasuries by 58 bps in December and by 56 bps in 2020. Non-Aaa Non-Agency CMBS outperformed Treasuries by 277 bps in December but underperformed by 360 bps in 2020 (Chart 10). We continue to recommend an overweight allocation to Aaa-rated Non-Agency CMBS and an underweight allocation to non-Aaa CMBS. Even with the expiry of TALF, Aaa CMBS spreads are already well below the cost of borrowing through TALF and thus will not be negatively impacted.7 Meanwhile, the structurally challenging environment for commercial real estate could lead to problems for lower-rated CMBS (panels 3 & 4). Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 50 basis points in December and by 105 bps in 2020. The average index spread tightened 7 bps in December to reach 49 bps (bottom panel). At its September meeting, the Fed decided to slow its pace of Agency CMBS purchases. It is no longer looking to increase its Agency CMBS holdings, but rather, will only purchase what is “needed to sustain smooth market functioning”. This is nonetheless a backstop of the market, and it does not change our overweight recommendation.   Appendix A: Buy What The Fed Is Buying The Fed rolled out a number of aggressive lending facilities on March 23. These facilities focused on different specific sectors of the US bond market. The fact that the Fed has decided to support some parts of the market and not others has caused some traditional bond market correlations to break down. It has also led us to adopt of a strategy of “Buy What The Fed Is Buying”. That is, we favor those sectors that offer attractive spreads and that benefit from Fed support. The below Table tracks the performance of different bond sectors since the March 23 announcement. We will use this to monitor bond market correlations and evaluate our strategy’s success. TablePerformance Since March 23 Announcement Of Emergency Fed Facilities Appendix B: Butterfly Strategy Valuations The following tables present the current read-outs from our butterfly spread models. We use these models to identify opportunities to take duration-neutral positions across the Treasury curve. The following two Special Reports explain the models in more detail: US Bond Strategy Special Report, “Bullets, Barbells And Butterflies”, dated July 25, 2017, available at usbs.bcaresearch.com US Bond Strategy Special Report, “More Bullets, Barbells And Butterflies”, dated May 15, 2018, available at usbs.bcaresearch.com Table 4 shows the raw residuals from each model. A positive value indicates that the bullet is cheap relative to the duration-matched barbell. A negative value indicates that the barbell is cheap relative to the bullet. Table 4Butterfly Strategy Valuation: Raw Residuals In Basis Points (As Of December 31ST, 2020) Table 5 scales the raw residuals in Table 4 by their historical means and standard deviations. This facilitates comparison between the different butterfly spreads. Table 5Butterfly Strategy Valuation: Standardized Residuals (As Of December 31ST, 2020) Table 6 flips the models on their heads. It shows the change in the slope between the two barbell maturities that must be realized during the next six months to make returns between the bullet and barbell equal. For example, a reading of 85 bps in the 5 over 2/10 cell means that we would only expect the 5-year to outperform the 2/10 if the 2/10 slope steepens by more than 85 bps during the next six months. Otherwise, we would expect the 2/10 barbell to outperform the 5-year bullet. Table 6Discounted Slope Change During Next 6 Months (BPs) Appendix C: Excess Return Bond Map The Excess Return Bond Map is used to assess the relative risk/reward trade-off between different sectors of the US bond market. It is a purely computational exercise and does not impose any macroeconomic view. The Map’s vertical axis shows 12-month expected excess returns. These are proxied by each sector’s option-adjusted spread. Sectors plotting further toward the top of the Map have higher expected returns and vice-versa. Our novel risk measure called the “Risk Of Losing 100 bps” is shown on the Map’s horizontal axis. To calculate it, we first compute the spread widening required on a 12-month horizon for each sector to lose 100 bps or more relative to a duration-matched position in Treasury securities. Then, we divide that amount of spread widening by each sector’s historical spread volatility. The end result is the number of standard deviations of 12-month spread widening required for each sector to lose 100 bps or more versus a position in Treasuries. Lower risk sectors plot further to the right of the Map, and higher risk sectors plot further to the left. Chart 11Excess Return Bond Map (As Of December 31ST, 2020)   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Special Report, “2021 Key Views: US Fixed Income”, dated December 15, 2020, available at usbs.bcaresearch.com 2 Please see US Bond Strategy Special Report, “2021 Key Views: US Fixed Income”, dated December 15, 2020, available at usbs.bcaresearch.com 3 Please see US Bond Strategy Special Report, “2021 Key Views: US Fixed Income”, dated December 15, 2020, available at usbs.bcaresearch.com 4 For more details on our model please see US Bond Strategy Weekly Report, “How Are Inflation Expectations Adapting?”, dated February 11, 2020, available at usbs.bcaresearch.com 5 Please see US Bond Strategy Weekly Report, “Preparing For A Dark Winter … But Do Markets Care?”, dated November 24, 2020, available at usbs.bcaresearch.com 6 Please see US Bond Strategy Weekly Report, “No Holding Back”, dated June 16, 2020, available at usbs.bcaresearch.com 7 Please see US Bond Strategy Weekly Report, “Preparing For A Dark Winter … But Do Markets Care?”, dated November 24, 2020, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation

Special Report

This is US Bond Strategy’s final report of the year. Our regular publication schedule will resume on January 5th with our Portfolio Allocation Summary for January 2021. We wish you a happy, healthy and prosperous new year.   Highlights Interest Rate Policy: The Fed has given us a checklist of three criteria that must be met before it will lift rates off the zero bound. After those criteria are met, the pace of the eventual rate hike cycle will be determined by how quickly inflation expectations move back to “well-anchored” levels. We don’t expect Fed liftoff in 2021. Balance Sheet Policy: The Fed will only increase the pace or lengthen the maturity of its asset purchases if the economy or risk assets undergo a significant negative shock in 2021. Absent that, Fed communications in late-2021 will increasingly focus on the eventual tapering of asset purchases. Given the current vague guidance about when tapering will start, a scaled-down repeat of the 2013 Taper Tantrum is possible in late-2021 or 2022. Emergency Lending Facilities: The Fed will not undertake efforts to subvert Congress and re-establish its emergency lending facilities in 2021. However, the absence of the facilities will not have a negative impact on financial markets. Fiscal/Monetary Coordination: Looking beyond 2021, we see the lines between fiscal and monetary policy continuing to blur. The Fed will be increasingly incentivized to dip its toes into the fiscal arena and fiscal policymakers will let it. Feature Chart 1An Eventful Year It would be an understatement to say that 2020 was a busy year for the Federal Reserve. The Fed cut rates to the zero bound when the recession struck in March. It also exploded its balance sheet to fresh all-time highs and rolled out brand-new emergency lending facilities to support flagging credit markets (Chart 1). Then, to top it all off, the Fed concluded a Strategic Review of its monetary policy strategy in August and officially adopted an Average Inflation Target. This report touches on the market implications of 2020’s big Fed moves, but its focus is on what the Fed is likely to do in 2021. The first three sections discuss how we see the Fed’s interest rate policy, balance sheet policy and emergency lending facilities evolving next year. The final section considers a longer time horizon as it discusses what might be the next frontier for monetary policy: increased cooperation between monetary and fiscal authorities. Interest Rate Policy With the fed funds rate at its effective lower bound, bond investors will spend 2021 trying to determine the eventual start date and magnitude of the next tightening cycle. This will be especially complicated because the Fed’s adoption of an Average Inflation Target means that old models of its reaction function must be discarded. We discussed the implications of the move toward Average Inflation Targeting in a September Special Report.1 To quickly recap, the Fed made three main changes that will influence our outlook for interest rate policy in 2021. First, the Fed edited its Statement on Longer-Run Goals and Monetary Policy Strategy to include a new interpretation of its price stability mandate. The new Statement reads: In order to anchor longer-term inflation expectations at [2 percent], the Committee seeks to achieve inflation that averages 2 percent over time, and therefore judges that, following periods when inflation has been running persistently below 2 percent, appropriate monetary policy will likely aim to achieve inflation moderately above 2 percent for some time.2 Second, the Fed modified its Statement on Longer-Run Goals and Monetary Policy Strategy to signal that it will rely less on labor market indicators to forecast future inflation. In its old Statement, the Fed talked about minimizing “deviations of employment from the Committee’s assessments of its maximum level.” The revised Statement talks about mitigating “shortfalls of employment from the Committee’s assessment of its maximum level.” In other words, the Fed is saying that it will be less inclined to view an unemployment rate below its estimated natural level (NAIRU) as a signal that inflation is about to accelerate. The Fed’s adoption of an Average Inflation Target means that old models of its reaction function must be discarded. Finally, at the September FOMC meeting, the Fed translated the changes it made to its Statement on Longer-Run Goals and Monetary Policy Strategy into more explicit guidance about when it will consider lifting rates off the zero bound. That guidance is as follows: … it will be appropriate to maintain this target range until labor market conditions have reached levels consistent with the Committee’s assessments of maximum employment and inflation has risen to 2 percent and is on track to moderately exceed 2 percent for some time.3 The Timing Of Liftoff Table 1A Checklist For Liftoff Digging into the new guidance, we identify three criteria for lifting rates off the zero bound (Table 1). First, the unemployment rate must reach levels consistent with the Committee’s assessments of NAIRU. Currently, those estimates range from 3.5% to 4.5% (Chart 2). Practically, we view this as the least important of the three criteria. NAIRU estimates are revised based on what happens with inflation, and the Fed has already acknowledged that it is now less inclined to view a sub-NAIRU unemployment rate as an inflationary signal. In short, if inflation were to rise sustainably above 2% with the unemployment rate still at 5%, the Fed would simply revise up its NAIRU estimates and begin the rate hike cycle. Chart 2Criteria For Lifting Rates The Fed’s second criterion for lifting rates is also the most specific. Inflation must rise to 2 percent before the Fed will consider hiking rates. In a recent speech, Fed Vice-Chair Richard Clarida said he interprets this to mean that 12-month PCE inflation must be at least 2% before the Fed will consider hiking (Chart 2, bottom panel).4 This is helpful for bond investors. We can be certain that no rate hikes will occur at least until 12-month PCE inflation reaches 2%. Finally, the Fed also wants to be certain that inflation is “on track to moderately exceed 2 percent for some time.” This means that the event of 12-month PCE reaching 2% won’t automatically lead to a rate increase. The Fed must also view inflation gains as sustainable. This will likely become an issue in the first half of 2021 when we know that base effects will push 12-month PCE sharply higher, possibly even above 2%. However, we also know that those gains will be short lived.5 Of course, the Fed also knows about the impact of base effects and it will look past any temporary jump in inflation in H1 2021. More generally, while we advise investors to not pay much attention to the Fed’s NAIRU estimates, the unemployment rate will play a role in the Fed’s determination of whether above-2% inflation is sustainable. That is, the Fed is more likely to view above-2% inflation as sustainable if the unemployment rate is 4% than it is if the unemployment rate is 6%. What Does The Market Think? The bond market has been quick to price-in the big shift in the Fed’s interest rate guidance. At present, the overnight index swap curve is priced for a single 25 basis point rate hike in mid-2023 and only one more by mid-2024 (Chart 3). We see a good chance that the Fed’s three liftoff criteria are met before then, a view that forms the basis of our below-benchmark portfolio duration recommendation for 2021.6 In addition, the New York Fed’s Survey of Market Participants shows that only 43% of respondents expect liftoff before the end of 2023 and only 31% before the end of 2022 (Table 2A). This is further evidence that bond yields have room to rise if it looks like the Fed’s three liftoff criteria will be met in 2022 or the first half of 2023. Finally, the New York Fed’s survey shows that market participants understand the Fed’s three liftoff criteria and that differences in opinion about the timing of liftoff reflect differences in views about the economic outlook, not differences in understanding the Fed’s reaction function. The bulk of survey respondents think that the unemployment rate will be between 3.5% and 4.2% (consistent with the Fed’s NAIRU estimates) and that 12-month PCE inflation will be between 2.2% and 2.5% at the time of liftoff (Table 2B). Chart 3The Fed May Lift Rates Sooner Than Markets Expect Table 2ALiftoff Expectations Table 2BMarkets Understand The Fed’s Guidance The Pace Of Tightening And Why We’re Watching Inflation Expectations We’ve seen the three criteria upon which the Fed will condition its decision to hike rates off the zero bound. But the timing of liftoff is not the only thing that bond investors need to consider. We also need to get a sense of how quickly rate hikes will proceed once the next tightening cycle begins. According to the Fed’s interest rate guidance, even after liftoff the Fed will seek to maintain accommodative monetary conditions until it has achieved its price stability goal under its new Average Inflation Target. Recall that this goal is defined as achieving “inflation that averages 2 percent over time”. This is somewhat vaguer than the Fed’s liftoff guidance. Over what time period should we seek to hit average 2% inflation? One option is to start calculating the average when the new regime was adopted in August. In that case, average PCE inflation is running at 0.96%, well below 2%. Alternatively, we could calculate average inflation since the Fed last cut rates to zero in March 2020 (1.50%) or average inflation since the Fed cut rates to zero in December 2008 (1.43%). The point is that the Fed has not given us a clearly defined target. Differences in opinion about the timing of liftoff reflect differences in views about the economic outlook. For this reason, it’s important for bond investors to understand why the Fed has shifted to an Average Inflation Target. The reason has to do with trying to re-anchor inflation expectations. Box 1 shows an example of an Expectations-Augmented Phillips Curve, the Fed’s go-to framework for thinking about inflation. As the accompanying quote from Janet Yellen explains, the Fed thinks about inflation’s long-run trend as being driven by expectations. Shocks to economic slack and import prices can cause inflation to deviate from its long-run trend, but expectations drive the trend itself. This makes it critical for a central bank to keep expectations well anchored near its inflation target. Box 1The Expectations-Augmented Phillips Curve (aka The Fed’s Inflation Model) This is the underlying rationale for the Fed’s Average Inflation Target. The Fed has observed that inflation expectations have been too low in recent years. In the Fed’s model, this signals that inflation’s long-run trend has shifted down. In order to get expectations back up to target, the Fed understands that it will probably need to accept a period of above-target inflation. Since economic agents have just experienced a long period of sub-2% inflation, it will probably require a significant period of above-2% inflation before their expectations sustainably shift higher.7 To sum it all up, the Fed will seek to keep monetary conditions accommodative, and thus supportive for risk assets, until inflation expectations are deemed to be re-anchored. At that point, monetary policy will shift to a neutral or restrictive stance and risk asset performance will be challenged. But don’t just take our word for it. Here is what Vice-Chair Clarida said in a recent speech (referenced above): It is important to note, however, that the goal of the new framework is to keep inflation expectations well anchored at 2 percent, and, for this reason, I myself plan to focus more on indicators of inflation expectations themselves – especially survey-based measures – than I will on the calculation of an average rate of inflation over any particular window of time. It is clear that inflation expectations will dictate the eventual pace of Fed tightening. But the question of what measure of inflation expectations to track remains unresolved. Measures of inflation expectations fall into three main categories: Market-based measures Survey measures Trend measures Market-based measures are derived from inflation-linked bonds. Specifically, we derive TIPS breakeven inflation rates for different time horizons by taking the difference between a nominal yield and TIPS yield of the same maturity. In this publication, we often refer to the 10-year and 5-year/5-year forward TIPS breakeven inflation rates and have found that a range of 2.3% to 2.5% has historically been consistent with periods when inflation expectations were deemed “well anchored” (Chart 4A). One potential issue with using market-based measures of inflation expectations is that TIPS prices can sometimes move around for reasons unrelated to changing inflation expectations. That is, regulations or broader portfolio diversification concerns could change the risk premium an investor is willing to accept from TIPS, even if that investor’s underlying inflation view is unchanged. Academics have made attempts to solve this problem by using affine term structure models to decompose yields into various components. Chart 4B presents one such model from D’Amico, Kim and Wei (DKW).8 The DKW model splits the TIPS breakeven inflation rate (or inflation compensation) into an inflation expectation, a liquidity premium that compensates investors for the lower liquidity in TIPS compared to nominal Treasuries and an inflation risk premium that represents the extra compensation investors require to take inflation risk. We are skeptical of the usefulness of affine term structure models. In general, these models have too few inputs to reliably generate the components they purport to measure. However, the Fed clearly pays some attention to the DKW decomposition. If a future increase in TIPS breakeven inflation rates is driven entirely by movement in the liquidity or inflation risk premium components, it would be reasonable to question whether the Fed will react. Chart 4AInflation Expectations: Market-Based Measures Chart 4BA Decomposition Of TIPS##br## Breakevens Survey measures of inflation expectations are exactly that: Responses from surveys, usually of professional forecasters or households (Chart 4C). One drawback of survey measures compared to market-based measures is that they are updated less frequently. Another is that survey respondents, particularly households, may only be able to distinguish very large swings in prices. That said, the Fed tracks a wide range of survey measures and they were even singled out by Vice-Chair Clarida as being particularly important in the above quote. Trend inflation measures are statistical measures of the trend in the actual PCE or CPI inflation data. Chart 4D shows both a very simple trend measure, the 10-year annualized rate of change, and a slightly more complex trend measure based on an exponential smoothing rule. Academics have developed even more complex trend inflation measures.9 The logic behind these measures is that expectations tend to adapt only slowly to changes in the actual inflation data. Chart 4CInflation Expectations: Survey Measures Chart 4DInflation Expectations: Trend Measures Finally, we should point out a relatively new measure that the Fed will be using to track inflation expectations going forward. It is called the Common Inflation Expectations Index and it is a composite of 21 different survey and market-based inflation measures (Chart 4E), no trend inflation measures are included.10 Chart 4EIntroducing The Common Inflation Expectations Index To summarize, the Fed has given us a checklist of three criteria that must be met before it will lift rates off the zero bound. After those criteria are met, the pace of the eventual rate hike cycle will be determined by how quickly inflation expectations move back to levels that are considered “well anchored”. Once that happens, the Fed will no longer have an incentive to keep monetary conditions accommodative and risk asset performance will be challenged. Charts 4A-4E in this report provide a wide array of different measures of inflation expectations to monitor. We will keep an eye on all of them, but in particular, we will track the Common Inflation Expectations Index’s progress back to 2.1% and the 5-year/5-year forward TIPS breakeven inflation rate’s progress back to a range of 2.3%-2.5%. While we don’t expect the Fed’s rate hike criteria to be met in 2021, a 2022 liftoff is possible if the COVID vaccine spurs a rapid economic recovery. However, we do expect that, in 2021, the market will start to price-in an earlier liftoff date and quicker pace of tightening than is currently discounted, thus pushing bond yields higher. The Financial Conditions Wildcard Chart 5Financial Conditions Our base case view is that the eventual pace of Fed tightening will be determined by inflation expectations. However, there is one wildcard that could cause the Fed to abandon its inflation expectations goal and tighten policy earlier. That wildcard is financial conditions. Presently, financial asset valuations are a mixed bag (Chart 5). Corporate bond spreads are tight, but not at all-time expensive levels. Equity P/E ratios are very elevated, but equities don’t look expensive compared to bonds. If these valuations stay relatively stable, the Fed will continue to rely on inflation expectations to guide the pace of tightening. However, if inflation expectations take a long time to rise, it is conceivable that such a long period of low interest rates could lead to historically stretched financial asset valuations. In short, if inflation doesn’t return within the next couple of years, the Fed may have to tighten policy to take the wind out of an asset bubble that might otherwise burst and lead to an economic recession. We stress that we are not yet close to this point and that the bar for the Fed to abandon its inflation goal will be very high, but we would place financial conditions alongside inflation expectations as the two most important variables to monitor to assess the eventual pace of Fed tightening. Balance Sheet Policy With the funds rate pinned at zero and the Fed’s interest rate guidance essentially set in stone, changes to the pace and composition of asset purchases are the principal tool that the Fed will use to provide more or less immediate monetary accommodation in 2021. The Fed is currently purchasing $80 billion of Treasuries and $40 billion of Agency MBS each month, with Treasury purchases spread out across the yield curve. If this pace and distribution of Treasury purchases is maintained in 2021, the Fed will end up purchasing less and less of the Treasury flow. The Treasury Department has a stated policy goal of increasing the average maturity of the outstanding debt and it has been pursuing that goal by raising the amount of coupon issuance at the expense of bills. The Treasury has already given us its planned coupon issuance schedule for Q4 2020 and Q1 2021. Chart 6 shows that net Fed coupon purchases will gradually decline as a percentage of gross issuance, assuming the Treasury follows through with its plan and the Fed’s balance sheet policy is unchanged. Chart 6The Path For Treasury Supply And Fed Demand Can The Fed Do More? … Will The Fed Do More? It is possible that the Fed will use its balance sheet to provide more monetary easing in 2021. There are two ways it could do this. First, it could simply increase the monthly pace of asset purchases. Alternatively, it could keep the same pace of purchases but shift Treasury buying toward the long-end of the curve. The idea here would be to prevent long-dated yields from rising too quickly. One or both of these changes could happen in 2021, but only if the economy experiences a negative growth shock or risk asset prices (equities and corporate credit) fall significantly. In that case, the Fed will want to be seen as responding to a negative shock, but absent that, the Committee seems comfortable with its current balance sheet strategy. Chart 7Rate-Sensitive Sectors Have Recovered Some have suggested that, even if the economic recovery stays on track, the Fed will try to use its balance sheet to lean against rising long-maturity bond yields. We doubt this. First, it is not obvious that the Fed would be able to stop the 10-year Treasury yield from rising to a range of, say, 1.25% to 1.5% by increasing bond purchases in that maturity range by a few billion dollars. As long as the Fed’s interest rate guidance is unchanged, the market’s interest rate expectations will continue to exert a powerful influence on bond yields across the entire curve. Unless the Fed announces a cap on long-dated bond yields, and pledges to buy enough securities to enforce that cap, we are skeptical about the effectiveness of just changing the quantity of asset purchases. Second, it is also not clear that a 10-year Treasury yield between 1.25% and 1.5%, in the context of a steepening yield curve and improving economic growth, would be a problem for either the economy or risk assets. In fact, these sorts of environments tend to be very positive for risk asset performance.11 It is only when the Fed is shifting to a more restrictive monetary policy stance and the yield curve is flattening that bond yields start to exert a negative influence on the economy and risk assets. Even if the Fed is not worried about a moderate bear-steepening of the Treasury curve, a case could be made for providing more easing right now in order to spur a quicker recovery. This question was posed to Chair Powell several times at the last FOMC press conference. In response, Powell noted that the sectors of the economy that are most sensitive to interest rates – residential investment and consumer spending on durable goods – have already recovered (Chart 7). The lagging sectors of the economy – particularly consumer spending on services – cannot recover until the COVID vaccine is widely distributed, irrespective of the level of interest rates. In our view, this is an acknowledgement that the Fed does not see much value in trying to provide further accommodation through the balance sheet channel. All in all, our base case scenario is that the Fed will maintain its current pace and maturity distribution of asset purchases throughout 2021. However, it will increase the pace and/or lengthen the maturity if there is a significant shock to the economy and/or financial markets. Later in 2021, if the recovery stays on track, Fed communications will increasingly take up the issue of when it will be appropriate to taper its pace of asset purchases. The Exit From Asset Purchases And The Possibility Of A Taper Tantrum Chart 8Remember The Taper Tantrum The Fed has already given us a timeline for how it will wind down its asset purchases. According to the minutes from the November FOMC meeting, most participants support a timeline where the Fed will start tapering its pace of asset purchases sometime before the first rate hike. It will then begin lifting interest rates and will stop purchases altogether sometime after that. At the December FOMC meeting, the Fed gave us additional guidance on when it will start the tapering process. Unfortunately, this guidance is quite vague and only confirms the fact that tapering will start before the liftoff date. Specifically, the Fed said that tapering will begin when “substantial further progress has been made toward the Committee’s maximum employment and price stability goals.” Because the guidance around the timing of the tapering process is quite vague, we think it’s possible that it could sneak up on investors and lead to a sharp upward re-adjustment in rate hike expectations, and thus a bond sell-off. In essence, a tamer version of the 2013 Taper Tantrum is possible in late-2021 or 2022. On May 22, 2013, Fed Chair Ben Bernanke explained the Fed’s plan to eventually start tapering its asset purchases. Because investors took this to mean that the rate hike cycle would start much sooner than anticipated, the bond market underwent a sharp re-adjustment. The market quickly went from pricing-in only 35 bps of rate hikes over the next 24 months to 116 bps, and Treasury returns fell precipitously as a result (Chart 8). The Fed has learned a few lessons about communications since then, and it will do its best to keep market expectations aligned with its own strategy. However, unless firmer guidance is provided about when tapering will begin, the risk of a hawkish surprise around the tapering announcement remains.  Bottom Line: The Fed will only increase the pace or lengthen the maturity of its asset purchases if the economy or risk assets undergo a significant negative shock in 2021. Absent that, Fed communications in late-2021 will increasingly focus on the eventual tapering of asset purchases. Given the current vague guidance about when tapering will start, a scaled-down repeat of the 2013 Taper Tantrum is possible in late-2021 or 2022. Emergency Lending Facilities In addition to cutting rates to zero and massively scaling up the size of its balance sheet, the Fed also responded to the COVID recession by launching a slew of emergency lending facilities, some re-treads from the financial crisis and some brand new. Facilities to support the corporate bond market (The Primary and Secondary Market Corporate Credit Facilities) and the Municipal Liquidity Facility were particularly successful at capping bond spreads versus Treasuries, even if their actual usage was quite low. In fact, corporate bond spreads peaked on the very day that the Fed announced its corporate credit facilities in March (Chart 1). More recently, however, Treasury Secretary Steve Mnuchin refused to authorize the continuation of most of the Fed’s emergency lending facilities beyond the end of the year. We wrote in November that, even with the Treasury taking back the funds used to set up the facilities, the Fed could re-launch them in 2021 if incoming Treasury Secretary Janet Yellen provides her approval.12 However, a late addition to the recently passed fiscal stimulus package appears to prohibit the re-authorization of the facilities without Congressional approval. At the time of publishing, we have not been able to see the details of the new provision, so there remains some uncertainty about what the Fed can and cannot do in this regard. Credit spreads are no longer trading at distressed levels, primary issuance markets are functioning properly, and the Fed’s facilities have hardly been used at all. Nonetheless, while the new bill raises interesting questions about Fed independence in the long-run, we doubt that markets will respond negatively to the absence of the Fed’s emergency facilities in 2021. Credit spreads are no longer trading at distressed levels, primary issuance markets are functioning properly, and the Fed’s facilities have hardly been used at all (Table 3). Table 3Usage Of The 2020 Federal Reserve Emergency Lending Facilities In short, we don’t see the Fed going out of its way to re-establish the facilities in 2021 because it will become clear that they are no longer needed. The Next Regime Shift In Fed Policy: Fiscal/Monetary Coordination The adoption of an Average Inflation Target represents a major regime shift in Federal Reserve policy. In the final section of this report, we expand our horizon beyond 2021 and speculate about what the next major regime shift for the Fed might be. The 2020 recession made two things crystal clear. First, traditional fiscal policy becomes essentially impotent once interest have been reduced to the zero-lower-bound. Once there, Fed policy is most impactful when it focuses on lending to the private sector. Lending to the private sector through an emergency lending facility is an act that blurs the distinction between monetary and fiscal policy. This was made abundantly clear by Congress’ recent push to legislate the Fed’s activities in this arena. Second, it is difficult for fiscal policy to act quickly enough during an economic downturn. While the CARES act was delivered in a timely manner, it has taken many months to pass a follow-up bill. The Fed’s independence allows it to act immediately when it is economically necessary, while Congress’ increasing polarization makes swift action a challenge. If we take these two observations to their logical conclusion, and throw in the strong chance that the traditional channels of monetary policy will be increasingly blocked in the future as rates bump up against zero, it seems to us that the strict separation of responsibilities between fiscal and monetary policymakers will fade over time. More specifically, greater time spent at the zero-lower-bound will incentivize the Fed to get more and more creative with its quasi-fiscal private lending facilities. Further, Congress will be more than happy to allow this encroachment as it finds itself unable to respond effectively in times of crisis. Of course, this will also lead to periodic push-back as some members of Congress fret about the Fed’s over-reach, but we expect this push-back will be the exception rather than the rule. Greater time spent at the zero-lower-bound will incentivize the Fed to get more and more creative with its quasi-fiscal private lending facilities.  In fact, the best idea might be for fiscal and monetary policymakers to join together proactively to craft programs that can be deployed during the next recession. One example of such an idea was recently presented by Julia Coronado and Simon Potter.13 Coronado and Potter’s idea relies on the use of instant payment processing technology (which the Fed is already working on) to create digital accounts at the Federal Reserve for every household. Once those accounts are in place, the federal government could issue Recession Insurance Bonds, zero coupon bonds with some pre-determined face value, and grant one bond to every household in the country. Then, during the next economic downturn, the Fed could decide to do a “people’s QE” where it buys all the Recession Insurance Bonds leaving every household with a direct cash payment equal to the face value of the bond. This scheme directly addresses the two main problems we named earlier. It is fiscal policy, not monetary policy, so it can still be effective at the zero-lower-bound. Also, Congress can take its time to deliberate on the bill that authorizes the creation of the Recession Insurance Bonds, as this can be done proactively during a period of economic recovery. Then, the Fed can use its ability to move quickly during the next downturn to “activate” the bonds and deliver the fiscal stimulus that Congress actually passed years earlier. While likely not a story for 2021, we see increased cooperation between monetary and fiscal policymakers as the next big regime shift for the Fed.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy / Global Fixed Income Strategy Special Report, “A New Dawn For US Monetary Policy”, dated September 1, 2020, available at usbs.bcaresearch.com 2 https://www.federalreserve.gov/monetarypolicy/guide-to-changes-in-statement-on-longer-run-goals-monetary-policy-strategy.htm 3 https://www.federalreserve.gov/monetarypolicy/files/monetary20200916a1.pdf 4 https://www.federalreserve.gov/newsevents/speech/clarida20201116a.htm 5 For a more complete discussion of our 2021 inflation outlook please see “BCA Outlook 2021: A Brave New World”, dated November 30, 2020, available at bca.bcaresearch.com 6 Please see US Bond Strategy Special Report, “2021 Key Views: US Fixed Income”, dated December 15, 2020, available at usbs.bcaresearch.com 7 This is the theory of adaptive expectations and we use it to model changes in the 10-year TIPS breakeven inflation rate. For further details please see US Bond Strategy Weekly Report, “How Are Inflation Expectations Adapting?”, dated February 11, 2020, available at usbs.bcaresearch.com 8 https://www.federalreserve.gov/econres/notes/feds-notes/tips-from-tips-update-and-discussions-20190521.htm 9 Fed Governor Lael Brainard references several of these trend measures in this recent speech: https://www.federalreserve.gov/newsevents/speech/brainard20201021a.htm&…; 10 More details on how this aggregate measure is constructed can be found here: https://www.federalreserve.gov/econres/notes/feds-notes/index-of-common-inflation-expectations-20200902.htm 11 For more details on how corporate credit performs during different yield curve environments please see US Bond Strategy Special Report, “2021 Key Views: US Fixed Income”, dated December 15, 2020, available at usbs.bcaresearch.com 12 Please see US Bond Strategy Weekly Report, “Preparing For A Dark Winter … But Do Markets Care?”, dated November 24, 2020, available at usbs.bcaresearch.com 13 https://www.piie.com/publications/policy-briefs/reviving-potency-monetary-policy-recession-insurance-bonds

Special Report

Many investors feel that the Phillips Curve has failed to predict weak inflation over the past decade. But this perception is due to a singular focus on the economic slack component of the modern-day version of the curve to the exclusion of inflation expectations, and a failure to fully consider the lasting impact of sustained periods of a negative output gap on those expectations. In addition, many investors tend to downplay the long-term balance sheet impact of two episodes of excesses and savings/capital misallocations on the relationship between the stance of monetary policy and the output gap, via a persistently negative shock to aggregate demand and a reduced sensitivity of economic activity to interest rates. The COVID-19 pandemic was certainly a major economic shock. But for now, it seems like this was a sharp income statement recession, not a balance-sheet recession. This fact, along with lower odds of negative supply-side shocks and several structural factors, suggest that inflation will be higher over the next ten years than it has over the past decade. Investors looking to protect against potentially higher inflation should look primarily to commodities, cyclical stocks, and US farmland. Gold is likely to remain well supported over the coming few years, but rich valuation suggests the long-term outlook for the yellow metal is poor. A hybrid TIPS/currency portfolio has historically been strongly correlated with the price of gold, and may provide investors with long-term protection against inflation – at a better price. Introduction Chart II-1A Surge In Long-Dated Inflation Expectations The pandemic, and the corresponding fiscal and monetary response is challenging the low-inflation outlook of many market participants. Chart II-1 highlights that long-dated market-based inflation expectations have surged past their pre-COVID levels after collapsing to the lowest-ever level in March. The shift in thinking about inflation has partly been a response to an extraordinary rise in government spending in many countries. But Chart II-1 shows that long-dated expectations in the US were mostly trendless from April to June as Federal support was distributed, and instead rose sharply in July and August in the lead-up to the Fed’s official shift to an average inflation targeting regime. This new dawn for US monetary policy has been prompted not just by the pandemic, but also by the extended period of below-target inflation over the past decade. In this report, we review how the past ten-year episode of low inflation can be successfully explained through the lens of the expectations-augmented (i.e. “modern-day”) Phillips Curve. Many investors fail to fully appreciate the impact that inflation expectations have on driving actual inflation, as well as the cumulative impact of two major capital and savings misallocations over the past 25 years on the responsiveness of demand to interest rates and on the level of inflation expectations. Using the modern-day Phillips Curve as a guide, we present several reasons in favor of the view that inflation will be higher over the next decade than over the past ten years. Finally, we conclude with an assessment of several ways for investors to protect their portfolios from rising inflation. Revisiting The “Modern-Day” Phillips Curve The original Phillips Curve, as formulated by New Zealand economist William Phillips in the late 1950s, described a negative relationship between the unemployment rate and the pace of wage growth. Given the close correlation between wage and overall price growth at the time, the Phillips Curve was soon extended and generalized to describe an inverse relationship between labor market slack and overall price inflation. Chart II-2Rising Unemployment And Inflation Challenged The Original Phillips Curve However, the experience of rising inflation alongside high unemployment from the late 1960s to the late 1970s underscored that prices are also importantly determined by inflation expectations and shocks to the supply-side of the economy (Chart II-2). In the 1980s and 1990s, the Federal Reserve’s success at reigning in inflation was achieved not only by raising interest rates to punishingly high levels, but also by sharply altering consumer, business, and investor expectations about future prices. The experience of the late 1960s and 1970s led to a revised form of the Phillips Curve, dubbed the “expectations-augmented” or “modern” version. As an equation, the modern Phillips Curve is described today by Fed officials, in terms of core inflation, as follows: πct = β1πet + β2πct-1 + β3πct-2 - β4SLACKt + β5IMPt + εt where: πct = Core inflation today πet = Expectations of inflation πct-n = Lagged core inflation SLACKt = Slack in the economy IMPt = Imported goods prices εt = Other shocks to prices Described verbally, this framework suggests that “economic slack, changes in imported goods prices, and idiosyncratic shocks all cause core inflation to deviate from its longer-term trend that is ultimately determined by long-run inflation expectations.1” This framework can easily be extended to headline inflation by adding changes in food and energy prices. In most formal models of the economy in use today, the modern Phillips Curve is combined with the New Keynesian demand function to describe business cycles: Yt = Y*t – β(r-r*) + εt where: Yt = Real GDP Y*t = Real potential GDP r = The real interest rate r* = The neutral rate of interest εt = Other shocks to output This equation posits that differences in the real interest rate from its neutral level, along with idiosyncratic shocks to demand, cause real GDP to deviate from potential output. Abstracting from import prices and idiosyncratic shocks, these two equations tell a simple and intuitive story of how the economy generally works: The stance of monetary policy determines the output gap and, The output gap, along with inflation expectations, determine inflation. The Modern-Day Phillips Curve: The Pre-2000 Experience This above view of inflation and demand was strongly accepted by investors before the 2008 global financial crisis, but the decade-long period of generally below-target inflation has caused a crisis of faith in the idea of the Phillips Curve. Charts II-3 and II-4 show the historical record of the New Keynesian demand function and the modern-day Phillips Curve, using five-year averages of the data in question to smooth out the impact of short-term and idiosyncratic effects. We use nominal GDP growth as our long-run proxy for the neutral rate of interest,2 the US Congressional Budget Office’s (CBO) estimate of potential GDP to determine the output gap, and a proprietary measure of inflation expectations based on an adaptive expectations framework3 (Chart II-5). Chart II-3With Just Two Exceptions, Monetary Policy Strongly Explained Demand Before 2000 Chart II-4Similarly, Pre-2000 The Output Gap Generally Explained Unexpected Inflation Chart II-3 shows that until 1999, the stance of monetary policy was highly predictive of the output gap over a five-year period, with just two exceptions where major structural forces were at play: the late 1970s, and the second half of the 1990s. In the case of the former, the disruptive effect of persistently high inflation negatively impacted output growth despite easy monetary policy, and in the latter case, economic activity was modestly stronger than what interest rates would have implied due to the beneficial impact of the technologically-driven productivity boom of that decade. Similarly, Chart II-4 shows that until 1999 there was a good relationship between the output gap and the deviation in inflation from expectations, again with the late 1970s and late 1990s as exceptions. Along with the beneficial supply-side effects of the disinflationary tech boom, persistent import price weakness (via dollar strength) seems to have also played a role in suppressing inflation in the late 1990s (Chart II-6). Chart II-5The Expectations Component Of The Modern Phillips Curve, Visualized Chart II-6A Strong Dollar Also Played A Role In Suppressing Inflation During The 1990s   The Modern-Day Phillips Curve Post-2000 Following 2000, deviations between the monetary policy stance, the output gap, and inflation become more prominent, particularly after 2008. As we will illustrate below, these deviations are more apparent on the demand side. In the case of inflation, the question should be why inflation was not even lower in the years immediately following the global financial crisis. On both the demand and inflation side, these deviations are explainable, and in a way that helps us determine future inflation. Charts II-7 and II-8 show the same series as in Charts II-3 and II-4, but focused on the post-2000 period. From 2000-2007, Chart II-8 shows that the relationship between the output gap and the deviation in inflation from expectations was not particularly anomalous. The output gap was negative from the end of the 2001 recession until the beginning of 2006, and inflation was correspondingly below expectations on average for the cycle. Chart II-7Post-2000, The Output Gap Decoupled From The Monetary Policy Stance Chart II-8Since The GFC, The Real Mystery Is Why Inflation Has Been So Strong   Chart II-7 shows that the anomaly during that cycle was in the relationship between the output gap and the stance of monetary policy. Monetary policy was the easiest it had been in two decades, yet the output gap was negative for several years following the recession. Larry Summers pointedly cited this divergence in his revival of the secular stagnation theory in November 2013, arguing that it was strong evidence that excess savings were depressing aggregate demand via a lower neutral rate of interest and that this effect pre-dated the financial crisis. Why was demand so weak during that period? Chart II-9 compares the annualized per capita growth in the expenditure components of GDP during the 2001-2007 expansion to the 1991-2001 period. The chart shows that all components of GDP were lower than during the 1991-2001 period, with investment – the most interest rate sensitive component of GDP – showing up as particularly weak. On the surface, this supports the idea of structural factors weighing heavily on the neutral rate, rendering monetary policy less easy than investors would otherwise expect. But Chart II-9 treats the 2001-2007 years as one period, ignoring what happened over the course of the expansion. Chart II-10 repeats the exercise shown in Chart II-9 from Q1 2001 to Q3 2005, and highlights that the annualized growth in per capita residential investment was much stronger than it was during the 1991-2001 period – and nonresidential fixed investment was much weaker. Spending on goods was roughly the same, which is impressive considering that the late 1990s experienced a productivity boom and robust wage growth. All the negative contribution to growth from residential investment during the 2001-2007 expansion came after Q3 2005, as the housing market bubble burst in response to rising interest rates. In short, Chart II-10 highlights that there was a strong relationship between easy monetary policy and the demand for housing, but that this was not true for the corporate sector. Chart II-9Looking At The Whole 2001-2007 Period, Investment Was Extremely Weak Chart II-10Housing Absolutely Responded To Easy Monetary Policy   Explaining Weak CAPEX Growth In The Early 2000s This leads us to ask why CAPEX was so weak during the 2001-2007 period. In addition to changes in interest rates, business investment is strongly influenced by expectations of consumer demand and corporate profitability. Chart II-11 shows that real nonresidential fixed investment and as-reported earnings moved in lockstep during the period, and that this delayed corporate-sector recovery also impacted the pace of hiring. Weak expectations for consumer spending do not appear to be the culprit. Chart II-12 highlights that while real personal consumption expenditure growth fell during the recession, spending did not contract (as it had done during the previous recession) and capital expenditures fell much more than what real PCE would have implied. Chart II-11Post-2001, Persistently Weak Profits Led To Weak Investment And Jobs Growth Chart II-12CAPEX Was Much Weaker In 2002 Than Justified By Consumer Spending   Instead, persistently weak CAPEX in the early 2000s appears to be best explained by the damaging impact of corporate excesses that built up during the dot-com bubble. The Sarbanes-Oxley Act of 2002 was passed in response to a series of corporate accounting frauds that came to light in the wake of the bubble, but in many cases had been occurring for several years. Chart II-13 highlights that widespread write-offs badly impacted earnings quality and the growth in the asset value of equipment and intellectual property products (IPP), both of which only began to improve again in early 2003. This occurred alongside an outright contraction in real investment in IPP as investors lost faith in company financial statements and heavily scrutinized corporate spending. Chart II-14highlights that a contraction in IP spending was a huge change from the double-digit pace of growth that occurred in the late 1990s. Chart II-13The Damaging Impact Of Corporate Excesses Chart II-14A Near-Unprecedented Collapse In IPP Investment Followed The Tech Bubble   In addition, corporate sector indebtedness also appears to have played a role in driving weak investment in the early 2000s. While the interest burden of nonfinancial corporate debt was not as high in 2000 as it was in the early 1990s, Chart II-15 highlights that debt to operating income surged in the late 1990s – which likely caused investors already skeptical about company financial statements to impose a period of elevated capital discipline on corporate managers following the recession. Chart II-16 shows that while the peak in the 12-month trailing corporate bond default rate in January 2002 was similar to that of the early 90s, it was meaningfully higher on average in the lead-up to and following the recession. Chart II-15The Late-1990s Saw A Major Increase In Corporate Debt Chart II-16Above-Average Corporate Defaults Before And After The 2001 Recession   To summarize, Charts II-10-16 underscore that management excesses, governance failures, and elevated debt in the corporate sector in the 1990s were the root cause of the seeming divergence between monetary policy and the output gap from 2001 to 2007. This was, unfortunately, the first of two major savings/capital misallocations that have occurred in the US over the past 25 years. Explaining The Post-GFC Experience In the early 2000s, the Federal Reserve was faced with a decision between two monetary policy paths: one that was appropriate for the corporate sector, and one that was appropriate for the household sector. The Fed chose the former, and it inadvertently contributed to the second major savings/capital misallocation to occur over the past 25 years: the enormous debt-driven bubble in US housing that culminated into the global financial crisis (GFC) of 2007-2009. Chart II-17It Is No Mystery Why Demand And Inflation Were Weak Last Cycle As a result, 2007 to 2013/2014 was a mirror image of the early 2000s. Unlike previous post-war downturns, the GFC precipitated a balance-sheet recession that deeply affected homeowners and the financial system. This lasting damage led to a multi-year household deleveraging process, which substantially lowered the responsiveness of the economy to stimulative monetary policy. On a year-over-year basis, Chart II-17 shows that total nominal household mortgage credit growth was continuously negative for six and a half years, from Q4 2008 until Q2 2015, underscoring that the large divergence during this period between the stance of monetary policy and the output gap should not, in any way, be surprising to investors. And this is even before accounting for the negative impact of the euro area sovereign debt crisis and double-dip recession, or the persistent fiscal drag in nearly every advanced economy last cycle. What is surprising about the post-GFC experience is that inflation was not substantially weaker than it was, which is ironic considering that the secular stagnation narrative was revived to help explain below-target inflation. Chart II-8 showed that actual inflation steadily improved versus expected inflation alongside the closing of the output gap and the decline in the unemployment rate, but that it was much stronger than the output gap would have implied – particularly during the early phase of the economic recovery. It is still an open question as to why this occurred. A weak dollar and a strong recovery in oil prices likely helped support consumer prices, but we doubt that these two factors alone explain the discrepancy. A more credible answer is that expectations stayed very well anchored due to the Fed’s strong record of maintaining low and stable inflation (thus preventing a disinflationary spiral). In addition, the fact that the Fed actively communicated to the public during the early recovery years that a large part of its objective was to prevent deflation may have helped support prices. For example, in a CBS interview following the Fed’s November 2010 decision to engage in a second round of quantitative easing (“QE2”), then-Chair Bernanke prominently tied the decision to the fact that “inflation is very, very low.” When asked whether additional rounds of easing might be required, Bernanke responded that it was “certainly possible” and again cited inflation as a core consideration. Chart II-18Rising US Oil Production Caused The Massive 2014 Oil Price Shock While inflation did not ultimately fall relative to expectations post-GFC as much as the output gap would have implied, the long-lasting weakness in demand left expectations vulnerable to exogenous shocks. In 2014, such a shock occurred: oil prices collapsed almost exactly at the point that US tight oil production crossed the four-million-barrels-per-day mark (Chart II-18), a level of output that many experts had previously believed would not be attainable (or would roughly mark the peak in production). We view this event as a truly exogenous shock to prices, given that research & development of shale technology had been ongoing since the late 1970s and only happened to finally gain traction around 2010. Chart II-19 shows that the 2014 oil price collapse caused a clear break lower in our measure of inflation expectations, to the lowest value recorded since the 1940s. This break also occurred in market-based expectations of inflation, such as long-dated CPI swap rates and TIPS breakeven inflation rates, and surveys of consumer inflation expectations (Chart II-20). This decline in inflation expectations meant that the output gap needed to be above zero in order for the Fed to hit its 2% target (absent any upwards shock to prices), and that the meaningful acceleration of inflation from 2016 to 2018 should actually be viewed as inflation “outperformance” because its long-term trend had been lowered by the earlier downward shift in expectations. Chart II-19The 2014 Oil Price Shock Collapsed Inflation Expectations... Chart II-20...No Matter What Inflation Expectations Measure Is Used   The Modern-Day Phillips Curve: Key Takeaways Based on the evidence presented above, we see the perceived “failure” of the Phillips Curve to predict weak inflation over the past decade as being due to: A singular focus on the output gap/slack component of the modern Phillips Curve, to the exclusion of expectations A failure to fully consider the lasting impact of sustained periods of a negative output gap on expectations Downplaying the long-term balance-sheet impact of two episodes of excesses and savings/capital misallocations on the relationship between the stance of monetary policy and the output gap, via a persistently negative shock to aggregate demand and a reduced sensitivity of economic activity to interest rates. One crucial takeaway from the modern-day Phillips Curve equation presented above is that if inflation expectations are largely formed based on the experience of past inflation, then inflation is ultimately determined by three dimensions of the output gap: whether it is rising or falling, whether it is above or below zero, and how long it has been above or below zero. The extended period of below-potential output over the past two decades, accelerated recently by a major negative shock to energy prices, has now lowered inflation expectations to a point that merely reaching the Fed’s target constitutes inflation “outperformance.” This realization, made even more urgent by the COVID-19 pandemic, has strongly motivated the Fed’s official shift to an average inflation targeting regime. That shift does not suggest that the Fed is moving away from the modern-day Phillips Curve framework; rather, the Fed’s new policy is aimed at closing the output gap as quickly as possible in order to prevent a renewed decline in inflation expectations (and thus inflation itself) from another long period of activity running below its potential. The Outlook For Inflation While the Fed has shifted its policy to prefer higher inflation, that does not necessarily mean it will get it. Why is it likely to happen this time, if the last economic cycle featured such a large divergence between monetary policy and the output gap? Chart II-21Above-Target Inflation Is Not Imminent First, to clarify, we do not believe that above-target inflation is imminent. The COVID-19 pandemic was an extreme event, and even given the very substantial recovery in the labor market, the unemployment rate remains almost 2½ percentage points above the Congressional Budget long-run estimate of NAIRU (Chart II-21). But based on our analysis of the modern-day Phillips Curve presented above, there are at least four main reasons to expect that inflation may be higher on average over the next ten years than over the past decade. Reason #1: This Appears To Be A Sharp Income Statement Recession, Not A Balance-Sheet Recession We highlighted above the importance of savings/capital misallocations in driving a gap between monetary policy and the output gap over the past two decades, but this recession was obviously not sparked by such an event. The onset of the pandemic came following a long period of US household sector deleveraging which, while painful, helped restore consumer balance sheets. Chart II-22 highlights that household debt to disposable income had fallen back to 2001 levels at the onset of the pandemic, and the interest burden of debt servicing had fallen to a 40-year low. From a wealth perspective, Chart II-23 highlights that total household liabilities to net worth have fallen below where they were at the peak of the housing market boom in 2005 for almost all income groups, and that a decline in leverage has been particularly noteworthy for the lowest income group since mid-2016. Chart II-22Households Have Repaired Their Balance Sheets... Chart II-23...Across Almost All Income Brackets   Total credit to the nonfinancial corporate sector rose significantly relative to GDP over the course of the last cycle, but subpar growth in real nonresidential fixed investment and a rise in share buybacks highlight that this debt went largely to fund changes in capital structure rather than increased productive capacity. Chart II-24 highlights that corporate sector interest payments as a percentage of operating income are low relative to history, and they do not seem to be necessarily dependent on extremely low government bond yields.4 Finally, the corporate bond default rate may have already peaked (Chart II-25) and the percentage of jobs permanently lost looks more like 2001 than 2007 (Chart II-26), signaling that a prolonged balance-sheet recession is unlikely. Chart II-24Corporate Sector Debt Is Currently High, But Affordable Chart II-25Corporate Defaults Have Already Peaked Chart II-26So Far, Permanent Job Losses Look Like The 2001 Recession, Not 2007/2008 The bottom line is that while the pandemic has not yet been resolved and that major and permanent economic damage cannot be ruled out, the absence of “balance-sheet dynamics” is likely to eventually lead to a stronger responsiveness of demand for goods and services to what is set to be an extraordinarily easy monetary policy stance for at least another two years. Reason #2: The Fed May Be Able To Jawbone Inflation Higher The Fed’s public commitment to set interest rates in a way that will generate moderately above-target inflation is highly reminiscent of its defense of quantitative easing in the early phase of the last economic expansion, and (in the opposite fashion) of Paul Volker’s campaign in the 1980s against the “self-fulfilling prophecy” of inflation. From 2008-2014, the Fed explicitly linked the odds of future bond buying to the pace of actual inflation in its public statements. On its own, this was not enough to cause inflation to rise, but we highlighted above that it may have contributed to the fact that inflation expectations did not collapse. Chart II-1 on page 12 showed that long-dated market-based expectations for inflation have already been impacted by the Fed’s regime shift, suggesting decent odds that Fed policy will contribute to self-fulfilling price increases if the US economy does indeed avoid “balance-sheet dynamics” as a result of the pandemic. Reason #3: The Odds Of Negative Supply Shocks Are Lower Than In The Past We noted above the impact that energy price shocks and large typically exchange-rate driven changes in import prices can have on inflation, with the 2014 oil price collapse serving as the most vivid recent example. On both fronts, a value perspective suggests that the odds of negative shocks to inflation over the coming few years from oil and the dollar are lower than they have been in the past. Chart II-27 shows that the cost of global energy consumption as a share of GDP has fallen below its median since 1970, and Chart II-28 highlights that the US dollar is comparatively expensive relative to other currencies – which raises the bar for further gains. Stable-to-higher oil prices alongside a flat-to-weak dollar implies reflationary rather than disinflationary pressure. Chart II-27Massive, Downward Shocks To Oil Prices Are Now Less Likely Chart II-28Valuation Favors A Declining Dollar, Which Is Inflationary   Reason #4: Structural Factors In addition to the cyclical arguments noted above, my colleague Peter Berezin, BCA’s Chief Global Strategist, has also highlighted several structural arguments in favor of higher inflation. Chart II-29 highlights that the world support ratio, calculated as the number of workers relative to the number of consumers, peaked early last decade after rising for nearly 40 years. This suggests that output will fall relative to spending the coming several years, which should have the effect of boosting prices. Chart II-30 also highlights that globalization is on the back foot, with the ratio of trade-to-output having moved sideways for more than a decade. Since the early 1990s, rising global trade intensity has corresponded with very low goods prices in many countries, and the end of this trend reduces the impact of a factor that has been weighing on consumer prices globally over the past two decades. Chart II-29Less Production Relative To Consumption Is Inflationary Chart II-30Trade Is Not Suppressing Prices As Much As It Used To   Positioning For Eventually Higher Inflation Below we present an assessment of several potential candidates across the major asset classes that investors can use to protect their portfolios from rising inflation once it emerges. We conclude with a new trade idea that may provide investors with inflation protection at a better valuation profile than more traditional inflation hedges. Fixed-Income Within fixed-income, inflation-linked bonds and derivatives (such as CPI swaps) are the obvious choice for investors seeking inflation protection. Inflation-linked bonds are much better played relative to nominal equivalents, as inflation expectations make up the difference between nominal and inflation-linked yields. But Table II-1 shows that 5-10 year TIPS are also likely to provide positive absolute returns over the coming year even in a scenario where 10-year Treasury yields are rising, so long as real yields do not account for the vast majority of the increase. Barring a major and positive change in the long-term economic outlook over the coming year, our sense is that the Fed would act to cap any outsized increase in real yields and that TIPS remain an attractive long-only option until the Fed becomes sufficiently comfortable with the inflation outlook. Table II-1TIPS Will Earn Positive Absolute Returns Next Year Barring A Surge In Real Yields Commodities Commodities are arguably the most traditional inflation hedge, and are likely to provide investors with superior risk-adjusted returns in an environment where inflation expectations are rising. Our Commodity & Energy Strategy service is positive on gold, and recently argued that Brent crude prices are likely to average between $65-$70/barrel between 2021-2025.5 Chart II-31Gold Is Expensive And Long-Term Returns May Be Poor One caveat about gold is that, unlike oil prices, it appears to be quite expensive relative to its history. Since gold does not provide investors with a cash flow, over time real (or inflation-adjusted) prices should ultimately be mean-reverting unless real production costs steadily trend higher. Chart II-31 highlights that the real price of gold is already sky-high and well above its historical average. Over a ten-year time horizon, gold prices fell meaningfully following the last two occasions where real gold prices reached current levels, suggesting that the long-term outlook for gold returns is poor. However, over the coming few years, gold prices are likely to remain well supported given our economic outlook, the Fed’s new monetary policy regime, and the consistently negative correlation between real yields and the US dollar and gold prices. As such, we would recommend gold as a hedge against the fear of inflation, which is likely to increase over the cyclical horizon. Equities We provide two perspectives on how equity investors may be able to protect themselves against rising inflation. The first is simply to favor cyclical versus defensive sectors. The former is likely to continue to benefit next year in response to a strengthening economy as COVID-19 vaccines are progressively distributed, and historically cyclical sectors have tended to outperform during periods of rising inflation. In addition, my colleague Anastasios Avgeriou, BCA’s Equity Strategist, presented Table II-2 in a June Special Report,6 and it highlights that cyclical sectors (plus health care) have enjoyed positive relative returns on average during periods of rising inflation. Table II-2S&P 500 Sector Performance During Inflationary Periods The second strategy is to favor companies that are more likely to successfully pass on increasing prices to their customers (i.e., firms with “pricing power”). Pricing power is a difficult attribute to identify, but one possible approach is to select industries that have experienced above-average sales per share growth over the past decade. While it is true that the past ten years have seen low rather than high inflation, it has also seen firms in general struggle to achieve robust top-line growth. Industries that have succeeded in this environment may thus be able to pass on higher costs to their customers without disproportionately suffering from lower sales. Chart II-32Last Decade's Revenue Winners: Potential Pricing Power Candidates Chart II-32 presents the historical relative performance of these industries in the US plus the materials and energy sector, equally-weighted and compared to an equally-weighted industry group portfolio (level 2 GICS). The chart shows that the portfolio has outperformed steadily over the past decade, although admittedly at a slower pace since 2018. An interesting feature of this approach is that, in addition to including industries within the industrials, consumer discretionary, and health care sectors (along with the food & staples retailing component of the consumer staples sector), tech stocks show up prominently due to their outstanding revenue performance over the past decade. Table II-2 above highlighted that tech stocks have historically performed poorly during periods of rising inflation, although it is unclear whether this is due to increasing prices or expectations of rising interest rates. Tech stocks are typically long-duration assets, meaning that they are very sensitive to the discount rate, but the Fed’s new monetary policy regime all but guarantees that investors will see a gap between inflation and rates for a time. It is thus an open question how tech stocks would perform in the future in response to rising inflation, and we plan to revisit this topic in a future report. Chart II-33Owners Of Existing Infrastructure Assets Are Primarily Utilities And Telecom Companies As a final point within the stock market, we would caution against equity portfolios favoring companies that are owners or operators of infrastructure assets. While increased infrastructure spending may indeed occur in the US over the coming several years, indexes focused on companies with sizeable existing infrastructure assets tend to be highly concentrated in the utilities and telecommunications sectors. Chart II-33 shows that the relative performance of the MSCI ACWI Infrastructure Index is nearly identical to that of a 50/50 utilities/telecom services portfolio, two sectors that are defensive rather than pro-cyclical and that have historically performed poorly during periods of rising inflation. Direct Real Estate Alongside commodities, direct real estate investment is also typically viewed as a traditional inflation hedge. For now, however, the outlook for important segments of the commercial real estate market is sufficiently cloudy that it is difficult to form a high conviction view in favor of the asset class. CMBS delinquency rates on office properties have remained low during the pandemic, but those of retail and accommodation have soared and the long-term outlook for all three may have permanently shifted due to the impact of the pandemic. By contrast, industrial and medical properties are likely to do well, with the former likely to be increasingly negatively correlated with the performance of retail properties in the coming few years (i.e., “warehouses versus malls”). I noted my colleague Peter Berezin’s structural arguments for inflation above, and Peter has also highlighted farmland as a real asset that is likely to do well in an environment of rising inflation.7 Chart II-34 further supports the argument: the chart shows that despite a significant increase in real farm real estate values over the past 20 years, returns to operators as a % of farmland values are not unattractive. In addition, USDA forecasts for 2020 suggest that operator returns will be the highest in a decade relative to current 10-year Treasury yields, underscoring both the capital appreciation and relative yield potential of US farmland. A Hybrid TIPS/Currency Inflation-Hedged Portfolio Finally, as we highlighted in Section 1, in a world of extremely low government bond yields, global ex-US investors have the advantage of being able to hedge against deflationary risks in a long-only portfolio by employing the US dollar as a diversifying asset. The dollar is consistently negatively correlated with global stock prices, and this relationship tends to strengthen during crisis periods. The flip side is that US-based investors have the advantage of being able to hedge against inflationary risks in a long-only portfolio by buying global currencies. Chart II-35 presents a 50/50 portfolio of US TIPS and an equally-weighted basket of six major DM currencies against the US dollar. The chart highlights that the portfolio is strongly positively correlated with gold prices, but with a better valuation profile. We already showed in Chart II-28 on page 28 that global currencies are undervalued versus the US dollar. TIPS valuation is not as attractive given that real yields are at record low levels, but the 10-year TIPS breakeven inflation rate currently sits at its 40th percentile historically (and thus has room to move higher). Chart II-34Farmland: Protection Again Inflation, At A Decent Yield Chart II-35A Hybrid TIPS/Currency Portfolio: Liquid, And Cheaper Than Gold   As such, while gold prices are likely to remain supported over the cyclical horizon, a hybrid TIPS/currency portfolio may also provide investors with long-term protection against inflation – at a better price. Jonathan LaBerge, CFA Vice President The Bank Credit Analyst Footnotes 1 “Inflation Dynamics and Monetary Policy,” Janet Yellen, Speech at the Philip Gamble Memorial Lecture, University of Massachusetts - Amherst, Amherst, Massachusetts, September 24, 2015. 2 The use of nominal GDP growth as our proxy for the neutral rate of interest is based on the idea that borrowing costs are stimulative if they are below that of income growth. 3 An adaptive expectations framework suggests that expectations for future inflation are largely determined by what has occurred in the past. Our proxy for inflation expectations is thus calculated using simple exponential smoothing of the actual PCE deflator, which provides us with a long and consistent time series for expectations. 4 The second debt service ratio shown in Chart II-24 would only rise to its 68th historical percentile if the 10-year Treasury yield were to rise to 3%, or the 75th with a 10-year yield at 4%. This would be elevated relative to history, but not extreme. 5 Please see Commodity & Energy Strategy Report “BCA’s 2021-25 Brent Forecast: $65-$70/bbl,” dated November 12, 2020, available at ces.bcaresearch.com 6 Please see US Equity Strategy Special Report “Revisiting Equity Sector Winners And Losers When Inflation Climbs,” dated June 1, 2020, available at uses.bcaresearch.com 7 Please see Global Investment Strategy Weekly Report “Will There Be A Fiscal Hangover?” dated May 29, 2020, available at gis.bcaresearch.com

Highlights Global growth will accelerate over the course of 2021 as COVID-19 vaccines are distributed and economic confidence improves in response. Longer-term global bond yields see some upward pressure as growth picks up, but global real yields will stay negative with on-hold central banks actively seeking an inflation overshoot. Maintain below-benchmark overall global duration exposure, and position for steeper government bond yield curves and wider inflation breakevens. The rise in global bond yields we anticipate will be relatively moderate, with US Treasury yields rising the most. Underweight the US in global bond portfolios, and favor countries where yields have a lower sensitivity to rising US yields (core Europe, Japan, UK). Also overweight Peripheral European debt given supportive monetary and fiscal policies that are helping to reduce credit risk (Italy, Spain, Portugal). The US dollar will remain soft in 2021, providing an additional reflationary impulse to the global economy. Overweight global inflation-linked bonds versus nominal government debt. Lower-quality global credit should outperform against a backdrop that will prove positive for risk assets: easy money policies, improving growth momentum and a reduction in virus-related uncertainty. Upgrade US high-yield to overweight through higher allocations to lower rated credit tiers, while downgrading US investment grade, where valuations are far less compelling, to neutral. Favor US corporates versus euro area equivalents, of all credit quality, based off less attractive euro area spread valuations. Within US$-denominated emerging market debt, favor corporates over sovereigns. Feature Dear Client, This report, detailing our global fixed income investment outlook for next year, will be our last for 2020. Please join me for a webcast this coming Friday, December 18 at 10:00 AM EST (3:00 PM GMT, 4:00 PM CET, 11:00 PM HKT) where I will discuss the outlook followed by a Q&A session. Best wishes for a very safe, healthy and prosperous 2021. We’ve all earned that after a difficult 2020 that none of us will soon forget. Rob Robis, Chief Global Fixed Income Strategist BCA Research’s Outlook 2021 report, “A Brave New World”, outlining the main investment themes for next year based on the collective wisdom of our strategists, was sent to all clients in late November.1 In this report, we discuss the broad implications of those themes for the direction of global fixed income markets in 2021. In a follow-up report to be published in the first week of the New Year, we will translate those themes into specific recommended allocations and weightings within our model bond portfolio framework. A Summary Of The 2021 BCA Outlook The tone of the BCA 2021 Outlook was generally positive, with conclusions that are supportive for the outperformance of risk assets relative to safe havens like government bonds (Chart 1). Chart 1How To Play Recovery & Reflation In 2021 Global growth will strengthen over the course of next year, after an initial soft patch related to the late-2020 COVID-19 economic restrictions in Europe and the US. Economic confidence will improve as the COVID-19 vaccines become more widely distributed, at a time of ongoing substantial monetary and fiscal stimulus in most important countries. A major release of pent-up demand is likely, fueled by the surge in private sector savings in the US and Europe after households and businesses cut back on spending because of the pandemic. The lingering impact of China’s substantial fiscal and credit stimulus in 2020 will still be felt throughout the world for most of 2021, even with Chinese authorities likely to begin curtailing the expansion of credit around mid-year. The tremendous amount of global spare capacity created by the virus and associated economic restrictions will keep inflation subdued in most countries. Thus, both monetary and fiscal policymakers will be under no pressure to pre-emptively tighten policy. The pace of monetary/fiscal stimulus will inevitably slow on a rate-of-change basis after the massive ramp up of government spending, income support, loan guarantees and central bank asset purchases. However, policymakers are expected to pull any and all of those levers once again in the event of a severe pullback in economic growth or a major bout of financial market turbulence. After a wild 2020 in a US election year, geopolitical uncertainty is expected to recede a bit next year. Although US-China tensions will remain elevated even under the incoming Biden administration, European politics are expected to be a tailwind for financial markets. A UK-EU Brexit deal is expected to be reached given economic realities, increased fiscal cooperation within the EU will support fiscally weaker countries like Italy, and the threat of the US imposing tariffs on Europe will disappear after Donald Trump leaves office. Our Four Main Key Views For Global Fixed Income Markets In 2021 The following are the main implications for global fixed income investment strategy based off the conclusions from the 2020 BCA Outlook: Key View #1: Maintain below-benchmark overall global duration exposure, and position for steeper government bond yield curves and wider inflation breakevens. Chart 2COVID-19 Lockdowns Will Not Last Forever COVID-19 was the elephant in the room for financial markets in 2020, influencing sentiment whenever cases flared up or subsided. Yet the impact diminished steadily since the first wave of the virus stretched beyond China in the spring. The broad span of global risk assets – equities, corporate credit, industrial commodities – has performed very well during the current, and much larger, surge in cases occurring in the US and Europe. One big reason for this is that investors now understand that lockdowns, and the associated drag on economic growth, do not last forever. In addition, investors know that policymakers in most countries will react to any sharp downturn in economic confidence with more fiscal and monetary stimulus to help offset the negative growth impact of the lockdowns. In Europe, many European governments enacted harsh national lockdowns in a bid to “flatten the curve” during the latest surge. This has helped successfully reduce the growth rate of new cases and hospitalizations (Chart 2). This will eventually lead to an easing of restrictions, and a recovery in economic activity, in early 2021. While US case numbers are also surging, the response by governments has been much less widespread, and severe, compared to Europe. There is little political appetite (even with a new president) for another wave of harsh restrictions along the lines of what took place last spring. Some slowing of economic activity is inevitable because of increased regional restrictions in large states like California and New York, as is already evident in some late-2020 data. However, any downturn should not be expected to last long with the growth rate of US COVID-19 hospitalizations having already peaked. The big game-changer, of course, is the introduction of COVID-19 vaccines which have already begun to be distributed in the UK and US. While there are uncertainties related to the operational logistics of a worldwide vaccine rollout, including whether enough people will voluntarily choose to be vaccinated to achieve herd immunity on a global scale, the very high announced efficacy levels of the various vaccines mean that an end of the pandemic is now achievable. Investors should see through the current surge in COVID-19 cases, and any short-term hiccup in economic growth, and focus on the bigger picture of the introduction of the vaccine and the positive implications for global economic confidence in 2021. Growth has already been holding up well in the US and China in the final months of 2020, with both manufacturing and services PMIs remaining solidly above the 50 line indicating expanding activity. As the euro area lockdowns begun to ease up, growth there will catch up, which already appears to be underway with the sharp uptick in the December PMI data (Chart 3). Those three regions account for one-half of worldwide GDP, so that is already a solid footing for global growth entering 2021. A sustained improvement in the pace of global economic activity is important, as it is becoming increasingly harder for governments to sustain the extreme levels of policy stimulus delivered in 2020. In China, policymakers are starting to rotate their focus away from aggressive stimulus and fighting deflation back to the cautious risk management approach to credit expansion that was in place prior to COVID-19. BCA Research’s China strategists expect the latest Chinese credit cycle to peak by mid-2021, with the credit impulse set to decline in the second half of the year (Chart 4). Combined with the tightening of monetary conditions through a strengthening yuan and higher local interest rates, some slowing of Chinese growth is inevitable. Although given the lags between stimulus and growth, the impact is more likely to be felt toward year-end and into 2022 – good news for much of the global economy that still relies heavily on exporting to China as an engine of growth. Chart 3A Growth Recovery Without Inflation Chart 4China Stimulus Will Peak Out By Mid-2021 Overall global fiscal policy is on track to be less supportive in 2021. The latest estimates from the IMF show that the “fiscal thrust”, or the change in the cyclically-adjusted primary budget balance relative to potential GDP, in most developed economies will turn negative next year (Charts 5A and 5B). Such a swing is inevitable given the sheer magnitudes of the fiscal stimulus measures first introduced to combat the economic damage from COVID-19 that will not be repeated in 2021. By the same token, less fiscal stimulus will be necessary if overall global growth improves, especially if vaccines can be successfully distributed to much of the world. Chart 5ANegative Fiscal Thrust In 2021 … Chart 5B… But Governments Will Spend More If Needed What does all this mean for global government bond yields? We believe that it signals a continuation of the trends seen towards the end of 2020 – a slow grind higher in longer-term yields, led by better growth and rising inflation expectations, but without any need to discount a move to tighter monetary policy because of a sustained overshoot of realized inflation. The current economic projections of the Fed, ECB, Bank of England (BoE), Bank of Canada (BoC) and Reserve Bank of Australia (RBA) all show that policymakers there expect unemployment rates to remain above pre-pandemic levels to at least 2023 (Chart 6). At the same time, central banks are also projecting inflation to be below their target levels/ranges over that same period. In response, the forward guidance from these central banks has been very dovish, with policy interest rates expected to remain at current levels at or near 0% for at least the next two to three years. Interest rate markets have taken the hint, with a very low expected path for rates over the next few years discounted in overnight index swap curves. Chart 6Central Banks Projecting A Slow Return To Full Employment Chart 7Markets Expect Years Of Negative Real Policy Rates The implication of this is that central banks are projecting a sustained, multi-year period where policy rates will remain below forecasted inflation (Chart 7). Or put more simply, central banks are consistently signaling that negative real interest rates will persist for a long time. This means that one of the most oft-discussed “oddities” of global bond markets in 2020 - the persistence of negative real long term bond yields in most major economies, most notably in the US Treasury market, even as inflation expectations increase – is unlikely to disappear in 2021. Those negative real yields reflect, to a large part, the expectation that real global policy rates will stay persistently negative (Chart 8). At some point in 2021, markets could challenge this dovish guidance from central banks that could temporarily push up both future interest rate expectations and longer-term real yields, especially in the US. However, it is more likely that central banks will not validate that move higher in yields for fears of pre-emptively short-circuiting an economic recovery. Such a hawkish shift could be more plausibly delivered in 2022 at the earliest, with the Fed the most likely candidate to change its guidance. Summing up all of the above points with regards to our recommendations on overall management of government bond portfolios, we arrive at the following conclusions (Chart 9): Chart 8Rising Inflation Breakevens With Stable Negative Real Yields Chart 9Moderately Higher Global Bond Yields In 2021 Duration exposure should be set below-benchmark. Our forward-looking Duration Indicator, comprised of leading economic indicators and economic expectations data, is strongly signaling that global yields should head higher in 2021. Position for a bearish steepening of yield curves. This will be driven more by rising longer-term inflation expectations, as the short-ends of yield curves will remain anchored by dovish on-hold central banks. Key View #2: Underweight the US in global bond portfolios, and favor countries where yields have a lower sensitivity to rising US yields Moving beyond the overall global duration view, there are significant country allocation decisions that derive from our outlook for 2021. First and foremost, we recommend underweighting US Treasuries in global bond portfolios, as we anticipate the biggest increase in developed market bond yields next year to occur in the US. We expect the benchmark 10-year Treasury yield to rise to the 1.25% to 1.5% range sometime in 2021. This move will come mostly through higher inflation expectations. The 10-year TIPS breakeven inflation rate is expected to reach the 2.3-2.5% range that we have long considered to be consistent with the market pricing in the Fed sustainably achieving its 2% inflation goal. Any additional Treasury yield increases beyond our 2021 forecast range would require the Fed to shift to a more hawkish stance signaling future rate hikes. With the Fed now operating with an Average Inflation Target framework, allowing for temporary overshoots of inflation after periods when inflation was below the Fed’s 2% target, the hurdle for such a shift in Fed guidance is much higher than in previous years. The Fed has also changed the nature of its forward guidance compared to years past, signaling that any future monetary tightening will only occur once actual inflation has sustainably returned to the 2% target. That means that the Fed will no longer pre-emptively choose to hike rates on merely a forecast of higher inflation – it will first need to see a sustained period of higher inflation materialize before considering any tightening. Thus, any move beyond our expected 1.25% to 1.5% range on US Treasuries would require a hawkish signal by the Fed that it intends to begin removing monetary accommodation through rate hikes. Under the Average Inflation Target framework, that will not happen in 2021 but could happen the following year if inflation stays at or above 2% over the course of next year. Turning to other countries, we recommend favoring bond markets with a lower historical “yield beta” to US Treasuries. In other words, we prefer overweighting counties where government bond yields are typically less correlated to changes in Treasury yields. We show those historical yield betas, using 10-year yields, in Chart 10. Importantly, the betas are calculated only for periods when Treasury yields are moving higher. We call this “upside beta”, which is a useful tool to identify which bond markets are more sensitive to selloffs in the US Treasury market. Chart 10Favor Lower Beta Government Bond Markets In 2021 The highest “upside beta” countries among the major developed markets are Australia, Canada and New Zealand, while the lowest “upside beta” countries are Germany, France and Japan. The UK is in the middle of those two groupings, although the trend over the past few years suggests that it is transitioning from a high-beta to low-beta country. Note that for all countries shown, the upside yield betas are below one, indicating that no market should be expected to see a bigger rise in yields than the US. Strictly based on our forecast of higher Treasury yields and calculated yield betas, we would recommend more overweight allocations to markets in the lower-beta group and more underweight allocations to the higher-beta group. We are comfortable recommending overweights to the lower-beta group of Germany, France, Japan and the UK. Although among the higher-beta group, we are reluctant to recommend underweighting all three countries because of the policy choices of their central banks. The RBA, BoC and Reserve Bank of New Zealand (RBNZ) have all enacted aggressively large quantitative easing (QE) programs in 2020 as a way to provide additional monetary stimulus after cutting policy rates to near-0%. The BoC stands out as being extremely aggressive on QE with its balance sheet expanding more than three-fold on a year-over-year basis (Chart 11). Chart 11More Divergence In The Pace Of Global QE None of these three central banks has discussed slowing the pace of purchases anytime soon. In the case of the RBA and RBNZ, they have gone as far as signaling the role of QE in dampening their bond yields to help stem the appreciation of their currencies. They may have limited success in driving down yields further, however. Measures of bond valuation like the term premium, which typically move lower when QE accelerates, have bottomed out across the developed markets even as central banks have absorbed a greater share of the stock of government debt in 2020 (Chart 12). Yet even if QE can no longer drive yields lower, it can limit how much yields can increase when under cyclical upward pressure. For this reason, we do not expect government bond yields in Australia, Canada or New Zealand to behave in line their historical higher yield beta that would make them clear underweight candidates in a period of rising US Treasury yields, as we expect. Net-net, we recommend that investors focus underweights solely on US Treasuries within global government bond portfolios. This suggests that yield spreads between Treasuries and other bond markets should continue to widen, as has been the case over the final few months of 2020 (Chart 13). We recommend neutral allocations to Australia, Canada and New Zealand, while overweighting core Europe, Japan and the UK. Chart 12More QE Is Less Impactful In Pushing Down Bond Yields Chart 13US Treasuries Will Continue To Underperform In 2021 We also are maintaining our overweight recommendation on Italian and Spanish government debt, which was one of our most successful calls of 2020. We view those markets more as a credit spread story versus core Europe, rather than a directional yield instrument like US Treasuries or German Bunds. On that basis, the spread of Italian and Spanish yields versus German yields has room to compress even further, as both are strongly supported by ECB bond purchases. Also, the introduction of the European Union’s €750bn Recovery Fund is a strong signal of greater fiscal co-operation within Europe – another important factor that has helped reduce the risk premium (credit spread) on Italy and Spain. When looking at the yields currently on offer in the developed world, Italy and Spain offer very attractive yields in a global low-yield environment (Table 1). Stay overweight. Table 1Developed Market Bond Yields, Both Unhedged & Hedged Into USD Key View #3: Overweight global inflation-linked bonds versus nominal government debt We have discussed the importance of rising inflation expectations as a core driver of the rise in global bond yields that we expect in 2021. This has been in the context of improving global growth, reduced spare economic capacity and central banks staying very dovish, all of which are necessary ingredients to boost depressed inflation expectations. A weaker US dollar will also play a significant role in that boost to inflation expectations and bond yields that we expect next year. The decline in the greenback seen in the latter half of 2020 has been driven by the typical factors (Chart 14): Chart 14More Negatives Than Positives For The USD The Fed’s aggressive rate cuts, dating back to 2019, have reduced much of the relative interest rate attractiveness of the US dollar Accelerating global growth after the sharp worldwide plunge in growth in Q2/2020 benefitted non-US economies more, eliciting a standard decline in the “anti-growth” US dollar Uncertainty and risk aversion declined after the initial COVID-19 shock at the start of 2020, easing the safe haven demand for dollars. Looking ahead, rate differentials continue to point to additional downward pressure on the US dollar, even with the moderate rise in longer-term US Treasury yields that we expect next year. Risk aversion and uncertainty should also decline in a dollar-bearish fashion with the US presidential election behind us and the COVID-19 vaccine ahead of us. Improving global growth should also be supportive of more dollar weakness, especially as Europe recovers from the current lockdown-driven slowdown. A weaker US dollar is a key variable to trigger faster global inflation through the link between the currency and global traded goods prices. On a rate-of-change basis, a weakening US dollar has a strong negative correlation to the growth rate of world export prices and commodity prices (Chart 15). Thus, more USD weakness in 2021 will lift realized global inflation through commodities and traded goods prices, especially against a backdrop of faster global growth. Chart 15Global Reflation Through A Weaker USD Chart 16Stay Overweight Global Inflation-Linked Bonds In 2021 BCA Research’s commodity strategists expect oil prices to move higher next year on the back of an improving demand/supply balance, with the benchmark Brent price of oil averaging $63/bbl over the course of 2021. A weaker USD could provide additional upside to that forecast, giving a further lift to realized inflation rates around the world. To position for this boost to inflation via a weaker dollar and rising commodity prices, we recommend that fixed-income investors continue holding a core allocation to inflation-linked bonds versus nominal government debt. We have maintained that recommendation since last spring after the collapse of global breakeven inflation rates that left breakevens very undervalued according to our fair value models (Chart 16).2 The valuation case is far less compelling now after the steady climb in breakevens over the latter half of 2020, with only French and Japan breakevens below fair value. However, given our expected backdrop of improving global growth and highly accommodative global monetary policy, breakevens are likely to continue to climb to more expensive levels. Our preferred allocations are to US and French inflation-linked bonds, while we would be cautious on Australian inflation-linked bonds which appear extremely overvalued on our models. Key View #4: Within an overweight allocation to global corporate debt, overweight US high-yield versus US investment grade and favor all US corporates versus euro area equivalents. Global corporate bond markets have enjoyed a spectacular rally over the final three quarters of 2020 after the huge pandemic related selloff of last February and March. The benchmark index yields for investment grade corporates in the US, euro area and UK have all fallen back below pre-COVID levels, while index yields for high-yield in the same three regions are back at the pre-COVID lows (Chart 17). The story is similar on a credit spread basis. The benchmark index option-adjusted spread (OAS) for investment grade corporates is only 11bps away from the pre-COVID low in the US and 4bps from the pre-COVID low in the euro area, with the UK spread now slightly below the pre-pandemic low (Chart 18). High-yield spreads still have some more room to compress with US, euro area and UK junk index spreads 67bps, 68bps and 110bps above the pre-pandemic low, respectively. Chart 17Corporate Bond Yields Falling To New Lows Chart 18Corporate Bond Spreads Approaching Pre-COVID Lows Supportive monetary policy has played a huge role in the global credit rally. Central banks have used their balance sheets aggressively to help ease financial conditions, including the direct buying of corporate bonds by the Fed, ECB and BoE. Looking ahead to 2021, it is clear that credit markets are still benefitting from loose monetary policy while also enjoying a tailwind from better global growth. The global high-yield default rate is rolling over and the US default rate has clearly peaked (Chart 19). There is now less of a need for direct buying of corporates by central banks with credit markets seeing major investor inflows with a robust pace of corporate bond issuance. Corporate bond markets can now walk on their own with the support of central bank crutches. This means that investors should pivot away from the more cautious “buy what the central banks are buying” approach that we had advocated for much of 2020 and be more selectively aggressive. First and foremost, that means increasing allocations to US high-yield corporate debt, both out of US investment grade and euro area corporates. Default-adjusted spreads in the US, which measure the high-yield index OAS net of realized default losses, will look far more attractive as the US default rate peaks (Chart 20). If the US default rate moves back below 5% over the next year from the current 8% rate, the US default-adjusted spread will climb back into positive territory. This will compare more favorably to the default-adjusted spread for euro area high-yield, which has been higher because the euro area default rate did not suffer a major spike this year despite the sharp downturn in euro area growth back in the spring. Chart 19Easy Money Policies Supporting Global Credit Chart 20High-Yield Looks More Attractive With Fewer Defaults In 2021 US high-yield also looks most attractive using our preferred metric of pure spread valuation, the 12-month breakeven spread. This measures the amount of spread widening that must occur over a one year period for corporate debt to have the same return as a duration-matched position in government bonds. We compare this “spread cushion” to its own history in a percentile ranking to determine if spreads look relatively attractive. Within US corporate debt, the 12-month breakeven spread for investment grade credit is down to the 5th percentile, suggesting virtually no room for additional spread tightening (Chart 21). For US high-yield credit, the 12-month breakeven spread is still relatively elevated at the 60th percentile level, suggesting more room for spread compression. Within euro area corporates, the 12-month breakeven percentile rankings for investment grade and high-yield are at the 27th and 28th percentile, respectively, suggesting a more limited scope for spread compression compared to US high-yield (Chart 22). Chart 21Move Down In Quality Within US Corporates Chart 22No Compelling Value In Euro Area Corporates When comparing the 12-month breakeven spreads of all corporate debt in the US, euro area and UK, broken down by credit tier, to a more pure measure of spread risk - duration times spread – the attractiveness of lower-rated US junk bonds is most compelling (Chart 23). In particular, US B-rated and Caa-rated junk spreads offer very high 12-month breakeven spreads relative to spread risk. Chart 23Comparing Value (Breakeven Spreads) With Risk (Duration Times Spread) Adding it all up, it is clear that lower-rated US high-yield debt offers an attractive value proposition for 2021. This is especially true given the positive global growth and monetary policy backdrop. The annual growth rate of the combined balance sheets of the Fed, ECB, BoE and Bank of Japan has been an excellent leading indicator of the excess return of US high-yield US Treasuries (Chart 24). The surge in balance sheet growth of 2020 is pointing to strong US high-yield bond performance versus Treasuries, and an outperformance of lower-rated US high-yield, in 2021. Chart 24Upgrade US High-Yield To Overweight Chart 25Within EM USD Credit, Favor Corporates Over Sovereigns This leads us to shift to an overweight stance on US high-yield, while downgrading US investment grade to neutral, as our key global spread product recommendation for 2020. Within other corporate credit markets, we recommend only a neutral allocation to euro area corporate credit, given the relatively less attractive valuations. Finally, within the emerging market US dollar denominated universe, we continue to recommend an overweight stance on corporates versus sovereigns, as the former will benefit more in 2021 from the lagged effect of Chinese credit stimulus and central bank balance sheet expansion in 2020 (Chart 25).   Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com Footnotes 1 Please see BCA Research The Bank Credit Analyst, "Outlook 2021: A Brave New World", dated November 30, 2020, available at bca.bcaresearch.com. 2 Our breakeven inflation models use the growth rate of oil prices in local currency terms and a long-term moving average of realized inflation as the inputs. Recommendations Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

Special Report

Highlights Below-Benchmark Portfolio Duration: The economic recovery will continue (and even accelerate) in 2021. Meanwhile, the Fed’s forward interest rate guidance is already as dovish as it will get. Keep portfolio duration below-benchmark in 2021, targeting a level of 1.25% to 1.5% for the 10-year Treasury yield.  Overweight TIPS Versus Nominal Treasuries: We remain overweight TIPS versus nominal Treasuries for the time being but are actively looking for an opportunity to get tactically underweight. This opportunity could emerge in the first half of 2021 when core and trimmed mean inflation re-converge and when the 10-year TIPS breakeven inflation rate looks expensive on our model.  Own Nominal Yield Curve Steepeners, Real Yield Curve Steepeners And Inflation Curve Flatteners: The nominal yield curve will continue to trade directionally with yields. Therefore, higher yields will coincide with a steeper nominal curve in 2021. Rising inflation and the Fed’s new Average Inflation Target both argue for a flatter inflation curve in 2021. We also recommend a real yield curve steepener as a high octane play on both a steeper nominal curve and flatter inflation curve. Overweight Spread Product Versus Treasuries: We see the economy as entering what we call “Phase 1” of the economic cycle in 2021, an environment of above-trend growth, low inflation and accommodative monetary policy. This is an environment where spread product typically performs very well relative to Treasuries. Move Down In Quality Within Corporates: Investment grade corporates will outperform Treasuries in 2021, but the potential for further spread compression is limited. Junk spreads have more room to tighten, and the Ba credit tier looks particularly attractive from a risk/reward perspective A Maximum Overweight Allocation To Municipal Bonds: Tax-exempt municipal bonds offer the best opportunity in the US fixed income space. Investors should adopt a maximum overweight allocation, and in particular, they should shift some allocation out of investment grade corporates and into Munis with the same credit rating and duration, but with a greater after-tax yield. Feature BCA published its 2021 Outlook on November 30. That report lays out the main macroeconomic themes that our strategists see driving markets next year. This Special Report explains how investors can profit from those themes in US fixed income markets. Specifically, we offer six key US fixed income views for 2021. This report is limited to the six key investment views listed on page 1, and only discusses Fed policy in the context of how it influences those views. Next week we will publish a more comprehensive “Fed In 2021” report that will delve into our outlook for the Fed next year. Outlook Summary First, a brief summary of the main economic views presented in BCA’s 2021 Outlook:1 The third wave of COVID infections will be a drag on economic activity in 2020 Q4 and 2021 Q1, but inventory re-stocking and the large build-up of household savings will prevent the US economy from falling into a double-dip recession. Ultimately, the vaccine roll-out will cause US GDP to grow well above trend in 2021. Inflation is likely to spike in the first half of 2021 due to base effects and the re-opening of some service sectors that were shuttered during the pandemic. But this initial surge will dissipate in the second half of the year. The wide output gap that opened in 2020 will persist in 2021 and will prevent a broad-based acceleration in consumer prices. The Fed’s forward interest rate guidance is as dovish as it will get. A large portion of the Outlook is devoted to considering longer-run economic and political trends that were accelerated by the global policy response to COVID-19. Specifically, rising populism, heavier corporate regulation and a greater appetite for MMT-like taxing and spending policies. The ultimate outcome of these trends will be significantly higher inflation, on the order of 3% to 5%, in the second half of the decade. Key View #1: Below-Benchmark Portfolio Duration Chart 1Treasury Yields In 2020 The economic recovery will continue (and even accelerate) in 2021. Meanwhile, the Fed’s forward interest rate guidance is already as dovish as it will get. Keep portfolio duration below-benchmark in 2021, targeting a level of 1.25% to 1.5% for the 10-year Treasury yield. Our recommendation to maintain below-benchmark portfolio duration rests on two key pillars. The first is BCA’s view that the economic recovery will continue in 2021 and will even accelerate once enough of the population has received the COVID vaccine. The second pillar is our view that the Federal Reserve’s reaction function is as dovish as it will get. In other words, having already laid out the conditions that must be in place for it to begin the next rate hike cycle, the Fed will not undertake further efforts to guide interest rates lower in the face of economic recovery. Chart 1 provides a bit more context for our assessment of Fed policy. This year, economic growth and inflation expectations troughed in March and moved rapidly higher throughout the summer. Bond yields, however, stayed relatively flat between March and August. The reason is that, even as the economic outlook improved, the Fed was steadily guiding markets towards a dramatic shift in its forward interest rate guidance. Specifically, the adoption of an Average Inflation Target – a pledge to allow a moderate overshoot of the 2% inflation target to make up for past downside misses. The result of the Fed’s dovish shift is that the increase in inflation expectations between March and August was entirely offset by falling real yields (Chart 1, panel 3), leaving nominal yields close to unchanged. However, the Fed made its Average Inflation Target official at the Jackson Hole Symposium in August. Then, in September, it formalized its forward rate guidance by promising not to lift rates off the zero bound until inflation reaches 2% and is expected to moderately overshoot for a while. These events changed the dynamic in the bond market. The Fed is no longer trying to guide markets towards a more dovish reaction function. That reaction function is now officially in place, and presumably in the market price. Indeed, nominal bond yields have risen in concert with improving economic conditions since August, and we expect that trend to continue in 2021. Our Golden Rule of Bond Investing states that we should set portfolio duration by considering our own expectations for future changes in the fed funds rate relative to what is already priced in the yield curve. Appendix A at the end of this report shows that the Golden Rule once again performed well in 2020. Looking ahead, the market is currently pricing-in one full 25 basis point rate hike by mid-2023 and then only one more by mid-2024 (Chart 2). We see high odds that inflation could sustainably reach 2% – the Fed’s stated criteria for lifting off the zero bound – before that, necessitating some Fed tightening in 2022. Chart 2Market Priced For Liftoff In 2023 How High Could Yields Go In 2021? To answer this question, we first look at the 5-year/5-year forward Treasury yield relative to survey estimates of the longer-run equilibrium fed funds rate. In theory, long-dated forward yields should be relatively insulated from near-term shifts in the policy rate and should settle near levels consistent with estimates of the equilibrium fed funds rate. In practice, we find that the 5-year/5-year forward Treasury yield does settle near these levels, but only during periods of global economic recovery when investors are presumably more inclined to envision the closing of the output gap and an eventual neutralizing of monetary policy. Notice that during the past two global growth upturns, 2013/14 and 2017/18, the 5-year/5-year forward Treasury yield peaked close to survey estimates of the long-run equilibrium fed funds rate from the New York Fed’s Survey of Market Participants and the Survey of Primary Dealers (Chart 3A). If the same thing happens next year, the 5-year/5-year forward Treasury yield will rise to a range of roughly 2% to 2.25%, 54 bps to 79 bps above current levels. Chart 3AHow High Can Yields Rise? Chart 3BLess Upside In 10y Than In 5y5y We see less upside next year for the benchmark 10-year yield than for the 5-year/5-year forward. Long-dated forward rates are not mathematically influenced by the near-term outlook for the policy rate, but the yield on the 10-year Treasury note embeds those expectations. Since it is unlikely that inflation will be strong enough to prompt a Fed rate hike in 2021, the yield curve will steepen as the economic outlook improves and the 10-year yield will rise by less than the 5-year/5-year forward. Looking at Chart 3B, next year’s bond market moves will look a lot more like 2013/14 than like 2017/18. The Fed kept rates at zero in 2013/14. This led to yield curve steepening and caused the 10-year Treasury yield to peak at a level well below survey estimates of the long-run equilibrium fed funds rate. In contrast, the Fed was hiking rates in 2017/18. This led to a flatter yield curve and caused the 10-year yield to peak at around the same level as the 5-year/5-year forward. All in all, while we could see the 5-year/5-year forward Treasury yield reach a range of 2% to 2.25% next year, we expect the 10-year Treasury yield to reach a range of 1.25% to 1.5%. Will The Fed Use Its Balance Sheet To Stop Treasury Yields From Rising? By far, the most common disagreement we’ve received from clients on our call for higher bond yields is that the Fed will simply use its balance sheet to prevent any increase in long-maturity yields. We don’t see this as having a meaningful impact. For one, the Fed will only take significant steps to ease monetary policy if it looks like the economic recovery is under threat. This would require a large tightening of financial conditions, meaning significantly lower stock prices and wider corporate bond spreads. We don’t see a 1.25% to 1.5% 10-year Treasury yield in the context of a steepening yield curve, low inflation and improving economic growth as likely to cause such an event. Granted, the Fed could take more minor actions, like keeping the same pace of purchases but shifting them further out the curve, but a significant tightening of financial conditions is likely required for them to increase the monthly pace of bond buying. Second, even if the Fed does decide to ramp up the pace of bond buying (either overall or only at the long-end of the curve), if it keeps the same forward interest rate guidance, then bond yields will be driven by the market’s perceived progress toward the conditions that would prompt the start of the next tightening cycle. It won’t matter how many bonds the Fed buys in the meantime. Our Golden Rule of Bond Investing has a strong track record that it achieves by focusing only on changes in the fed funds rate relative to expectations. It does not consider asset purchases at all, and we are also inclined to view them more as a distraction. Key View #2: Overweight TIPS Versus Nominal Treasuries Chart 4Adaptive Expectations Model We remain overweight TIPS versus nominal Treasuries for the time being but are actively looking for an opportunity to get tactically underweight. This opportunity could emerge in the first half of 2021 when core and trimmed mean inflation re-converge and when the 10-year TIPS breakeven inflation rate looks expensive on our model. TIPS breakeven inflation rates fell dramatically when the COVID crisis struck in March, but they then rebounded just as quickly and are now near fair value according to our Adaptive Expectations Model (Chart 4). Our model forecasts the future 12-month change in the 10-year TIPS breakeven inflation rate based on where the rate currently sits relative to several different measures of actual CPI inflation. Right now, our model is looking for a 12 basis point decline in the 10-year breakeven rate during the next year, but this forecast will rise if CPI prints strongly in the coming months, which is exactly what we expect. Chart 5Expect Higher Inflation In H1 2021 As noted in the above Outlook Summary, base effects and the re-opening of some service sectors will cause inflation to jump in the first half of 2021. A good way to see this is to look at the gap between 12-month core and trimmed mean CPI (Chart 5). Core inflation fell dramatically in March and April and is now in the process of bouncing back. Meanwhile, trimmed mean inflation measures were much more stable in the spring because they filtered out those sectors that experienced huge negative inflation prints during quarantine.   We think the gap between core and trimmed mean CPI is a good guidepost for our TIPS strategy. As long as the gap remains wide, we see upside risks to inflation. However, once the gap closes, that will signal that the “snapback phase” from re-opening the economy is over and that inflation pressures will moderate in line with the wide output gap. Shelter inflation is one of the components of inflation that is most sensitive to the output gap, and it has already been rolling over in line with the rising unemployment rate (Chart 5, bottom panel). Overall, our TIPS strategy in 2021 is to remain overweight TIPS versus nominal Treasuries for the time being. However, we are actively looking for an opportunity to get tactically short TIPS versus nominals. This could occur sometime in the first half of 2021 when core and trimmed mean inflation have re-converged and when (hopefully) the 10-year TIPS breakeven inflation rate looks more expensive on our model. Key View #3: Own Nominal Yield Curve Steepeners, Real Yield Curve Steepeners and Inflation Curve Flatteners Chart 62/5/10 Butterfly Spread Valuation The nominal yield curve will continue to trade directionally with yields. Therefore, higher yields will coincide with a steeper nominal curve in 2021. Rising inflation and the Fed’s new Average Inflation Target both argue for a flatter inflation curve in 2021. We also recommend a real yield curve steepener as a high octane play on both a steeper nominal curve and flatter inflation curve. Nominal Yield Curve With the funds rate pinned at zero and the Fed unlikely to actually lift it until 2022 (at the earliest), it is quite clear that the slope of the nominal yield curve will continue to trade directionally with yields as we head into 2021. That is, with volatility at the front-end of the curve completely suppressed, the yield curve will steepen when yields rise and flatten when they fall. In that context, we recommend complementing our below-benchmark portfolio duration view with nominal yield curve steepeners. Our preferred way to implement a nominal yield curve steepener is to buy the 5-year Treasury note and short a barbell consisting of the 2-year note and 10-year note. Allocations to the 2-year and 10-year should be weighted so that the duration of the 2/10 barbell matches that of the 5-year note. As we have explained in prior research, this sort of position is designed to profit from 2/10 yield curve steepening and it has worked well during the past few months (Chart 6).2  The one problem with this 5 over 2/10 trade is that it is not cheap. The 5-year yield is below the yield on the 2/10 barbell (Chart 6, panel 3) and the 5-year bullet looks expensive on our fair value model (Chart 6, bottom panel). However, we should also note that the 5-year looked much expensive during the last period of zero-bound rates in 2012. Given today’s very similar policy environment, we could see the 5-year yield getting even more expensive in 2021. Inflation Curve Chart 7Favor Inflation Curve Flatteners... Our second recommended yield curve position relates to the inflation curve, either the TIPS breakeven inflation curve or the CPI swap curve. Here, we recommend owning inflation curve flatteners for two reasons. First, short-maturity inflation expectations are more sensitive to the actual inflation data than long-maturity expectations. We saw a prime example of this relationship in 2020. The 2-year CPI swap rate plunged into negative territory when inflation fell in March while the 10-year CPI swap rate held relatively stable in comparison (Chart 7). Subsequently, the 2-year CPI swap rate rose much more quickly than the 10-year rate this summer as inflation rebounded. Looking ahead, with inflation biased higher in the first half of 2021, we should see greater upside in short-maturity inflation expectations than in long-maturity ones. The Fed’s adoption of an Average Inflation Target is the second reason to favor inflation curve flatteners. If the Fed is ultimately successful at achieving an overshoot of its 2% inflation target, it will mean that the Fed will be attacking its inflation target from above rather than from below for the first time since the 1980s. Logically, the inflation curve should be inverted in this sort of environment. This means that the inflation curve still has a lot of room to flatten from current levels (Chart 7, bottom panel). Real Yield Curve Chart 8...And Real Yield Curve Steepeners The Fisher Equation tells us that real yields are simply the difference between nominal yields and inflation expectations. Viewed that way, it is easy to see that – all else equal – a steeper nominal curve will lead to a steeper real yield curve. Meanwhile, a flatter inflation curve will also lead to a steeper real yield curve. In that sense, a real yield curve steepener is just a combination of the nominal curve steepener and inflation curve flattener that we already mentioned (Chart 8). As inflation rises, it will pressure short-dated inflation expectations higher relative to long-dated ones. This will exert bull-steepening pressure on the real yield curve. Meanwhile, investors starting to price-in eventual rate hikes will lead to nominal yield curve steepening. This will exert bear-steepening pressure on the real yield curve. With that in mind, a real yield curve steepener is a high conviction position for us in 2021. We have less conviction on the outright direction for real yields, though we suspect that long-maturity real yields have already troughed for the cycle. Key View #4: Overweight Spread Product Versus Treasuries We see the economy as entering what we call “Phase 1” of the economic cycle in 2021, an environment of above-trend growth, low inflation and accommodative monetary policy. This is an environment where spread product typically performs very well relative to Treasuries.  Most spread sectors will likely end the year having underperformed duration-equivalent Treasuries in 2020. However, this simple fact obscures the actual pattern of spread movements that was witnessed during the year. Spreads widened sharply when COVID struck but they peaked on March 23, the same day that the Federal Reserve announced its slew of emergency lending facilities.3 Spread product has been outperforming Treasuries since then (see Appendix B), a trend we expect will continue in 2021. The phase of the economic cycle when the economy is just emerging from a recession is typically one where risk assets perform well. The principal reason to expect spread product outperformance to continue is that the phase of the economic cycle when the economy is just emerging from a recession is typically one where risk assets perform well. It tends to be an environment where economic activity is growing at an above-trend pace, but inflation is still low and monetary conditions are accommodative. This is the perfect environment for credit spreads to tighten. The slope of the yield curve is a useful variable for summarizing the above macro conditions and we often use it to define three phases of the economic cycle (Chart 9): Chart 9The Three Phases Of The Cycle Phase 1 is defined as the time between the end of the last recession and when the 3-year/10-year Treasury slope flattens to below 50 bps. Phase 2 is defined as when the 3-year/10-year Treasury slope is between 0 bps and 50 bps. Phase 3 is defined as the time between when the 3-year/10-year Treasury slope turns negative and the start of the next recession. As we are just now emerging from recession and the 3-year/10-year slope is above 50 bps and steepening, we see the economy as being firmly in Phase 1 of the cycle. Historically, this phase has been the best one for spread product returns relative to duration-matched Treasuries (Table 1). Table 1Corporate Bond Performance In Different Phases Of The Cycle The main risk to this view of spread product is that we are not yet emerging from the recession and the corporate default rate may have another leg higher. Our sense, however, is that the default rate has already peaked. Gross leverage (the ratio between total corporate debt and pre-tax corporate profits) and job cut announcements are two variables that correlate very tightly with the default rate (Chart 10). Starting with leverage, net earnings revisions – a leader profit indicator – have already troughed and the corporate financing gap has turned negative (Chart 11). A negative financing gap means that the corporate sector has sufficient retained earnings to cover its capital expenditures. In other words, most firms are flush with cash and they won’t need to issue more debt in the coming quarters. Further, job cut announcements have come down sharply during the past few months (Chart 11, bottom panel). Chart 10The Default Rate Correlates With Gross Leverage And Job Cuts Chart 11Firms Have Enough Cash The above trends in corporate profits, corporate debt and job cut announcements are consistent with what we’re already seeing on the default front. The US corporate sector was experiencing upwards of 20 default events per month back in May, June and July. But only seven defaults occurred in November, following five in October and six in September (Chart 12). Chart 12The Default Rate Has Peaked The bottom line is that the macro environment of above-trend growth, low inflation and accommodative monetary policy is one where we should expect spread product to outperform Treasuries. Relative valuation dictates which spread sectors we prefer over other ones, and the next two Key Views address this issue. Key View #5: Move Down In Quality Within Corporates Investment grade corporates will outperform Treasuries in 2021, but the potential for further spread compression is limited. Junk spreads have more room to tighten, and the Ba credit tier looks particularly attractive from a risk/reward perspective. As noted in the previous section, the macroeconomic environment is one where spread product should flourish. However, valuation in certain sectors could limit how much further spread tightening is possible. In particular, valuation looks to be a constraint for investment grade corporates. In absolute terms, investment grade corporate spreads look like they still have some room to compress (Chart 13). The overall index spread is 12 bps above its pre-COVID level. The Aa, A and Baa-rated spreads are 16 bps, 11 bps and 13 bps above, respectively. Only seven defaults occurred in November, following five in October and six in September. However, valuation looks much worse in risk-adjusted terms. Chart 14 shows the 12-month breakeven spread, i.e. the spread widening required for the sector to underperform Treasuries on a 12-month investment horizon. In addition, we re-weight the overall corporate index to ensure that it maintains a constant credit rating distribution over time, and we show all breakeven spreads as percentile ranks relative to their own histories. For example, a reading of 8% for the Baa credit tier means that the 12-month breakeven spread for the Baa credit tier has only been lower than it is today 8% of the time since our data begin in 1995. Chart 13IG Spreads Still Above ##br##Pre-COVID levels Chart 14IG Looks More Expensive In Risk-Adjusted Terms Adding it all up, we think there is scope for investment grade corporates to modestly outperform Treasuries in 2021, but there are also more attractively priced sectors that investors may want to consider. Municipal bonds are one particularly attractive alternative to investment grade corporates (we discuss our view on municipal bonds in the next section), but investors are also advised to pick-up additional spread by moving down in quality within the corporate credit space. High-Yield corporate bonds have significantly more scope for tightening than their investment grade counterparts, with the overall junk index spread still 69 bps above its pre-COVID level (Chart 15). Within junk, the Ba credit tier looks like the best place to camp out from a risk/reward perspective. The incremental spread offered by Ba-rated junk bonds compared to Baa-rated corporates is elevated compared to history, 111 bps above its 2019 low (Chart 15, panel 2). In contrast, the additional spread pick-up you get from moving into the lower junk tiers (B & Caa) is more in line with typical historical levels (Chart 15, bottom 2 panels). Chart 15Ba-Rated Bonds Look Best Another reason to be cautious about chasing the extra spread in the B-rated and below credit tiers is that the High-Yield index is pricing-in a fairly rapid decline in the default rate for the next 12 months (Chart 16). If we assume a 25% recovery rate and target an excess spread of 150 bps above default losses,4 then we calculate a spread-implied default rate of 3.1%. That is, we should only expect junk bonds to outperform duration-matched Treasuries if the default rate comes in below 3.1% during the next 12 months. This would represent a steep decline of 5.3% from the 8.4% default rate we just witnessed during the past 12 months, but this sort of big drop in the default rate would not be out of line with what typically happens when the economy emerges from recession. For example, in the last recession, the 12-month default rate peaked at 14.6% in November 2009 and then fell to 3.6% by November 2010, a decline of 11%! Chart 16Spread-Implied Default Rate All in all, we view the Ba-rated credit tier as the sweet spot within corporate credit in terms of offering the best combination of risk and reward. We also expect the default rate to fall quickly enough that the lower-rated junk credit tiers will outperform Treasuries, but the risk here is greater and the potential additional compensation is not historically elevated. Investment grade corporate spreads will remain tight, but have limited room to compress further. Investors are advised to look at Ba-rated corporates and municipal bonds instead.  Key View #6: A Maximum Overweight Allocation To Municipal Bonds Tax-exempt municipal bonds offer the best opportunity in the US fixed income space. Investors should adopt a maximum overweight allocation, and in particular, they should shift some allocation out of investment grade corporates and into Munis with the same credit rating and duration, but with a greater after-tax yield. At present, we think that tax-exempt municipal bonds represent the best opportunity in US fixed income. Muni spreads have certainly tightened since March, but valuation remains attractive relative to both Treasuries and investment grade corporates. First, let’s consider value relative to Treasuries (Chart 17). Spreads between Aaa-rated municipal bonds and maturity-matched Treasuries are elevated compared to history across the entire yield curve. 2-year Munis even offer a 3 bps yield pick-up over 2-year Treasuries before adjusting for the tax advantage. Further out the curve, value is worst at the 5-year part of the curve where the breakeven effective tax rate between Munis and Treasuries is 42%, slightly above the top marginal tax rate of 37%. But value improves again for longer maturities. The breakeven effective tax rate between 10-year Munis and Treasuries is 24% and it is a mere 10% for 30-year bonds.5 Next, we can look at relative value between Munis and credit. This is where the attractiveness of munis really stands out (Chart 18). After controlling for credit rating and duration, municipal revenue bonds offer a yield advantage over the Bloomberg Barclays Credit Index across the entire yield curve, before any adjustment is made for the municipal tax exemption. General Obligation (GO) Munis only offer a before-tax yield advantage over credit beyond the 12-year maturity point, but the GO Muni/credit spread is nonetheless historically elevated for all maturity buckets. Chart 17Muni/Treasury Yield Spreads Chart 18Munis Versus Credit This is all well and good, but it could easily be countered that municipal bonds only offer such attractive valuations because the COVID recession has been an historically challenging period for state & local government balance sheets. If this period leads to a spate of downgrades and defaults, then municipal bonds no longer look cheap. All this is true, but we think investors’ worst fears in this regard will not be realized. For one thing, state & local governments have been very quick to clamp down on spending and cut employment (Chart 19). Coming out of the last recession, Muni/Treasury yield spreads had almost fully recovered by the time that state & local government austerity began. Also, state budgets were in pretty good shape heading into the COVID downturn, with all-time high Rainy Day Fund balances (Chart 19, bottom panel). Chart 19State & Local Austerity Has Begun We recommend that investors take advantage of historically attractive municipal bond spreads by adopting a maximum overweight allocation. In particular, investors should shift allocation out of investment grade rated corporate bonds, where valuations are stretched, and into municipal bonds that offer the same credit rating and duration with a greater yield pick-up. Finally, Chart 20 shows the spread between different municipal bond sectors and the Bloomberg Barclays US Credit Index. We match the credit rating and duration in each case, but we make no adjustments for the municipal tax exemption. The message from Chart 20 is that the yield advantage in investment grade Munis is broad based, with the exception of the Electric sector. We also see that attractive valuations do not extend to high-yield Munis, which appear expensive relative to High-Yield Credit. Chart 20Municipal Bond Sector Valuation Appendix A:  The Golden Rule Of Bond Investing Our Golden Rule of Bond Investing says that we should determine what change in the fed funds rate is priced into the overnight index swap curve for the next 12 months, and then decide whether the Fed will deliver a hawkish or dovish surprise relative to that expectation. We contend that if the Fed delivers a hawkish surprise, then a below-benchmark portfolio duration positioning will pay off. Conversely, if the Fed delivers a dovish surprise, then an above-benchmark portfolio duration positioning will profit. Chart A1 shows how the Golden Rule has performed in every calendar year going back to 1990. We include year-to-date performance for 2020. In 31 years of historical data, our Golden Rule performed well in 23. It provided the wrong recommendation in 8 years, though 3 of those years were during the zero-lower-bound period between 2009 and 2015 when 12-month rate expectations were essentially pinned at zero.6 Chart A1The Golden Rule's Track Record At the beginning of this year, the market was priced for 13 bps of rate cuts in 2020. The funds rate actually fell by 146 bps, leading to a dovish surprise of 133 bps. Based on a historical regression, we would expect a dovish surprise of 133 bps to coincide with a Treasury index yield that falls by 81 bps. In actuality, the index yield fell by 122 bps, more than our Golden Rule predicted. Chart A2 shows how close changes in the Treasury index yield have been to our Golden Rule’s prediction in each of the past 31 years. This regression between the change in Treasury index yield and the monetary policy surprise is the main source of error in our Treasury return forecasts. Chart A2Treasury Index Yield Changes Versus Fed Funds Surprises Based on our expected -81 bps index yield change, we would have expected the Treasury index to deliver 6.5% of total return in 2020 and to outperform cash by 5.5%. In actuality, the index earned 7.9% of total return and outperformed cash by 7%. Charts A3 and A4 show how index total and excess returns have performed relative to our Golden Rule’s expectations in each of the past 31 years. Chart A3Treasury Index Total Returns Versus The Golden Rule’s Predictions Chart A4Treasury Index Excess Returns Versus The Golden Rule’s Predictions Appendix B: Spread Product Performance In 2020 Table B1Spread Product Year-To-Date Performance Table B2Performance Since March 23 Announcement Of Emergency Fed Facilities Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see The Bank Credit Analyst, “Outlook 2021: A Brave New World”, dated November 30, 2020, available at bca.bcaresearch.com 2 Please see US Bond Strategy Special Report, “Bullets, Barbells And Butterflies”, dated July 25, 2017, available at usbs.bcaresearch.com 3 We discussed these facilities in detail in two Special Reports published jointly this year with our US Investment Strategy team. US Investment Strategy / US Bond Strategy Special Report, “Alphabet Soup: A Summary Of The Fed’s Anti-Virus Measures”, dated April 14, 2020 and US Investment Strategy / US Bond Strategy Special Report, “Alphabet Soup Part 2: Shocked And Awed”, dated July 28, 2020. Both reports available at usbs.bcaresearch.com 4 Our research has shown that this is the minimum excess spread investors should require to be confident that junk bonds will outperform duration-matched Treasuries. For more details please see US Bond Strategy Weekly Report, “No Holding Back”, dated June 16, 2020, available at usbs.bcaresearch.com 5 The breakeven effective tax rate is the effective tax rate that makes the after-tax muni yield the same as the Treasury yield. If the investor’s personal tax rate is above the breakeven effective tax rate, they will get an after-tax yield pick-up from owning the municipal bond over the Treasury. 6 We say the Golden Rule “worked” if a dovish surprise coincided with positive Treasury index excess returns versus cash, or if a hawkish surprise coincided with negative Treasury excess returns versus cash. Recommended Portfolio Specification

Dear Client, We are sending you our Strategy Outlook today, where we outline our thoughts on the macro landscape and the direction of financial markets for 2021 and beyond. Next week, please join me for a webcast on Thursday, December 17 at 10:00 AM EST (3:00 PM GMT, 4:00 PM CET, 11:00 PM HKT) where I will discuss the outlook. Our publishing schedule will resume early next year. On behalf of the entire Global Investment Strategy team, I would like to wish you a Merry Christmas, Happy Holidays, and a Healthy New Year! Best regards, Peter Berezin, Chief Global Strategist Highlights Macroeconomic outlook: The global economy will strengthen in 2021 as the pandemic winds down. Inflation will remain well contained for the next 2-to-3 years before moving sharply higher by the middle of the decade. Global asset allocation: Stocks are technically overbought and vulnerable to a short-term correction. Nevertheless, investors should favor equities over bonds in 2021 given the likelihood that earnings will accelerate while monetary policy stays accommodative. Equities: This year’s losers will be next year’s winners. In 2021, international stocks will outperform US stocks, small caps will outperform large caps, banks will outperform tech, and value stocks will outperform growth stocks. Fixed income: Bond yields will rise modestly next year, implying that investors should maintain below average duration exposure. Spread product will outperform safe government bonds. Favor inflation-protected securities over nominal bonds. Currencies: The US dollar will continue to weaken in 2021. The collapse in US interest rate differentials versus its trading partners, stronger global growth, and a widening US trade deficit are all bearish for the greenback. Commodities: Tight supply conditions and a cyclical recovery in oil demand will support crude prices. Investors should favor gold over bitcoin as a hedge against long-term inflation risk. I. Macroeconomic Outlook V Is For Vaccine Chart 1Efficacy Rates Of Seasonal Flu Vaccines Are Well Below Those Of The Covid-19 Vaccines Ten months after the start of the pandemic, there is a light at the end of the tunnel. Both of the vaccines developed by Pfizer-BioNTech and Moderna using mRNA technology have demonstrated efficacy rates of around 95%. AstraZeneca’s vaccine, produced in collaboration with Oxford University, showed an efficacy rate of 90% in one of its clinical arms. Russia and China have also launched vaccines. The Russian vaccine, Gamaleya, displayed an efficacy rate of 91% based on 22,000 test participants. Such high efficacy rates are on par with the measles and smallpox vaccines, and well above the typical 30%-to-50% success rate for the seasonal flu vaccine (Chart 1). Inoculating most of the world’s population will not be easy. Nevertheless, large-scale vaccine production has already begun. More than half of the professional forecasters enrolled in the Good Judgement Project expect enough doses to be available to vaccinate 200 million Americans (about 60% of the US population) by the end of the second quarter of 2021 (Chart 2). Chart 2Mass Distribution Of Covid-19 Vaccines Expected By Mid-2021 According to opinion polls, public concern about the potential side effects from the vaccines, while still high, has diminished over the past few weeks (Chart 3). Most countries will start by vaccinating health care workers and other at-risk groups. Assuming no major side effects are reported, the successful deployment of the vaccines among health care professionals should bolster confidence within the general public. Chart 3The Public Is Slowly Becoming Less Worried About Covid-19 Vaccines Vaccines And Growth: A Short-Term Paradox? There is no doubt that the availability of a safe and effective vaccine will bolster economic activity over the medium-to-long term. The short-term impact, however, is ambiguous. On the one hand, vaccine optimism could reduce household precautionary savings. It could also prompt more firms to invest in new capacity. On the other hand, the expectation that a vaccine is coming could motivate people to take even greater efforts to avoid getting sick in the interim. Think about what happens when you take cover under a tree after it starts to rain. Your decision to stay under the tree depends on how long you expect the rain to continue. If the rain will last for only 10 minutes, staying put makes sense. However, if it will rain continuously for the next two days, you are better off going home. You are going to get wet anyway. Who wants to get sick just as the pandemic is winding down? It is like being the last soldier killed on the battlefield. Growth In Europe Suffering More Than In The US… So Far The number of new daily cases has declined by 45% in the EU from the highs reached in the second week of November. That said, progress on the disease front has come at a cost. As Covid infections surged, European governments were forced to reimplement a variety of lockdown measures (Chart 4). Correspondingly, growth indicators have weakened across the region (Chart 5). At this point, it looks highly likely that GDP will contract in the euro area and the UK in the fourth quarter. Chart 4The Latest Viral Surge Led To Lockdowns In Europe   In contrast to Europe, the US economy should expand in the fourth quarter. The Atlanta Fed’s GDPNow model is pointing to growth of 11.2% in Q4, led by a recovery in personal consumption, strength in residential and nonresidential investment, and inventory restocking. Nevertheless, dark clouds are forming. After a short-lived dip in late November, the number of new daily cases in the US is on the rise again. The 7-day average of confirmed new cases has jumped to around 200,000. The Centers for Disease Control  (CDC) estimates that for every single case that is caught, seven go undiagnosed.1 This implies that over 11 million people are being infected each week, or about 3% of the US population. With the weather getting colder and the Christmas holiday season approaching, a further viral surge looks probable. Just as in Europe, we may see more lockdowns and more voluntary social distancing in the US over the coming weeks. Building A Fiscal Bridge To A Post-Pandemic World Lockdowns would be less of a problem if governments provided enough income support to struggling households and businesses. Unfortunately, at least in the US, considerable uncertainty remains about whether such support will be forthcoming. After a burst of stimulus earlier this year, US fiscal policy has tightened sharply. Since peaking in April, real disposable personal income has dropped by 9%, reflecting a steep decline in government transfer payments (Chart 6). The latest data suggest that real disposable income will be down in Q4 compared to the preceding quarter. Chart 5Renewed Lockdowns Are Weighing On Economic Activity In The Euro Area Chart 6Less Transfers Mean Less Income   President Trump tried to offset some of the sting from the expiration of emergency unemployment benefits in the CARES Act by diverting funds from the Federal Emergency Management Agency (FEMA) to support jobless workers. However, this money has now run out (Chart 7). Likewise, the resources in the Paycheck Protection Program for small businesses have been depleted, and many state and local governments are facing a cash crunch. Chart 7Drastic Drop In Unemployment Insurance Payments Chart 8People Are Eager For More Stimulus The US Congress has been squabbling over a new stimulus bill since May. Ultimately, we think a bill will be passed, potentially as part of a year-end omnibus spending package. Public opinion still very much favors maintaining stimulus. A survey conducted by Pew Research after the election found that about 80% of respondents supported passing a new stimulus package (Chart 8). Similarly, according to a recent NY Times/Siena College poll, 72% of voters supported a hypothetical $2 trillion stimulus package that would extend emergency unemployment insurance benefits, distribute direct cash payments to households, and provide financial support to state and local governments (Table 1). Such a package is basically what the Democrats are proposing. Strikingly, when this package is described in non-partisan terms, even the majority of Republicans are in favor of it. Table 1Even Republicans Want More Stimulus Peak Chinese Stimulus Even though it originated there, China has weathered the pandemic better than any other major country. Chinese export growth accelerated to 21.1% year-over-year in November. The Caixin manufacturing PMI rose to 54.9 on the month, the strongest reading since November 2010. The service sector PMI increased to a healthy 57.8. The “official” PMIs published by the National Bureau of Statistics also rose. Chinese growth will moderate over the coming months. The magnitude of China’s policy support has peaked, as evidenced by the rise in bond yields and interbank rates (Chart 9). The authorities have also permitted more corporate issuers to default, while tightening rules on online lending. Turning points in Chinese domestic demand and imports tend to lag policy developments by about 6-to-9 months (Chart 10). Thus, the tailwind from Chinese stimulus should fade by the middle of next year, hopefully just in time for the baton to be passed to a more organic, vaccine-driven global growth recovery. Chart 9China: Bond Yields And Interbank Rates Have Been Rising Chart 10Tailwind From Chinese Stimulus Will Fade By The Middle Of Next Year Japan: Near-Term Wobbles Japan is in the midst of its third wave of the pandemic. While not as bad as the latest waves in the US and Europe, it has still been disruptive enough to slow the economy. Although it did tick up in November, the manufacturing PMI remains below the crucial 50 boom/bust line, notably weaker than in other APAC countries. The outlook component of the Economy Watchers Survey fell to 36.5 in November (from 49.1), while the current situation component slid to 45.6 (from 54.5). Nevertheless, there are some encouraging signs. The number of new Covid cases seems to be stabilizing. Machine tool orders rose to 8% year-over-year in November, the first positive print since September 2018. Retail sales have recovered from a low of -14% year-over-year in April to around +6% in October. Broad money growth has reached a record high. The Japanese government is also considering a new ¥73 trillion fiscal stimulus package to fight the pandemic. Global Monetary Policy To Stay Accommodative Chart 11Service And Shelter Inflation Tend To Be Largely Determined By Labor Market Slack Could a vaccine-led economic recovery cause central banks to remove the punch bowl? We think not. Inflation is likely to rise in the first half of 2021 as the “base effects” from the pandemic-induced drop in prices reverse. However, central banks will see through these short-term oscillations in inflation. Inflation in modern economies is largely driven by services and shelter (goods account for only 25% of the US core CPI and 37% of the euro area core CPI). Both service inflation and shelter inflation tend to be largely determined by labor market slack (Chart 11). In its October 2020 World Economic Outlook, the IMF projected that the unemployment rate in the main developed economies would fall back to its full employment level by around 2025 (Chart 12). While this is too pessimistic in light of the subsequent progress that has been made on the vaccine front, it is probable that unemployment will remain too high to generate an overheated economy for the next 2-to-3 years. Chart 12Unemployment Rate Is Projected To Decline Towards Pre-Covid Lows In The Coming Years Chart 13Long-Term Inflation Expectations Are Still Subdued   Moreover, despite vaccine optimism, long-term inflation expectations are still below target in most of the major economies (Chart 13). Not only do central banks want inflation to return to target, they want inflation to overshoot their targets in order to make up for the shortfall in inflation in the post-GFC era. Had the core PCE deflator in the US risen by 2% per year since 2012, the price level would be about 3.3% higher than it currently is. In the euro area, the price level is about 9.5% below where it would have been if consumer prices had risen by 2% over this period. In Japan, the price level is 11.6% below target (Chart 14). Chart 14Central Banks Have Missed Their Inflation Targets II. Financial Markets A. Global Asset Allocation Remain Overweight Equities Versus Bonds On A 12-Month Horizon Equities have run up a lot since the start of November. Bullish sentiment has surged in the American Association of Individual Investors weekly bull-bear poll, while the put-to-call ratio has fallen to multi-year lows (Chart 15). This makes equities vulnerable to a short-term correction. Nevertheless, rising odds of an effective vaccine and continued easy monetary policy keep us bullish on stocks over a 12-month horizon. Stronger economic growth should lift earnings estimates. Stocks have usually outperformed bonds when growth has been on the upswing (Chart 16). Chart 15A Lot Of Bullishness Chart 16Stocks Rarely Underperform Bonds When The Global Economy Is Strengthening Valuations also favor stocks. As Chart 17 illustrates, the global equity risk premium – which we model by subtracting real bond yields from the cyclically-adjusted earnings yield – remains quite high. Along the same lines, dividend yields are above bond yields in the major markets. Even if one were to pessimistically assume that nominal dividend payments stay flat for the next 10 years, real equity prices would have to fall by 24% in the US for stocks to underperform bonds (Chart 18). In the euro area, real equity prices would need to tumble 32%. In Japan, they would have to drop 20%. Chart 17Equity Risk Premia Remain Elevated Chart 18Stocks Would Need To Fall A Lot For Equities To Underperform Bonds As such, investors should overweight global equities relative to bonds. We recommend a neutral allocation to cash to take advantage of any short-term dip in stock prices. Our full suite of asset allocation and trade recommendations are shown at the back of this report.   B. Equity Sectors, Regions, Styles This Year’s Losers Will Be Next Year’s Winners The “pandemic trade” is giving way to the “reopening trade.” We are still in the early innings of this transition. Hence, going into next year, it makes sense to favor stocks that were crushed by lockdown measures but could thrive once restrictions are lifted. Chart 19 shows relative 12-months forward earnings estimates for US/non-US, large caps/small caps, and tech/overall market. In all three cases, the tables have turned: Estimates are now rising more quickly for non-US stocks, small caps, and non-tech sectors. Non-US Stocks To Outperform Stocks outside the US are significantly cheaper than their US peers based on price-to-earnings, price-to-book, price-to-sales, and dividend yields (Chart 20). The macro outlook also favors non-US stocks, which tend to outperform when global growth is strengthening and the US dollar is weakening (Chart 21). Chart 19Relative Earnings Expectations For Non-US Stocks, Small Caps, And Non-Tech Are Improving   Chart 20Non-US Stocks Are Cheaper Chart 21Non-US Equities Tend To Outperform Their US Peers When Global Growth Is Improving And The Dollar Is Weakening As we discuss below, the dollar is likely to depreciate further over the next 12 months. A weaker dollar benefits cyclical sectors of the stock market more than defensives (Chart 22). Deep cyclicals are overrepresented outside the US (Table 2). Being more cyclical in nature, small caps usually outperform when the dollar weakens (Chart 23). Chart 22Cyclicals Tend To Outperform Defensives In A Falling Dollar Environment Table 2Financials Are Overrepresented In Ex-US Indices, While Tech Dominates The US Market Chart 23Small Caps Also Tend To Outperform When Global Growth Strengthens And The Dollar Weakens   Chart 24Banks’ Net Interest Margins Will Receive A Boost Buy The Banks Banks comprise a larger share of non-US stock markets. Stronger growth in 2021 will put upward pressure on long-term bond yields. Since short-term rates will stay where they are, yield curves will steepen. Steeper yield curves will boost banks’ net interest margins (Chart 24).   In addition, faster economic growth will put a lid on defaults. Banks have set aside considerable capital for pandemic-related loan losses. Yet, the wave of defaults that so many feared has failed to materialize. According to the American Bankruptcy Institute, commercial bankruptcies are lower now than they were this time last year (Chart 25). Personal loan delinquencies have also been trending down. The 60-day delinquency rate on credit card debt fell to 1.16% in October from 2.02% a year earlier. The delinquency rate for mortgages fell from 1.54% to 0.98%. Only auto loan delinquencies registered a tiny blip higher (Table 3). Chart 25Commercial Bankruptcies Are Well Contained Table 3Personal Loan Delinquencies Have Also Been Trending Lower Just A “Value Bounce”? In our conversations with clients, many investors are open to the idea that value stocks are due for a cyclical bounce. However, most still believe that growth stocks will fare best over a longer-term horizon. Such a view is understandable. After all, profit growth is the principal driver of equity returns. If, by definition, growth companies enjoy faster earnings growth, does it not stand to reason that growth stocks will outperform value stocks over the long haul? Well, actually, it doesn’t. What matters is profit growth relative to expectations, not absolute profit growth. If earnings rise quickly, but by less than investors had anticipated, stock prices could still go down. Historically, investors have tended to extrapolate earnings trends too far into the future, which has led them to overpay for growth stocks. Chart 26 demonstrates this point analytically. It features the results of a study by Louis Chan, Jason Karceski, and Josef Lakonishok. The authors sorted companies by projected five-year earnings growth and then compared the analysts’ forecasts with realized earnings. For the most part, they found that there was no relationship between expected profit growth and realized profit growth beyond horizons of two years. In general, the higher the long-term earnings growth estimates, the more likely actual earnings were to miss expectations. Chart 26Investors Tend To Overpay For Growth The Paradox Of Growth Given the difficulty of picking individual stocks that will consistently surpass earnings estimates, should investors simply allocate the bulk of their capital to sectors such as technology that have the best long-term growth prospects while eschewing structurally challenged sectors such as energy and financials? Again, the answer is not as obvious as it may seem. As Chart 27 illustrates, stocks in industries that experience a burst of output growth do tend to outperform other stocks. However, over the long haul, companies in fast growing industries do not outperform their peers (Chart 28). In other words, stock prices seem to respond more to unanticipated changes in industry growth rather than to the trend level of growth. Chart 27Stocks In Industries That Experience A Burst Of Output Growth Do Tend To Outperform Other Stocks … Chart 28… But Over The Long Haul, Companies In Fast-Growing Industries Do Not Outperform Their Peers Explaining Tech Outperformance In this vein, it is useful to examine what has powered the performance of US tech stocks over the past 25 years. Chart 29 shows that faster sales-per-share growth explains less than half of tech’s outperformance since 1996 and none of tech’s outperformance in the period up to 2011. The majority of tech’s outperformance is explained by greater margin expansion and an increase in the P/E ratio at which tech stocks trade relative to the rest of the stock market. Chart 29Decomposing Tech Outperformance What accounts for the significant increase in tech profit margins? In two words, the answer is “monopoly power.” Tech companies are particularly susceptible to network effects: The more people who use a particular tech platform, the more attractive it is for others to use it. Facebook is a classic example. Second, tech companies benefit significantly from scale economies. Once a piece of software has been written, creating additional copies costs almost nothing. Even in the hardware realm, the marginal cost of producing an additional chip is tiny compared to the fixed cost of designing it. All of this creates a winner take-all environment where success begets further success. Normally, structurally fast-growing industries attract more competition, which increases the odds that up-and-coming firms will displace incumbents. The growth of tech monopolies has subverted that process, allowing profits to rise significantly. A Tougher Path Forward For Tech A key question for investors is how much additional scope today’s tech monopolies have to expand profits. While it is difficult to generalize, two broad forces are likely to curtail future earnings growth. First, many tech titans have become so big that their future growth will be driven less by their ability to take market share from competitors and more by the overall size of the markets in which they operate. As it is, close to three-quarters of US households have an Amazon Prime account. Slightly over half have a Netflix account. Nearly 70% have a Facebook account. Google commands 92% of the internet search market. Together, Google and Facebook generate about 60% of all online advertising revenue. Second, the monopoly power wielded by tech companies makes them vulnerable to governmental action, including higher taxes, increased regulation, and stronger anti-trust enforcement. Importantly, it is not just the left that wants greater scrutiny of tech companies. According to a recent Pew Research study, more than half of conservative Republicans favor increasing government regulation of the tech sector (Chart 30). Chart 30Conservatives Favor Increased Government Regulation Of Big Tech Companies We do not expect tech stocks to decline in absolute terms since they still have a variety of tailwinds supporting them. Nevertheless, our bet is that the cyclical shift in favor of value stocks we are seeing now will usher in a period of outperformance for value names that could last for much of this decade. Not only are value stocks exceptionally cheap compared to growth stocks (Chart 31), but as we discuss below, bond yields likely reached a secular bottom this year. This could set the stage for a period of lasting outperformance for value plays. Chart 31Value Stocks Are Extremely Cheap Relative To Growth Stocks C. Fixed Income Position For Steeper Yield Curves As discussed earlier, central banks are unlikely to raise rates over the next 2-to-3 years. In fact, short-term real rates will probably decline further in 2021 as inflation expectations rise towards central bank targets. What about longer-term bond yields? Chart 32 displays the expected path of policy rates in the major developed economies now and at the start of 2020. The chart suggests that there is still scope for rate expectations in the post-2023 period to recover some of the ground they have lost since the start of the pandemic. This implies that bond investors should position for steeper yield curves, while keeping duration risk at below-benchmark levels. They should also favor inflation-linked securities over nominal bonds. Chart 32Policy Rate Expectations Remain Below Pre-Pandemic Levels Avoid “High Beta” Government Bond Markets The highest-yielding bond markets tend to have the highest “betas” to the general direction of global bond yields (Chart 33). This means when global bond yields are rising, higher-yielding markets such as the US usually experience the biggest selloff in bond prices. Chart 33High-Yielding Bond Markets Are The Most Cyclical This pattern exists because faster growth has a more subdued impact on rate expectations in economies such as Europe and Japan where the neutral rate of interest is stuck deep in negative territory. For example, if stronger growth lifts the neutral rate in Japan from say, -4% to -2%, this would still not warrant raising rates. In contrast, if stronger growth lifts the neutral rate from -1% to +1% in the US, this would eventually justify a rate hike.  As such, we would underweight US Treasurys in global government bond portfolios. We expect the 10-year Treasury yield to increase to around 1.3%-to-1.5% by the end of 2021, which is above current expectations of 1.15% based on the forward curve. Conversely, we would overweight European and Japanese government bond markets. After adjusting for currency-hedging costs, US Treasurys offer only a small yield pickup over European and Japanese bonds but face a much greater risk of capital losses as rate expectations recover (Table 4). Table 4Bond Markets Across The Developed World BCA’s global fixed-income strategists have a neutral recommendation on Canadian and Australian bonds. While Canadian and Australian yields are also “high beta,” both the BoC and the RBA are very active purchasers in their domestic markets. Stay Overweight High-Yield Developed Market Corporate Debt In fixed-income portfolios, we would overweight corporate debt relative to safer government bonds. In an economic environment where monetary policy remains accommodative and growth is rebounding, corporate default rates should remain contained, which will keep spreads from widening. Within corporate credit, we favor high yield over investment grade. Geographically, we prefer US corporate bonds over euro area bonds. The former trade with a higher yield and spread than the latter (Charts 34A & B). Chart 34AFavor High-Yield Bonds Over Investment-Grade ... Chart 34B… And US Corporates Over Euro Area One way to gauge the attractiveness of credit is to look at the percentile rankings of 12-month breakeven spreads. The 12-month breakeven spread is the amount of credit spread widening that can occur before a credit product starts to underperform a duration-matched, risk-free government bond over a one-year horizon. For US investment-grade corporates, the breakeven spread is currently in the bottom decile of its historic range, which is rather unattractive from a risk-adjusted perspective. In contrast, the US high-yield breakeven spread is currently in the 62nd percentile, which is quite enticing. In the UK, high-yield debt is more appealing than investment grade, although not quite to the same extent as in the US. In the euro area, both high-yield and investment-grade credit are fairly unattractive (Chart 35). Chart 35Corporate Bond Breakeven Spread Percentile Rankings Outside the corporate sector, our US bond strategists like consumer ABS due to the strength of household balance sheets. They also see value in municipal bonds. However, they would avoid MBS, as prepayment risks are elevated in that sector. EM credit should also benefit from the combination of stronger global growth and a weaker US dollar. Long-Term Inflation Risk Is Underpriced As noted earlier in the report, inflation is unlikely to rise significantly over the next three years. Beyond then, a more inflationary environment is probable. Chart 36 shows that the wage-version of the Phillips curve in the US is alive and well. It just so happens that over the past three decades, the labor market has never had a chance to overheat. Something always came along that derailed the economy before a price-wage spiral could develop. This year it was the pandemic. In 2008 it was the Global Financial Crisis. In 2000 it was the dotcom bust and in the early 1990s it was the collapse in commercial real estate prices following the Savings and Loan Crisis. Admittedly, only the pandemic qualifies as a true “exogenous” shock. The prior three recessions were endogenous in nature to the extent that they were preceded by growing economic imbalances, laid bare by a Fed hiking cycle. One can debate the degree to which the global economy is suffering from imbalances today, but one thing is certain: no major central bank is keen on raising rates anytime soon. Central banks want higher inflation. They are likely to get it. D. Currencies, Commodities, And Yes, Bitcoin Dollar Bear Market To Continue In 2021 The dollar faces a number of headwinds going into next year. First, interest rate differentials have moved significantly against the greenback. At the start of 2019, US real 2-year rates were about 190 basis points above rates of other developed economies; today, US real rates are around 60 basis points lower than those abroad. In fact, as Chart 37 shows, the trade-weighted dollar has weakened less than one would have expected based on the decline in interest rate differentials. This suggests that there could be some “catch-up” weakness for the dollar next year even if rate differentials remain broadly stable. Chart 36Is The Phillips Curve Really Dead? Chart 37A Relatively Muted Decline In The Dollar Given The Move In Real Yield Differentials   Second, the US dollar is a counter-cyclical currency, meaning that it tends to move in the opposite direction of the global business cycle (Chart 38). If the global economy strengthens next year thanks to an effective vaccine, the dollar should weaken. Chart 38The Dollar Is A Countercyclical Currency Chart 39USD Remains Overvalued Third, the US dollar remains about 13% overvalued based on Purchasing Power Parity (PPP) exchange rates (Chart 39). This overvaluation is also reflected in the large US current account deficit, which rose in the second quarter to the highest level since 2008 and is on track to swell even further in the second half of the year. Technicals Are Dollar Bearish Admittedly, many investors are now bearish on the dollar. Shouldn’t one be a contrarian and adopt a bullish dollar view? Not necessarily. In most cases, being contrarian makes sense. However, this does not apply to the dollar. The dollar is a high-momentum currency (Chart 40). When it comes to trading the dollar, it pays to be a trend follower. Chart 40The Dollar Is A High Momentum Currency One of the simplest and most profitable trading rules for the dollar is to go long the greenback when it is trading above its moving average and go short when it is trading below its moving average (Chart 41). Today, the trade-weighted dollar is trading below its 3-month, 6-month, 1-year, and 2-year moving averages. Along the same lines, the dollar performs best when sentiment is bullish and improving. In contrast, the dollar does worse when sentiment is bearish and deteriorating, as it is now (Chart 42). Chart 41Being A Contrarian Doesn’t Pay When It Comes To Trading The Dollar (I) Chart 42Being A Contrarian Doesn’t Pay When It Comes To Trading The Dollar (II) The bottom line is that both fundamental factors – interest rate differentials, global growth, valuations, current account dynamics – and technical factors – moving average rules and sentiment – all point to dollar weakness next year. Top Performing Currencies In 2021 EUR/USD is likely to rise to 1.3 by the middle of next year. The ECB does not want a stronger currency, but with euro area interest rates already in negative territory, there is not much it can do. The Swedish krona, as a highly cyclical currency, should strengthen against the euro. In contrast, the Swiss franc, a classically defensive currency, will weaken against the euro. It is more difficult to forecast the direction of the pound given uncertainty about ongoing Brexit talks. The working assumption of BCA’s geopolitical team is that Prime Minister Boris Johnson has sufficient economic and political incentives to arrive at a trade deal, a parliamentary majority to get it approved, and a powerful geopolitical need to mollify Scotland. This bodes well for sterling. The yen is a very defensive currency. Thus, in an environment of strengthening global growth, the yen is likely to trade flat against the dollar, and in the process, lose ground against most other currencies. We are most bullish about the prospects for EM and commodity currencies going into next year. China is likely to let its currency strengthen further in return for a partial rollback of tariffs by the Biden administration. A stronger yuan will allow other currencies in Asia to appreciate. Stay Bullish On Commodities And Commodity Currencies The combination of a weaker US dollar and stronger global growth should support commodity prices in 2021. Industrial metals outperformed oil this year, but the opposite should be true next year. Chart 43Oil Prices Are Expected To Recover While the long-term outlook for crude is murky in light of the shift towards electric vehicles, the near-term picture remains favorable due to the cyclical rebound in petroleum demand and ongoing OPEC and Russian supply discipline. BCA’s commodity strategists expect the average price of Brent to exceed market expectations by about $14 in 2021, which should help the Norwegian krone, Canadian dollar, Russian ruble, Mexican peso, and Colombian peso (Chart 43). Favor Gold Over Bitcoin As An Inflation Hedge Gold has traditionally served as the go-to hedge against inflation. These days, however, there is a new competitor in town: bitcoin. In traditional economic parlance, money serves three purposes: as a medium of exchange; as a unit of account; and as a store of value. Both gold and bitcoin flunk the test for the first two purposes. Few transactions are conducted in either gold or bitcoin. It is even rarer for prices of goods and services to be set in ounces of gold or units of bitcoin. Gold arguably does better as a store of value. It has been around for a long time and if all else fails, it can always be melted down and turned into nice jewelry. Bitcoin’s Achilles Heel Bitcoin’s defenders argue that the cryptocurrency does serve as a store of value because one day, it will reach a critical mass that will make it a viable medium of exchange and a functional unit of account. Yet, this argument is politically naïve. Countries with fiat currencies derive significant benefits from their ability to create money out of thin air that can then be used to pay for goods and services. In the US, this “seigniorage revenue” amounts to over $100 billion per year. The existence of fiat currencies also gives central banks the power to set interest rates and provide liquidity backstops to the financial sector. Bitcoin’s ability to facilitate anonymous transactions is also its Achilles heel. The widespread use of bitcoin would make it more difficult for governments to tax their citizens. All this suggests that bitcoin will never reach a critical mass where it becomes a viable medium of exchange or functional unit of account. Governments will step in to ban or greatly curtail its usage before then. And without the ability to reach this critical mass, bitcoin’s utility as a store of value will disappear. Hence, investors looking for some inflation protection in their portfolios should stick with gold. Peter Berezin Chief Global Strategist peterb@bcaresearch.com Footnotes 1  Heather Reese, A. Danielle Iuliano, Neha N. Patel, Shikha Garg, Lindsay Kim, Benjamin J. Silk, Aron J. Hall, Alicia Fry, and Carrie Reed, “Estimated incidence of COVID-19 illness and hospitalization — United States, February–September, 2020,” Clinical Infectious Diseases (Oxford Academic), November 25, 2020. Global Investment Strategy View Matrix Special Trade Recommendations This table provides trade recommendations that may not be adequately represented in the matrix on the preceding page. Current MacroQuant Model Scores

Highlights Chart 1Bond Yields & The CRB/Gold Ratio In our last report of November, we noted that the rising COVID case count was likely to lead to a challenging few months for the US economy, but we also questioned whether financial markets would pay attention or whether they would stay focused on the vaccine roll-out and eventual economic recovery. We now have our answer. November’s employment report was the worst since April, but the Treasury curve has bear-steepened, credit spreads have come in and TIPS have outperformed nominals. What’s more, the jump in the CRB Raw Industrials / Gold ratio suggests that the 10-year Treasury yield has even more near-term upside (Chart 1). With a vaccine on the horizon and Congress closing in on a fiscal relief package, investors should stay positioned for the reflation trade on a 6-12 month horizon: below-benchmark portfolio duration, nominal and real yield curve steepeners, inflation curve flatteners, overweight TIPS versus nominals and overweight corporate bonds rated Ba and higher. Feature Investment Grade: Overweight Investment grade corporate bonds outperformed the duration-equivalent Treasury index by 233 basis points in November, bringing year-to-date excess returns up to -74 bps. The strong rally in corporate bonds since March has culminated in extremely tight valuations for investment grade corporates. The 12-month breakeven spread for the Bloomberg Barclays Corporate Index (adjusted to keep the average credit rating constant) has only been tighter 4% of the time since 1995 (Chart 2). The same figure for the Baa-rated credit tier is 5%. We retain a positive outlook on corporate credit despite these stretched valuations. In our view, an environment where the economy is recovering and where the Fed will be very cautious about scaling back accommodation is the exact sort of environment where we should expect a lot of enthusiasm for spread product and, as a result, extremely tight spreads. We will not be surprised if our 12-month breakeven spread percentile rank valuation measure reaches its all-time expensive level within the next couple of months. While the macro environment makes it difficult to turn negative on investment grade corporates, we acknowledge that other sectors may offer better opportunities, particularly in the higher credit tiers. Specifically, we find better value in tax-exempt municipal bonds than in corporates and recommend that investors favor the former over the latter. At the sector level, we continue to recommend overweight allocations to subordinate Bank bonds, Healthcare and Energy bonds. We also advise underweight allocations to Technology and Pharmaceutical bonds. Chart 2Investment Grade Market Overview Table 3ACorporate Sector Relative Valuation And Recommended Allocation* Table 3BCorporate Sector Risk Vs. Reward* High-Yield: Neutral High-Yield outperformed the duration-equivalent Treasury index by 382 basis points in November, bringing year-to-date excess returns up to -5 bps. After last month’s strong outperformance, Ba-rated junk bonds are now beating duration-equivalent Treasuries by 267 bps, year-to-date. The B and Caa credit tiers are lagging by 179 bps and 548 bps, respectively. We still view Ba-rated junk bonds as the sweet spot within the corporate credit space. The sector is relatively insulated from default risk and yet still offers a sizeable spread pick-up over investment grade corporates (Chart 3). We remain underweight B-rated and lower junk bonds for now as those securities are pricing-in a relatively optimistic outlook for the default rate. But, an imminent vaccine roll-out makes that outlook appear more realistic and we could soon upgrade the lower-rated junk credit tiers when we think the value is exhausted in the Ba-rated and higher securities. Looking at value for the junk index as a whole, we see that the index is pricing-in a default rate of 3% for the next 12 months, significantly below the 8.3% that was observed during the most recent 12-month period (panel 3). However, only four corporate issuers defaulted in October down from a monthly peak of 22 in July. Job cut announcements, an excellent indicator of the default rate, are also falling rapidly (bottom panel). At the sector level, we advise overweight allocations to high-yield Technology and Energy bonds. We are underweight the Healthcare and Pharmaceutical sectors. Chart 3High-Yield Market Overview MBS: Underweight Mortgage-Backed Securities underperformed the duration-equivalent Treasury index by one basis point in November, dragging year-to-date excess returns down to -39 bps. The conventional 30-year MBS index option-adjusted spread (OAS) tightened 1 bp on the month, and it currently sits at 64 bps (Chart 4). This is significantly higher than the 59 bps offered by Aa-rated corporate bonds, the 53 bps offered by Agency CMBS and the 25 bps offered by Aaa-rated consumer ABS. Despite the relatively attractive OAS, we continue to view the elevated primary mortgage spread as a risk for MBS investors. It suggests that mortgage rates need not rise alongside Treasury yields in the near-term, meaning that mortgage refinancings can continue at their current rapid pace (panel 3). All else equal, this elevated refinancing activity will pressure MBS spreads wider. The recent spike in the mortgage delinquency rate does not pose a near-term risk to spreads as it is being driven by households that have been granted forbearance from the federal government (panel 4). The risk for MBS holders only comes into play if many households are unable to resume their regular mortgage payments when the forbearance period expires early next year. But even in that case, further government intervention to either support household incomes or extend the forbearance period would mitigate the risk. Chart 4MBS Market Overview Government-Related: Underweight The Government-Related index outperformed the duration-equivalent Treasury index by 64 basis points in November, bringing year-to-date excess returns up to -222 bps. Sovereign debt outperformed duration-equivalent Treasuries by 157 bps on the month, bringing year-to-date excess returns up to -269 bps. Foreign Agencies outperformed the Treasury benchmark by 46 bps in November, bringing year-to-date excess returns up to -647 bps. Local Authority debt outperformed Treasuries by 139 bps in November, bringing year-to-date excess returns up to -228 bps. Domestic Agency bonds outperformed by 10 bps, bringing year-to-date excess returns up to -23 bps. Supranationals outperformed by 9 bps, bringing year-to-date excess returns up to +2 bps. US dollar weakness is usually a boon for Emerging Market (EM) Sovereign and Foreign Agency returns. However, this year’s dollar weakness has occurred mostly relative to other Developed Market currencies (Chart 5). Value has improved somewhat for EM Sovereigns during the past few weeks, but the index continues to offer less spread than the Baa-rated US Credit index (panel 4). At the country level, Turkey, Colombia, Mexico, Russia and South Africa are the only countries that offer a spread pick-up relative to duration and quality-matched US corporates. Of those, only Mexico looks attractive on a risk/reward basis. Chart 5Government-Related Market Overview Municipal Bonds: Overweight Municipal bonds outperformed the duration-equivalent Treasury index by 130 basis points in November, bringing year-to-date excess returns up to -340 bps (before adjusting for the tax advantage). Municipal bond spreads tightened sharply relative to both Treasuries and Corporates in November, but they remain exceptionally attractive relative to history (Chart 6). In fact, as we showed in a recent report, the Bloomberg Barclays Revenue Bond index offers a greater yield than the quality-matched Credit index across the entire maturity spectrum (before adjusting for the tax advantage).1 This is also true for the Bloomberg Barclays General Obligation (GO) index beyond the 12-year maturity point. Eight-to-twelve-year maturity GO bonds trade only 1 basis point through the Credit index, implying a breakeven effective tax rate of 4%. Six-to-eight-year maturities trade 11 bps through the Credit index, implying a breakeven effective tax rate of 16%. Extraordinary valuation is the main reason for our recommendation to overweight municipal bonds. The severe ongoing state & local government credit crunch is a concern, but it is a risk we are willing to take. It now looks possible that a relief package containing some federal funds for state & local governments will be passed before the end of the year. This would alleviate a lot of the concern. But even in the absence of federal assistance, the combination of austerity measures (bottom panel) and all-time high State Rainy Day Fund balances should help stave off a wave of municipal downgrades. Chart 6Municipal Market Overview Treasury Curve: Buy 5-Year Bullet Versus 2/10 Barbell The Treasury curve bull-flattened in November, but then bear-steepened sharply during the first week of December. All told, the 2/10 Treasury slope is currently 81 bps, 7 bps steeper than at the end of October. The 5/30 Treasury slope is 131 bps, 4 bps steeper than at the end of October. Our expectation is that continued economic recovery will cause investors to price-in eventual monetary tightening at the long-end of the Treasury curve. With the Fed maintaining a firm grip on the front end, this will lead to Treasury curve bear steepening. A timely vaccine roll-out and/or further fiscal stimulus will speed this process up. We recommend positioning for a steeper curve by owning the 5-year Treasury note and shorting a duration-matched barbell consisting of the 2-year note and 10-year notes. This position is designed to profit from 2/10 curve steepening.     Valuation is a concern with our recommended steepener, as the 5-year yield is below the yield on the duration-matched 2/10 barbell (Chart 7). However, the 5-year looked much more expensive during the last zero-lower-bound period between 2010 and 2013 (bottom 2 panels). We anticipate a return to similar levels. Chart 7Treasury Yield Curve Overview TIPS: Overweight TIPS outperformed the duration-equivalent nominal Treasury index by 70 basis points in November, bringing year-to-date excess returns up to -23 bps. The 10-year and 5-year/5-year forward TIPS breakeven inflation rates rose 8 bps and 5 bps on the month. They currently sit at 1.91% and 1.96%, respectively. Core CPI was flat in October and the year-over-year rate dropped from 1.73% to 1.63%. The 12-month trimmed mean CPI fell even more – from 2.37% to 2.22% – so the gap between core and trimmed mean inflation continued to narrow (Chart 8). We expect further narrowing in the months ahead, and therefore expect core CPI to come in relatively hot. For this reason, we recommend maintaining an overweight allocation to TIPS versus nominal Treasuries for the time being, even though the 10-year TIPS breakeven rate is now somewhat expensive according to our Adaptive Expectations Model (panel 2).2 Inflation pressures may moderate once core and trimmed mean inflation measures converge, and this could give us an opportunity to tactically reduce TIPS exposure sometime next year. We also recommend holding real yield curve steepeners and inflation curve flatteners. With the Fed now officially targeting an overshoot of its 2% inflation goal, we would expect the cost of 2-year inflation protection to rise above the cost of 10-year inflation protection (panel 4). With the Fed also exerting more control over short-dated nominal yields than over long-term ones, we expect that short-maturity real yields will come under downward pressure relative to the long end (bottom panel). Chart 8TIPS Market Overview ABS: Overweight Asset-Backed Securities outperformed the duration-equivalent Treasury index by 11 basis points in November, bringing year-to-date excess returns up to +82 bps. Aaa-rated ABS outperformed the Treasury benchmark by 10 bps on the month, bringing year-to-date excess returns up to +68 bps. Non-Aaa ABS outperformed by 17 bps, bringing year-to-date excess returns up to +174 bps (Chart 9). On paper, the Treasury department’s decision to let the Term Asset-Backed Loan Facility (TALF) expire at the end of the year is quite negative for ABS. However, as we explained in a recent report, we don’t expect a material impact on spreads.3 For one thing, Aaa ABS spreads are already well below the borrowing cost offered by TALF. But more importantly, consumer credit quality remains quite robust. As we first explained back in June, the stimulus received from the CARES act led to a significant increase in disposable income and a jump in the savings rate (panel 4).4  Faced with an income boost and few spending opportunities, many households took the opportunity to pay down consumer debt. Granted, further income support from Congress is needed now that the CARES act’s enhanced unemployment benefits have expired. But given the substantial boost to savings that has already occurred, we are confident that more stimulus will arrive in time to prevent a wave of consumer bankruptcies. Chart 9ABS Market Overview Non-Agency CMBS: Neutral Chart 10CMBS Market Overview Non-Agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 85 basis points in November, bringing year-to-date excess returns up to -168 bps. Aaa Non-Agency CMBS outperformed Treasuries by 71 bps on the month, bringing year-to-date excess returns up to -2 bps. Non-Aaa Non-Agency CMBS outperformed by 127 bps, bringing year-to-date excess returns up to -620 bps (Chart 10). We continue to recommend an overweight allocation to Aaa-rated Non-Agency CMBS and an underweight allocation to non-Aaa CMBS. Even with the imminent expiry of TALF, Aaa CMBS spreads are already well below the cost of borrowing through TALF and thus will not be negatively impacted.5  Meanwhile, the structurally challenging environment for commercial real estate could lead to problems for lower-rated CMBS (panels 3 & 4). Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 38 basis points in November, bringing year-to-date excess returns up to +55 bps. The average index spread tightened 6 bps on the month. It currently sits at 53 bps, above typical historical levels (bottom panel). At its September meeting, the Fed decided to slow its pace of Agency CMBS purchases. It is no longer looking to increase its Agency CMBS holdings, but rather, it is only purchasing what is “needed to sustain smooth market functioning”. This is nonetheless a Fed back-stop of the market, and it does not change our overweight recommendation.    Appendix A: Buy What The Fed Is Buying The Fed rolled out a number of aggressive lending facilities on March 23. These facilities focused on different specific sectors of the US bond market. The fact that the Fed has decided to support some parts of the market and not others has caused some traditional bond market correlations to break down. It has also led us to adopt of a strategy of “Buy What The Fed Is Buying”. That is, we favor those sectors that offer attractive spreads and that benefit from Fed support. The below Table tracks the performance of different bond sectors since the March 23 announcement. We will use this to monitor bond market correlations and evaluate our strategy’s success. Table Performance Since March 23 Announcement Of Emergency Fed Facilities Appendix B: Butterfly Strategy Valuations  The following tables present the current read-outs from our butterfly spread models. We use these models to identify opportunities to take duration-neutral positions across the Treasury curve. The following two Special Reports explain the models in more detail: US Bond Strategy Special Report, “Bullets, Barbells And Butterflies”, dated July 25, 2017, available at usbs.bcaresearch.com US Bond Strategy Special Report, “More Bullets, Barbells And Butterflies”, dated May 15, 2018, available at usbs.bcaresearch.com Table 4 shows the raw residuals from each model. A positive value indicates that the bullet is cheap relative to the duration-matched barbell. A negative value indicates that the barbell is cheap relative to the bullet. Table 4Butterfly Strategy Valuation: Raw Residuals In Basis Points (As Of December 4TH, 2020) Table 5 scales the raw residuals in Table 4 by their historical means and standard deviations. This facilitates comparison between the different butterfly spreads. Table 5Butterfly Strategy Valuation: Standardized Residuals (As Of December 4TH, 2020) Table 6 flips the models on their heads. It shows the change in the slope between the two barbell maturities that must be realized during the next six months to make returns between the bullet and barbell equal. For example, a reading of 70 bps in the 5 over 2/10 cell means that we would only expect the 5-year to outperform the 2/10 if the 2/10 slope steepens by more than 70 bps during the next six months. Otherwise, we would expect the 2/10 barbell to outperform the 5-year bullet. Table 6Discounted Slope Change During Next 6 Months (BPs) Appendix C: Excess Return Bond Map The Excess Return Bond Map is used to assess the relative risk/reward trade-off between different sectors of the US bond market. It is a purely computational exercise and does not impose any macroeconomic view. The Map’s vertical axis shows 12-month expected excess returns. These are proxied by each sector’s option-adjusted spread. Sectors plotting further toward the top of the Map have higher expected returns and vice-versa. Chart 11Excess Return Bond Map (As Of December 4TH, 2020) Our novel risk measure called the “Risk Of Losing 100 bps” is shown on the Map’s horizontal axis. To calculate it, we first compute the spread widening required on a 12-month horizon for each sector to lose 100 bps or more relative to a duration-matched position in Treasury securities. Then, we divide that amount of spread widening by each sector’s historical spread volatility. The end result is the number of standard deviations of 12-month spread widening required for each sector to lose 100 bps or more versus a position in Treasuries. Lower risk sectors plot further to the right of the Map, and higher risk sectors plot further to the left.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Weekly Report, “Political Risk Will Dominate In A Pivotal Month For The Bond Market”, dated October 13, 2020, available at usbs.bcaresearch.com 2 For more details on our model please see US Bond Strategy Weekly Report, “How Are Inflation Expectations Adapting?”, dated February 11, 2020, available at usbs.bcaresearch.com 3 Please see US Bond Strategy Weekly Report, “Preparing For A Dark Winter … But Do Markets Care?”, dated November 24, 2020, available at usbs.bcaresearch.com 4 Please see US Bond Strategy Weekly Report, “No Holding Back”, dated June 16, 2020, available at usbs.bcaresearch.com 5 Please see US Bond Strategy Weekly Report, “Preparing For A Dark Winter … But Do Markets Care?”, dated November 24, 2020, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification Corporate Sector Relative Valuation And Recommended Allocation

Highlights Inflation Breakeven Trades: We are taking profits on our recommended inflation breakeven widening trades in Italy and Canada, as breakevens in both countries are no longer below the fair values implied by our models. We are initiating a new trade this week, going long French 10-year inflation-linked bonds versus French nominal OATs, as French breakevens remain below fair value. Yield Curve Butterfly Trades: We are closing three of our four outstanding government bond yield curve trade recommendations, taking profits in France and Italy and realizing a loss in the UK. We are maintaining our US 5/7/10 butterfly trade, which is the cheapest way to position for an expected steepening of the Treasury curve based on our valuation models. Cross-Country Spread Trades: We are cutting our losses in our New Zealand-UK government bond spread trade, with the odds of the RBNZ shifting to a negative interest rate policy severely curtailed by political pressure over surging New Zealand house prices. We are maintaining our US-Germany spread widening trade, as the spread is too narrow based on our fair value model and we see more scope for US Treasury yields to drift higher in the coming months. Feature Dear Client, Next week, we will be jointly publishing our semi-annual Central Bank Monitor Chartbook along with our colleagues at BCA Research Foreign Exchange Strategy. You will receive that report a few days later than usual on Friday, December 11. We will return to our regular publishing schedule on Tuesday, December 15 with our 2021 Key Views report outlining our main investment themes and ideas for the upcoming year. Best Regards, Rob Robis As we enter the final weeks of an incredibly eventful and (unfortunately) all too memorable 2020, our attention now turns to investment ideas for the coming New Year. This week, all BCA Research clients will receive the 2021 Outlook report, detailing the key themes and recommendations from all our strategists. We will follow that up with our own 2021 Global Fixed Income Strategy outlook report later this month. The waning days of the year also offer a good time to review our more short-term trade recommendations currently in our Tactical Overlay portfolio. In addition, the waning days of the year also offer a good time to review our more short-term trade recommendations currently in our Tactical Overlay portfolio (Table 1). Several of our suggested trades have generated a solid profit (like inflation breakeven wideners) but have now outlived their original rationale. Others, like some of our yield curve trades in Europe, have not gone as we expected and should therefore be closed out. Table 1Changes To Our Tactical Overlay Portfolio As a reminder to our regular readers, our Tactical Overlay is a portfolio of individual trade ideas within the global fixed income space with an investment horizon of six months or less. These differ from our more typical strategic (6-12 month) recommendations that also populate our model bond portfolio. Ideas for our Tactical Overlay trades often stem from our fair value models, but can also be plays on events that we expect will be market relevant on a near-term basis, like central bank meetings. All recommended trades are implemented using specific securities, rather than generic Bloomberg tickers or bond indices. This allows for a more transparent process where clients can follow along with the performance of our trades. Evaluating Our Tactical Inflation-Linked Breakeven Trades We currently have two open tactical trade recommendations involving inflation-linked bonds: Long 10-year Italian inflation-linked bonds vs short 10-year Italian bond futures Long 10-year Canadian inflation-linked bonds vs short 10-year Canadian bond futures We initiated both of these trades back in June of this year, as well as an additional trade involving US TIPS, based on the output of our inflation breakeven fair value framework. In our models, we regress 10-year inflation breakevens on the annual rate of change of oil prices in local currency terms and a multi-year moving average of realized headline inflation.1 At the time of our mid-year report, inflation breakevens were too low on our models in the majority of developed market countries with inflation-linked bonds – a lingering after-effect of the COVID-19 shock to global growth in the second quarter of 2020 (Chart 1). Since then, 10-year inflation breakevens have caught up to fair value in the US, Germany, Italy and Canada, and have even moved above fair value in the UK and Australia. Chart 1A Big Shift In Inflation Breakeven Valuations In June, we also entered into a US 10-year TIPS breakeven widening trade, but we took profits on the trade once US breakevens returned back to our model fair value estimate in September. We now see a similar situation in Canada (Chart 2) and Italy (Chart 3) where breakevens have converged to our model-implied fair value. Chart 2Canadian 10-Year Inflation Breakeven Model A move above fair value is possible, but could be harder to achieve with the Canadian dollar and euro steadily trending higher which could weigh on the market’s view on future inflation in Canada and Italy. We are taking profits on our Canada and Italy 10-year breakeven trades, realizing profits of 4.7% and 5.6% respectively.  Thus, we are taking profits on our Canada and Italy 10-year breakeven trades, realizing profits of 4.7% and 5.6% respectively. The Italian returns were boosted considerably by the long side of this trade, as we entered the position when the 10-year real yield was +1.05% and which has since collapsed to -0.05% on the back of the massive rally in Italian bonds. One place where breakevens still look attractively cheap, trading close to one standard deviation below our model fair value, is in France (Chart 4). This contrasts with the breakevens in Italy and Germany that have fully converged to fair value. Thus, we are entering a new trade this week, going long the on-the-run 10yr French inflation-linked bond (OATi) and shorting French bond futures (Euro-OATs). The hedge ratio used for this trade to keep both legs duration matched, given the much shorter duration of the OATi relative to nominal French bonds, is 0.49 (see the Tactical Overlay table on page 17 for specific details on the securities used in the trade). Chart 3Italian 10-Year Inflation Breakeven Model Chart 4French 10-Year Inflation Breakeven Model Bottom Line: We are taking profits on our recommended inflation breakeven widening trades in Italy and Canada, while initiating a new breakeven widening position in France, based on the output of our breakeven fair value models. Evaluating Our Yield Curve/Butterfly Spread Trades Back in July, we initiated a series of yield curve butterfly spread trades in the US, UK, Italy and France.2 Butterfly spreads compare the yield of a single bond (bullets) to that of a duration-neutral combination of bonds with shorter and longer maturities relative to the bullet (barbells). Our valuation models produce fair value estimates of various butterfly combinations based on the relation of the butterfly spreads to the slope of the yield curve. We then combine those valuations with our own macro views on the future slope of yield curves to come up with potential value-based curve trades.3 We now evaluate our four existing curve trades in turn. Long UK 3/20 Barbell vs. 10-Year Bullet Our original rationale for entering this trade was two-fold. Firstly, this position was the most attractive butterfly combination in terms of the standardized deviation of the spread from its model-implied fair value. Secondly, there was a relatively low correlation between nominal UK bond yields and inflation breakevens--meaning that we could see a rise in long-dated inflation expectations that did not also push up nominal bond yields by a proportional amount. This made the trade consistent with our overall macro view back in July that the Gilt curve would flatten (the same rationale applies to the other two long barbell versus short bullet trades, or “flatteners”, in France and Italy that we discuss below). Unfortunately, our rationale did not play out as expected (Chart 5). Instead of reverting to fair value, the butterfly spread was mostly flat while the bullet grew more expensive relative to the barbell, driven by a rise in the model fair value. This in turn was due to significant steepening in the underlying 3/20 curve, contrary to our expectations. We also saw a significant overall upward shift in the overall UK Gilt curve, which generated losses on our long barbell position (which has a higher interest rate convexity) that overwhelmed the profits on our short bullet position. Going forward, there are good technical and strategic reasons to exit this trade. The butterfly spread is not yet at levels where it tends to mean-revert (second panel). In addition, Joe Biden’s US election victory has also increased the odds of a Brexit deal, which would put bear-steepening pressure on the UK Gilt curve. With that in mind, we are closing our Long UK 3/20 Barbell vs. 10-Year Bullet for a loss of -0.17%. Long France 2/30 Barbell vs. 5-Year Bullet Our rationale for entering this flattener was the same as in the UK. However, we fared quite a bit better here. The underlying 2/30 curve did flatten, as we expected, however, the butterfly spread itself moved further away from fair value, with the bullet component becoming relatively more expensive (Chart 6). So, as with the UK, the returns on this trade can be largely explained by the relative outperformance of the barbell component due to its higher convexity. In France, however, the effect worked to our favor as the yield curve shifted downwards significantly. The positive returns on the long French 30-year OAT component, where yields have been nearly slashed in half since July, dominated the other parts of the trade - even with the 30-year bond only being a small piece (11%) of the duration-weighted barbell Chart 5UK 3/10/20 Spread Fair Value Model Chart 6France 2/5/30 Spread Fair Value Model Although we did make profits on the flattener, it turned into a convexity bet that was not our original intention. Seeing as our underlying logic did not work out as expected, we are not comfortable remaining in this position. Thus, we are closing our France butterfly trade for a profit of 0.56%. Long Italy 5/30 Barbell vs. 10-Year Bullet As with the UK and France, we entered this trade based on its attractive model-based valuation and the relatively low correlation between inflation breakevens and nominal yields in France. Our expectation of flattening in the underlying 5/30 curve did not bear out as it remained mostly flat (Chart 7). We did see some reversion in the butterfly spread towards our model-implied fair value, which helped us make profits on our trade. Again, we cannot ignore the effect of convexity when looking at the outperformance of the barbell component. Yields fell dramatically across the Italian curve in one of the clearest examples of the yield-chasing behavior we have been describing this year.4 As Italian yields continue their race to the bottom, supported by ECB asset purchases and perceptions of more fiscal co-operation between the countries of Europe, there is a chance that this trade will continue to perform by virtue of its exposure to the long end of the Italian curve. However, as our original bias towards curve flattening did not play out, we prefer to maintain our exposure to Italian government debt via an overweight allocation in our model bond portfolio instead. We therefore close our Long Italy 5/30 Barbell vs. 10-Year Bullet for a profit of 0.83% Long US 7-Year Bullet vs. 5/10 Barbell The US was the only region where we initiated a “steepener” trade, with a long bullet versus short barbell combination that does well when the yield curve steepens. We chose this particular 5/7/10 butterfly as it was the most attractive steepener available based on our model-implied valuation that also fit our fundamental macro bias back in July towards US Treasury curve steepening – a view that we still hold today. With signs pointing towards further bear steepening of the Treasury curve, we feel comfortable keeping this US 5/7/10 butterfly spread trade open. Our rationale for initiating the trade was borne out, with the underlying 5/10 Treasury curve steepening and the butterfly spread tightening towards fair value (Chart 8). Our trade was supported by a continued rebound in long-dated US inflation expectations as well as the US election result, the most bond-bearish event of the year. Chart 7Italy 5/10/30 Spread Fair Value Model Chart 8US 5/7/10 Spread Fair Value Model Going forward, we see good reasons to maintain this trade. The butterfly spread, after briefly reaching expensive levels, is back to being attractively valued. Even if the residual were to dip back below zero, it would still have room to become more expensive, shoring up our trade. This trade also remains the most attractive of all the steepener trades on a model-implied valuation basis, removing any incentive to rotate towards another part of the curve. The odds favor more reflationary Treasury curve steepening after the US election. President-elect Biden has a stated goal of more fiscal stimulus, while his selection of Janet Yellen as Treasury Secretary signaling increased cooperation between monetary and fiscal authorities. With signs pointing towards further bear steepening of the Treasury curve, we feel comfortable keeping this US 5/7/10 butterfly spread trade open. Bottom Line: We are closing three of our four outstanding government bond yield curve trade recommendations, taking profits in France and Italy and realizing a loss in the UK. We are maintaining our US 5/7/10 butterfly trade, which is the cheapest way to position for an expected steepening of the Treasury curve based on our valuation models. Evaluating Our Cross-Country Yield Spread Trades We currently have two recommended trades involving plays on the spread between government bonds: Long 5-year New Zealand government bonds versus short 5-year UK Gilts, currency-hedged into GBP We initiated this trade on August 25, and to date the trade is severely underwater with a total return of -1.8%.5 That loss comes from the long New Zealand leg of the trade, as the 5-year NZ bond yield has increased by 34bps from our entry level. Chart 9A Rapid Shift Upward In NZ Rate Expectations The rationale for this trade was based on our assessment of the relative probability of the Bank of England (BoE) and Reserve Bank of New Zealand (RBNZ) moving to a negative interest rate policy. Both central banks hinted strongly at such a move throughout the summer months as part of their efforts to support pandemic-stricken economies. Our view back in late August was that it was more likely that the RBNZ would choose negative rates, as New Zealand had far lower inflation expectations than the UK and, unlike the British pound, the New Zealand dollar was not undervalued. This trade was initially profitable, but all that changed rapidly during the month of November. The RBNZ disappointed investor expectations on a move to negative rates at the November 11 monetary policy meeting. The central bank elected instead to increase the size of its existing quantitative easing program, while giving no hint that negative rates were coming soon. The response was a sharp move higher in both New Zealand bond yields and the New Zealand dollar (Chart 9). There was an even more violent adjustment in yields and the currency last week, after New Zealand Finance Minister Grant Robertson wrote a letter to RBNZ Governor Adrian Orr asking the central bank to change its policy remit to include controlling New Zealand house price inflation. Markets interpreted this blatant political pressure on the central bank as the end of any hopes of negative rates in New Zealand, with bond yields and the currency spiking higher once again. House prices have surged after the RBNZ aggressively cut interest rates earlier this year, with a rapidly rising share of new mortgages having higher loan-to-value ratios (Chart 10). House price inflation is now running at 19.8%, and Finance Minister Robertson did cite deteriorating housing affordability and inequality as the basis for his letter to the RBNZ. It is clear that a move to negative interest rates – which could further fuel the explosion in house prices – is now very difficult for the RBNZ to pull off without facing intense criticism. It is clear that a move to negative interest rates – which could further fuel the explosion in house prices – is now very difficult for the RBNZ to pull off without facing intense criticism. This shatters the underlying rationale for our long New Zealand/short UK yield spread trade (Chart 11). Chart 10RBNZ-Fueled Boom In House Prices Thus, we are choosing to cut our losses and close out our recommended trade. Long 10-year German Bunds versus short 10-year US Treasuries Chart 11Time To Cut Our Losses On The NZ-UK Trade We initiated this recommendation on October 27, and to date the trade is running a small loss of -0.17%.6 The rationale behind the trade was two-fold: Our valuation model for the 10-year UST-Bund yield spread showed that the spread was far below fair value; We turned more bearish on US Treasuries just before the US presidential election, downgrading our recommended allocation to underweight while also upgrading more defensive Germany – with its low yield-beta to US Treasuries - to overweight. The trade initially performed well, driven by faster growth and inflation in the US versus the euro area (Chart 12). The Treasury selloff has stalled of late, but we view this as more a consolidative pause than a near-term peak in yields. Chart 12Fundamentals Justify A Wider UST-Bund Spread With our Treasury-Bund valuation model still showing that the spread is too tight, and with the spread not looking overly stretched versus its 200-day moving average (Chart 13), we are keeping our US versus Germany trade in our Tactical Overlay portfolio. Chart 13Valuation & Momentum Point To A Wider UST-Bund Spread Bottom Line: We are cutting our losses in our New Zealand-UK government bond spread trade, with the odds of the RBNZ shifting to a negative interest rate policy severely curtailed by political pressure over surging New Zealand house prices. We are maintaining our US-Germany spread widening trade, as the spread is too narrow based on our fair value model and we see more scope for US Treasury yields to drift higher in the coming months   Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com Shakti Sharma Research Associate ShaktiS@bcaresearch.com Footnotes 1 Please see BCA Research Global Fixed Income Strategy Weekly Report, " How To Play The Revival Of Global Inflation Expectations", dated June 23, 2020, available at gfis.bcaresearch.com. 2 Please see BCA Research Global Fixed Income Strategy Weekly Report, "Global Yield Curve Trades: Netting Returns With Butterflies", dated July 7, 2020, available at gfis.bcaresearch.com. 3 Readers looking for more detailed background on butterfly trades and our yield curve modelling framework should refer to the July 7, 2020 Strategy Report where we initiated these trades. 4 Please see BCA Research Global Fixed Income Strategy Weekly Report, "We’re All Yield Chasers Now", dated August 11, 2020, available at gfis.bcaresearch.com. 5 Please see BCA Research Global Fixed Income Strategy Weekly Report, "Assessing The Leading Candidates To Join The Negative Rates Club", dated August 26, 2020, available at gfis.bcaresearch.com. 6 Please see BCA Research Global Fixed Income Strategy Weekly Report, "The Global Bond Implications Of Rising Treasury Yields", dated October 27, 2020 available at gfis.bcaresearch.com. Recommendations The GFIS Recommended Portfolio Vs. The Custom Benchmark Index Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

Highlights US Election & COVID-19: Joe Biden’s apparent victory in the US presidential race, as well as the announcement of a potential successful COVID-19 vaccine trial, are both bond-bearish outcomes. This is especially so for US Treasuries given the more resilient growth momentum in the US. Fixed Income Strategy: The big news announcements do not motivate us to change our fixed income investment recommendations. Stay below-benchmark on overall duration, and underweight the US in global bond portfolios. Stay overweight global inflation-linked bonds versus nominal government debt, particularly in the US and Italy. Maintain an overweight stance on global spread product, focused on US corporates (investment grade and Ba-rated high-yield) and emerging market US dollar denominated corporates. Feature Chart of the WeekUS Yields Leading The Way Higher Investors have digested two major pieces of news over the past few days – the projected election of Joe Biden as the 46th US President and the positive results of Pfizer’s COVID-19 vaccine trial. Both outcomes are bond-bearish, but the bigger response came after the news of a potential vaccine, with the 10-year US Treasury yield hitting an 8-month high of 0.96% yesterday. Yields in other countries rose by a lesser amount, continuing the recent trend of US Treasury underperformance (Chart of the Week). After the US election result, however, we remain comfortable with our recommended below-benchmark overall duration stance and underweight allocation to US Treasuries in global bond portfolios.  The introduction of a successful vaccine would obviously be a game-changer for all financial markets, not just fixed income, as it would allow investors to see an end to the pandemic and a return to more normal economic activity. While we are heartened by the vaccine trial announcement, there are still many hurdles that need to be cleared before any vaccine is approved and distributed around the world. It is still too soon to adjust our bond investment strategy in anticipation of a post-COVID world. After the US election result, however, we remain comfortable with our recommended below-benchmark overall duration stance and underweight allocation to US Treasuries in global bond portfolios. While a Biden victory combined with the Republicans likely keeping control of the US Senate was the least bond-bearish outcome - thus avoiding the big surge in government spending likely after a Democratic “blue wave” - there is clear upward momentum in US economic growth that suggests more upside for Treasury yields on both an absolute basis and relative to other countries. Cross-Country Divergences Are Starting To Appear Our recent decision to cut our recommended overall global duration stance to below-benchmark was motivated by our more bearish view on US Treasuries. However, a more defensive duration posture was justified by the rapid rebound in global growth seen since the depths of the COVID-19 recession. Our Global Duration Indicator, comprised of leading economic data, has been calling for a bottom in global bond yields toward the end of 2020 (Chart 2). The rise in global yields we are witnessing now appears to be right on cue. There are now more relative growth, inflation and policy divergences opening up that will allow country allocation to become a bigger source of outperformance for fixed income investors. Chart 2Global Yields Are Bottoming Importantly, inflation expectations across the developed world have yet not risen by enough to force central banks to become less dovish. This suggests that global yield curves will have a steepening bias over at least the next six months, with longer-term yields rising more on the back of faster growth (and additional increases in inflation expectations) than shorter-maturity yields which are more sensitive to monetary policy shifts. Those trends will not be seen equally across all countries, though. There are now more relative growth, inflation and policy divergences opening up that will allow country allocation to become a bigger source of outperformance for fixed income investors. For example, the October US manufacturing ISM and Payrolls data released last week showed robust strength, even in a month where new US COVID-19 cases rose sharply. Europe, on the other hand, has seen an even bigger surge in new cases, resulting in a wave of national lockdowns that has already begun to weigh on domestic economic activity. Thus, core European bond yields have remained stable, even with the euro area manufacturing PMI remaining elevated (Chart 3). We see similar divergences in other developed economies, with generally strong manufacturing PMIs and mixed responses from bond yields. When looking at the breakdown of nominal bond yields into the real yield and inflation expectations components, even more divergences are evident (Chart 4).1 Chart 3Mixed Responses To Rebounding Growth Chart 4Real Yield Trends Are Starting To Diverge Chart 5Discounting An Extended Period Of Negative Real Rates The real yields on benchmark 10-year inflation-linked bonds are slowly rising in the US and Canada, but remain stable in Germany, the UK and Australia. Market expectations for central bank policy rates, extracted from overnight index swap (OIS) curves, are currently priced for an extended period of low policy rates over the next few years. This is no surprise, as central banks have told the markets this would be the case via dovish forward guidance. Yet central banks are also projecting inflation rates to move higher between 2021 and 2023, even as they are signaling unchanged interest rates over that same period (Chart 5). Central banks are effectively telling markets that they want an extended period of negative real policy rates - a major reason why real bond yields are negative across the developed world. At some point, however, markets will begin to challenge the need for deeply negative real policy rates as economies recover from the COVID-19 shock to growth. Unemployment in the US and Canada has already declined sharply since spiking during the first wave of COVID-19 lockdowns. In the US, the unemployment rate has fallen from a peak of 14.7% to 6.9%; in Canada, the decline has been from 13.7% to 8.9% (Chart 6). This contrasts sharply to trends in Europe and Australia, where unemployment rates remain elevated. Chart 6Diverging Trends In Unemployment At some point, however, markets will begin to challenge the need for deeply negative real policy rates as economies recover from the COVID-19 shock to growth. With the Fed and Bank of Canada (BoC) projecting additional declines in unemployment over the next few years, markets are starting to discount a less dovish stance from both central banks. The US and Canadian OIS curves are now discounting one full 25bp policy rate hike by Aug 2023 and May 2023, respectively. This is a bit sooner than signaled by the forward guidance of the Fed and BoC. Thus, markets are now pricing in a less negative path for real policy rates – and, by association, real bond yields. Chart 7Markets Still Discounting Low Yields For Longer This contrasts to the euro area, Australia and the UK, where unemployment rates remain elevated. The recent surge in coronavirus cases across Europe means that the ECB and Bank of England will be under no pressure by markets to reconsider their current easy money policies. While in Australia, persistently weak inflation and, more recently, worries about an appreciating Australian dollar are keeping expectations for Reserve Bank of Australia (RBA) policy ultra-dovish. Given the likely hit to longer-term potential growth from the COVID-19 pandemic, coming at a time of elevated debt levels (both government and private), markets are justified in pricing in a structurally lower level of policy rates for longer (Chart 7). Yet even in such a world, there will be cyclical upswings in growth and inflation that will upward pressure on bond yields. At the moment, those pressures seem greatest in the developed world in the US and Canada. This suggests that global bond investors should underweight both the US and Canada. However, the Fed seems more willing to accept a period of rising bond yields than the BoC, which has been very aggressive in the expansion of its quantitative easing (QE) program, which leaves us to only consider the US as a recommended underweight. Bottom Line: Joe Biden’s apparent victory in the US presidential race, as well as the announcement of a potential successful COVID-19 vaccine trial, are both bond-bearish outcomes. This is especially so for US Treasuries given the more resilient growth momentum in the US. Recommended Fixed Income Strategy After A Busy Few Days Joe Biden’s election victory and the potential COVID-19 vaccine do not lead us to make any changes to our main fixed income investment recommendations, which generally have a pro-growth, pro-risk bias that would benefit from the reduction in US political uncertainty and, potentially, the beginning of the end of the pandemic. On duration, we continue to recommend a moderate below-benchmark overall exposure. Our main fixed income investment recommendations, which generally have a pro-growth, pro-risk bias that would benefit from the reduction in US political uncertainty and, potentially, the beginning of the end of the pandemic. On country allocation, we remain underweight the US, neutral Canada and Australia, and overweight the UK, core Europe, Italy, Spain and Japan. The country allocations are determined by each country’s sensitivity to changes in US Treasury yields, particularly during periods of rising yields. We are overweight the countries with a lower “yield beta” to changes in US yields. We view Italy and Spain as credit instruments, supported by large-scale ECB purchases and more fiscal cooperation within Europe. We are not recommending underweights to higher-beta Canada and Australia, however, with both the BoC and RBA being very aggressive with bond purchases (Chart 8). On credit, the backdrop remains very conducive to spread product outperformance versus government bonds, particularly with the monetary policy backdrop remaining highly accommodative (Chart 9). Chart 8Global QE Has Been Aggressive We expect some additional spread tightening for developed market corporate debt as well also emerging market US dollar denominated corporates. In terms of regions and credit tiers, we prefer US investment grade and Ba-rated high-yield to euro area credit. Chart 9Central Bank Liquidity Still Supportive For Global Credit Chart 10More Global QE Is Good For Inflation-Linked Bonds Finally, we continue to recommend overweight allocations to inflation-linked bods versus nominal government debt in the US, Italy and Canada. Central banks will continue to err on the side of maintaining stimulative monetary policy settings to keep financial conditions easy to support economic growth. That means no hawkish surprises on the interest rate front, while also continuing to buy bonds via quantitative easing (Chart 10) – reflationary policies that should help boost inflation expectations.   Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com Footnotes 1 We have deliberately left Japan out of this analysis, as the Bank of Japan’s Yield Curve Control policy has effectively short-circuited the link between Japanese economic growth, inflation and bond yields. Recommendations The GFIS Recommended Portfolio Vs. The Custom Benchmark Index Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

Highlights Economy: The Democrats did not sweep the US election, but the Democratic House and Republican Senate will likely find some middle ground on a stimulus deal. This will keep the economic recovery on track. A highly effective COVID vaccine that becomes widely available would supercharge it. Rates: Investors should continue to position for a bear-steepening nominal Treasury curve on a 6-12 month horizon. They should also overweight TIPS versus nominal Treasuries, favor inflation curve flatteners and real yield curve steepeners. Treasury Supply & Fed Demand: The Treasury department will continue to increase coupon issuance at the expense of bills. If current policy remains in place, Fed purchases will decline as a percent of coupon issuance in the coming quarters. But the Fed could take steps to modify its asset purchase strategy during the next few months. Feature We’ve seen enough. After a week of checking and re-checking the numbers, BCA’s US Bond Strategy service has concluded that a moderate bear-steepening of the Treasury curve remains the most likely outcome for the next 6-to-12 months. Of course, the dust has not completely settled on the US election. President Trump has issued several legal challenges and control of the Senate won’t be decided until early January when two run-off elections take place in Georgia. However, it now looks safe to assume that Joe Biden will take over as President next year. We also expect, with slightly less conviction, that the Republican party will keep control of the Senate, as Democrats need to win both Georgia races in order to secure a majority. In this week’s report, we assess the fall-out from the election on different sectors of the US bond market. We also consider how the election result impacts the outlook for fiscal stimulus and provide an update on supply and demand trends in the Treasury market. The Election Fall-Out In Bond Markets Nominal Yields Chart 1A Moderate Bear-Steepening The 10-year Treasury yield got as high as 0.90% in advance of election day, as the market was pricing-in a Democratic sweep that would have led to a substantial increase in government spending. This outcome is still technically possible, but it now looks unlikely. The 10-year Treasury yield fell back to 0.78% as the election results came in but returned to 0.90% yesterday morning on news that Pfizer’s COVID vaccine was 90% effective in phase 3 trials. This market action affirms our recommended positioning. The Bloomberg Barclays Treasury Index has been underperforming cash since August, the uptrend in the 10-year Treasury yield remains intact and the yield curve continues to steepen (Chart 1). All these trends will remain in place as long as the economic recovery continues, and timely distribution of an effective COVID vaccine will certainly speed that process up. The biggest risk to our view is that a Democrat-controlled House and Republican-controlled Senate are unable to agree on a follow-up fiscal stimulus package during the next few months, and that the economic recovery stalls as a result. This is possible, but our base case scenario is that a compromise will be easier to reach now that the election is over. We expect a moderately-sized relief bill to be delivered relatively soon, possibly even before the end of the year. The Inflation Curve Chart 2Own Inflation Curve Flatteners... The 10-year TIPS breakeven inflation rate fell 8 bps immediately after the election, but unlike with nominal yields, the trend in the cost of inflation compensation had been relatively flat heading into election day (Chart 2). It’s not hard to see why. Inflationary pressures in the economy have clearly moderated compared to the summer. The oil price has taken a step down (Chart 2, panel 3) and month-over-month CPI growth has been trending lower (Chart 2, bottom panel). We don’t expect this deceleration in inflation to continue. Global economic recovery will keep commodities well bid, and core inflation will slowly recover back to target. This argues for staying overweight TIPS versus nominal Treasuries. We also recommend owning inflation curve flatteners. The inflation curve has been steepening since August, as the short-dated cost of inflation compensation has fallen by more than the long-dated cost (Chart 2, panel 2). This steepening is typical for periods when TIPS breakeven rates are falling, and it will reverse when breakevens start rising again. Looking further out, the Fed’s commitment to allow a temporary overshoot of its 2% inflation target means that we should expect the inflation curve to invert. This means that inflation curve flatteners have a lot of room to run. Real Yields With almost no volatility in short-maturity nominal yields, short-maturity real yields are simply the mirror image of short-maturity inflation expectations. For this reason, the 2-year real yield has been moving up since August as the 2-year cost of inflation compensation has declined (Chart 3). This dynamic doesn’t hold for long maturities, where nominal yields have been rising as markets price-in eventual Fed tightening (Chart 3, top panel). Inflationary pressures in the economy have clearly moderated compared to the summer. The different behavior of long- and short-maturity real yields gives us high conviction in recommending a real yield curve steepener (Chart 3, bottom panel). A recovery in inflation expectations will push short-maturity real yields lower but will not have the same impact at the long-end where nominal yields will also rise. Chart 3…And Real Yield Curve Steepeners Corporate Credit Chart 4Credit Spreads Welcomed The Election Results Interestingly, the election result of a Biden presidency and divided House and Senate was viewed positively by both the “risk-free” Treasury market and risky credit spreads (Chart 4). Treasury yields fell on expectations of less fiscal stimulus, but credit spreads also tightened because a Republican Senate will keep corporate tax hikes at bay and a Biden presidency will ratchet down trade tensions with China. We maintain our positive outlook on credit and continue to recommend overweight allocations to corporate bonds rated Ba and higher. We remain underweight low-rated junk bonds (B & below) for now, because those spreads are pricing-in a rapid drop in the default rate. We may soon shift into low-rated junk as well, depending on how quickly an effective vaccine can be distributed. One less discussed risk for corporate spreads is the expiration of the Fed’s emergency lending facilities at the end of the year. The facilities are currently scheduled to expire on December 31st, though Fed Chair Powell seemed to imply last week that he would like to extend them. The one hitch could be that the Treasury department will also have to sign-on to an extension. It is currently unclear whether it is interested in doing so. Municipal Bonds Chart 5Munis Still Very Attractive The strong relative performance of municipal bonds since election day has been the most confounding market move (Chart 5). All logic tells us that Municipal / Treasury yield spreads should have widened as it became clear that the Republicans will likely keep control of the Senate. A Republican Senate will prevent Joe Biden from raising income taxes, which would have made tax-advantaged munis look more attractive on a relative basis. A Republican Senate has also been staunchly opposed to providing federal aid to cash-strapped state & local governments. But munis have outperformed Treasuries despite these obvious negative catalysts, possibly in part due to Mitch McConnell’s post-election comments that suggested he is open to compromise on a fiscal relief bill and would even be open to including some funds for state & local governments. Despite McConnell’s comments, the prospect of federal funds for state & local governments is uncertain at best. But we nonetheless maintain an overweight allocation to municipal bonds due to continued extraordinary valuations relative to both Treasuries (Chart 5, panel 2) and corporates (Chart 5, bottom panel). The Stimulus Risk As we alluded to above, the biggest risk to our bond-bearish view is that the failure to pass a follow-up to the CARES act results in a spate of negative economic data that spooks investors. This negative data would likely first show up in consumer spending, which so far continues to recover (Chart 6). However, we think the odds are that, in the absence of stimulus, we will see a disappointing consumer spending report within the next few months. How markets react to that news will depend on the status of stimulus talks at that time, as well as news about a potential vaccine roll-out. Disposable personal income was still above pre-COVID levels in September, but it continues to be buttressed by income support from the federal government. Notice that non-transfer income remains below pre-COVID levels (Chart 6, panel 3). Further, the drop in the savings rate during the past few months has outpaced the improvement in consumer sentiment (Chart 6, bottom panel). This suggests that any excess savings that households may have accumulated in the spring are now close to being exhausted. In the absence of stimulus, we will see a disappointing consumer spending report within the next few months. Elsewhere, the labor market continues to make steady improvements, but it could also use some help from policymakers. Excluding temporary census employment, nonfarm payrolls rose 786k in October, slightly above September’s pace but below the rapid gains seen in May and June (Chart 7). Further, initial jobless claims remain above 700k per week and real-time employment data from Homebase has been steady at a low level. All this to say that the labor market is making only modest gains and there remains a gap of 10 million jobs between current nonfarm payrolls and those from February (Chart 7, top panel). Chart 6Keep Monitoring Consumer Spending Chart 7A Slow Recovery In Employment The bottom line is that, without further fiscal stimulus, the odds are high that the economic data will disappoint at some point during the next few months. This will cause bond yields to fall and credit spreads to widen, unless it looks like Congress is close to a deal or like a vaccine will be available in a timely manner. Fortunately, we do think the odds are relatively high that a Republican Senate and Democratic House will be able to reach a compromise stimulus deal, albeit a modest one in the range of $700 billion to $1 trillion. The political incentives against compromise have faded now that the election is over, and we expect a deal either this year or early next year. Treasury Supply And Fed Demand The Treasury department recently released its financing estimates for the next two quarters. A few trends are worth mentioning. First, the Treasury will continue to increase coupon issuance as it seeks to extend the average maturity of the outstanding debt (Chart 8, top panel). Chart 8The Path For Treasury Supply And Fed Demand Second, the Treasury will continue to operate with an historically elevated cash balance, but it will seek to reduce its cash holdings to $800 billion from $1.6 trillion currently (Chart 9). Chart 9Treasury Will Deploy Some Cash Third, the Treasury assumed in its projections that Congress will deliver another $1 trillion of stimulus. The combination of (i) increased coupon issuance, (ii) a falling cash balance and (iii) stimulus projections that may be too high, points to a continued drop in T-bill issuance (Chart 10). In fact, the Treasury acknowledged that bill issuance will likely fall going forward and said that it would be comfortable with a distribution where bills account for 15%-20% of the outstanding debt (Chart 11). Chart 10Expect T-Bill Issuance To ##br##Keep Falling… Chart 11…And To Settle At Around 15-20% Of Outstanding Debt Fed Chairman Powell also addressed the media last week, after the conclusion of the November FOMC meeting, and announced that the Fed made no changes to its asset purchases. For the time being, the Fed will continue to purchase “at least” $80 billion of Treasuries and $40 billion of MBS per month. However, Powell did indicate that FOMC participants discussed different ways in which they might modify the asset purchase program in the future. Presumably this means that if the Committee feels the need to deliver further monetary stimulus it will do so by either shifting its Treasury purchases to the long-end of the curve – in order to remove more duration risk from the market – or by increasing the outright pace of purchases. Powell made it clear that he sees these sorts of balance sheet moves as viable forms of monetary stimulus, though the tone of the questions he received during the press conference suggests that the consensus increasingly senses that the Fed might be out of ammo. Several questioners noted Powell’s repeated calls for fiscal stimulus and asked directly whether the Fed has done all it can. In conclusion, if the Fed maintains the current pace and distribution of Treasury purchases (Chart 8, panel 2), its asset purchases will continue to trend down compared to gross Treasury issuance (Chart 8, bottom panel). However, we could see the Fed taking a step to mitigate that decline at the long-end of the curve by shifting the maturity distribution of its asset purchases towards longer maturities. This move could occur as early as next month. The Treasury will continue to operate with an historically elevated cash balance, but it will seek to reduce its cash holdings to $800 billion from $1.6 trillion currently. The bar for actually increasing the monthly pace of purchases is likely much higher, and it would require a significant tightening of financial conditions or drop in economic activity to push the Fed into action. The bigger question, however, is whether the market even cares anymore about tweaks to the Fed’s asset purchase program. The tone of questions at last week’s press conference suggests it might not. Appendix: Buy What The Fed Is Buying The Fed rolled out a number of aggressive lending facilities on March 23. These facilities focused on different specific sectors of the US bond market. The fact that the Fed has decided to support some parts of the market and not others has caused some traditional bond market correlations to break down. It has also led us to adopt of a strategy of “Buy What The Fed Is Buying”. That is, we favor those sectors that offer attractive spreads and that benefit from Fed support. The below Table tracks the performance of different bond sectors since the March 23 announcement. We will use this to monitor bond market correlations and evaluate our strategy’s success. TablePerformance Since March 23 Announcement Of Emergency Fed Facilities Ryan Swift US Bond Strategist rswift@bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification