Yield Curve

Highlights Treasuries: Bond investors should maintain below-benchmark portfolio duration and continue to short the 5-year note versus a duration-matched 2/10 barbell. For those investors who want to take an outright long position in US Treasuries, the 2-year Treasury note looks like the best security to choose. Municipal Bonds: This week we upgrade our recommended allocation to municipal bonds from overweight (4 out of 5) to maximum overweight (5 out of 5). Investors who can take advantage of the muni tax exemption should favor municipal bonds over Treasuries and over corporate bonds with the same credit rating and duration. In particular, we recommend that investors focus on long-maturity municipal bonds. Fed: Given our view that inflation will fall during the next 12 months, we still view December 2022 as the most likely liftoff date. However, we will continue to monitor our Five Factors For Fed Liftoff to see if our forecast needs to be revised. Feature Chart 1 Our call for a bear-flattening of the US Treasury curve has worked out well during the past few weeks. Long-maturity Treasury yields have almost risen back to their March highs, and the short-end of the curve has also participated in the recent bout of selling (Chart 1). In light of these moves, it makes sense to re-evaluate our nominal Treasury curve positioning. First, we consider whether, at current yield levels, it still makes sense to run below-benchmark portfolio duration. Second, we consider whether our current recommended yield curve trade (short the 5-year note versus a duration-matched 2/10 barbell) remains the best way to extract returns from changes in the yield curve’s shape. The next section of this report answers these questions by looking at forecasted returns for different Treasury maturities across a variety of plausible economic and monetary policy scenarios. Later in the report we look at municipal bond valuation and provide a quick update on last week’s Fedspeak. Forecasting Treasury Returns Chart 2 Three sources of Treasury bond return need to be considered when creating a forecast. Income Return: The return earned from the bond’s coupon payments. Rolldown Return: The return that a bond accrues simply by moving closer to its maturity date in an unchanged yield curve environment. Capital Gains/Losses: The return earned by a bond due to changes in the level and slope of the yield curve. We like to combine the income and rolldown return into one measure called “carry”. The carry can be thought of as the return an investor will earn in a specific bond if the yield curve remains unchanged throughout the investment horizon. Though carry is not the be all and end all of bond returns, it can be illuminating to look at the yield curve in terms of carry instead of the typical yield-to-maturity. Chart 2 shows the usual par coupon yield curve alongside the 12-month carry for each Treasury security. At present, the steepness of the 3-7 year part of the curve means that bonds of those maturities benefit a lot from rolldown. In fact, we see that a 7-year Treasury note will earn more than a 10-year Treasury note during the next 12 months if the curve remains unchanged. After calculating carry, the next step is to calculate capital gains/losses for each bond. To do this, we create some possible scenarios for future changes in the fed funds rate and assume that the yield curve moves to fully price-in that funds rate path over the course of a 12-month investment horizon.1  Next, we calculate the capital gains/losses for each bond based on the new shape of the yield curve in each scenario. Tables 1A-1D show the results from four different scenarios where the Fed starts to lift rates in December 2022. We then assume that the Fed will lift rates at a pace of 75-100 bps per year and that the funds rate will level-off at a terminal rate of either 2.08% or 2.58%. The 2.08% terminal rate corresponds to the median estimate of the long-run neutral fed funds rate from the New York Fed’s Survey of Market Participants. The 2.58% terminal rate corresponds to the median forecast from the Fed’s Summary of Economic Projections.2  Chart Chart Chart Chart The scenario shown in Table 1B is the closest to our base case. In this scenario, some short-maturity bonds deliver positive returns, but returns are negative for the 5-year maturity and beyond. Also, the 5-year note delivers the worst total return of all the maturities we examine. Unsurprisingly, expected returns for the longer maturities drop significantly if we raise our terminal rate assumption to 2.58% (Tables 1C & 1D). Therefore, any call to short the 5-year note versus the long-end relies on an assumption that the market will trade as though the terminal rate is closer to 2% than to 2.5% during the next 12 months. This is in line with our expectation. Finally, we observe that slowing our pace assumption from 100 bps per year to 75 bps raises expected returns across the board, but the 5-year still performs worse than the other maturities (Table 1A). Due to our expectation that inflation will fall during the next 12 months, a December 2022 liftoff remains our base case.3  However, the market has recently moved to price-in an earlier start to rate hikes. As of last Friday’s close, the fed funds futures curve was priced for liftoff in September 2022 and for a total of 49 bps of tightening by the end of 2022 (Chart 3). Chart 3Market Priced For September 2022 Liftoff Market Priced For September 2022 Liftoff Market Priced For September 2022 Liftoff Tables 2A-2D incorporate these recent market moves into our forecast by looking at the same scenarios as in Tables 1A-1D but assuming a September 2022 liftoff instead of December. The results are not all that different. Expected returns are worse across the board, but the 5-year still looks like the worst spot on the curve unless the market starts to price-in a higher terminal rate. Chart Chart Chart Chart Investment Conclusions Most of the scenarios we examined had negative expected returns for most maturities. We therefore still think it makes sense to keep portfolio duration low. Further, in every scenario the best expected returns can be found in the shorter maturities. In fact, the 2-year Treasury note offers positive returns in every scenario we examined. An outright long position in the 2-year Treasury note looks like a decent trade for investors forced to hold bonds. As for the yield curve, our results suggest that we should continue with our current positioning: short the 5-year note versus a duration-matched 2/10 barbell. The 5-year note performs worst in every scenario that assumes a 2.08% terminal rate. While it’s conceivable that investors will eventually push their terminal rate expectations higher, we think this is more likely to occur once the Fed has already lifted rates a few times. Bottom Line: Bond investors should maintain below-benchmark portfolio duration and continue to short the 5-year note versus a duration-matched 2/10 barbell. For those investors who want to take an outright long position in US Treasuries, the 2-year Treasury note looks like the best security to choose. The Duration Drift In Municipal Bond Valuations One under-discussed aspect of municipal bonds is that the securities tend to pay higher coupons than other bonds. That is, the bonds will often be issued with coupon rates well above prevailing yields. Investors therefore must pay a higher price to purchase the bonds, but they receive more return in the form of coupon payments. This feature of municipal bonds has important implications for how we should value them. For example, while the average maturity of the Municipal Bond index is much higher than the average maturity of the Treasury index, the muni index’s higher coupon rate makes its average duration significantly lower (Chart 4). This means that any valuation measure that compares a municipal bond’s yield with the yield of another bond with the same maturity will be unflattering for the muni. Chart 4Munis Pay High Coupons, Have Low Durations Munis Pay High Coupons, Have Low Durations Munis Pay High Coupons, Have Low Durations Further, since Treasury securities and corporate bonds tend to issue at par, the coupon rates paid by those securities have fallen alongside yields during the past few decades. Meanwhile, municipal bond coupons have been relatively stable (Chart 4, panel 3). This means that, over time, municipal bond durations have fallen significantly compared to the durations of other US bond sectors. A fair valuation measure would compare municipal bond yields with equivalent-duration Treasury yields and that is exactly what we’ve done. Chart 5A shows the spread between General Obligation (GO) muni bond yields and equivalent-duration Treasury yields. Chart 5B shows the spreads expressed as percentile ranks. For example, a percentile rank of 50% means that the spread is at its historical median, a percentile rank of 10% means the spread has only been tighter 10% of the time. Chart 5AGO Muni/Treasury Spreads I GO Muni/Treasury Spreads I GO Muni/Treasury Spreads I Chart 5BGO Muni/Treasury Spreads II GO Muni/Treasury Spreads II GO Muni/Treasury Spreads II The first thing that jumps out from our analysis is that municipal bonds are not that expensive. Shorter-maturity spreads were tighter than current levels as recently as 2019/20 and the long-maturity (17-year+) spread is positive, despite the muni tax exemption. In terms of percentile rank, spreads for all GO maturity buckets are only just below the historical median. However, spreads traded much tighter prior to the 2008 financial crisis and it may not be reasonable to expect munis to return to those tight mid-2000 valuations. Charts 6A and 6B repeat the exercise from Charts 5A and 5B but for Revenue bonds instead of GOs. The message is similar. Muni valuations are not that stretched compared to history, and investors can earn a before-tax spread pick-up in munis versus Treasuries if they focus on the long maturities. Chart 6ARevenue Muni/Treasury Spreads I Revenue Muni/Treasury Spreads I Revenue Muni/Treasury Spreads I Chart 6BRevenue Muni/Treasury Spreads II Revenue Muni/Treasury Spreads II Revenue Muni/Treasury Spreads II In fact, municipal bonds offer a before-tax yield advantage versus Treasuries for Revenue bonds beyond the 12-year maturity point and for GO bonds beyond the 17-year maturity point. Further, the breakeven tax rate for 12-17 year GOs versus Treasuries is a mere 1% and the breakeven tax rate for 8-12 year Revenue bonds is only 8%. Investors facing a tax rate above the breakeven rate will earn an after-tax yield pick-up in munis versus duration-matched Treasuries (Table 3). Table 3Muni/Treasury And Muni/Credit Yield Ratios The Best & Worst Spots On The Yield Curve The Best & Worst Spots On The Yield Curve Of course, municipal bonds also carry a small credit risk premium relative to duration-matched Treasuries. The GO and Revenue indexes have average credit ratings of Aa1/Aa2 and Aa3/A1, respectively, compared to a Aaa rating for US Treasuries. But we can control for credit risk as well by comparing municipal bonds to the US Credit Index and matching both the duration and credit rating. Even this comparison looks favorable for municipal bonds. Once again, long-maturity munis offer a before-tax yield advantage compared to credit rating and duration-matched US Credit. Meanwhile, breakeven tax rates for other maturities are low enough to attract most investors. Bottom Line: This week we upgrade our recommended allocation to municipal bonds from overweight (4 out of 5) to maximum overweight (5 out of 5). Investors who can take advantage of the muni tax exemption should favor municipal bonds over Treasuries and over corporate bonds with the same credit rating and duration. In particular, we recommend that investors focus on long-maturity municipal bonds, noting that the relatively low duration of these bonds makes them attractive relative to other bonds with similar risk profiles. Five Fed Factors A lot of Fedspeak hit the tape last week. Of particular interest were an interview with Chair Jay Powell on Friday and speeches by Fed Governors Randy Quarles and Chris Waller on Wednesday and Tuesday. One takeaway from their remarks is that a tapering announcement at the next FOMC meeting is very likely, with net asset purchases expected to hit zero by the middle of next year. The market, however, seems to have already taken the taper announcement on board. The more interesting aspects of the speeches were the discussions about how the Fed will decide when to lift rates and how elevated inflation readings may or may not influence that decision. We’ve noted in prior reports that five factors will determine when the Fed finally decides to lift rates, and last week’s comments gave us confidence that we’re on the right track. We run through our Five Factors For Fed Liftoff below, with some additional comments on why each factor is important (Table 4). Table 4Five Factors For Fed Liftoff The Best & Worst Spots On The Yield Curve The Best & Worst Spots On The Yield Curve 1. The Unemployment Rate The Fed has officially pledged through its forward guidance not to lift rates until “maximum employment” is reached. While the exact definition of “maximum employment” can be debated, there is widespread agreement that it includes an unemployment rate below its current adjusted level of 4.9%.4 More specifically, we inferred from the September Summary of Economic Projections that most FOMC participants view an unemployment rate of around 3.8% as consistent with “maximum employment” (Chart 7).5 Chart 7Defining "Maximum Employment" Defining "Maximum Employment" Defining "Maximum Employment" We expect that the Fed will refrain from lifting rates until the unemployment rate reaches 3.8%. 2. Labor Force Participation We explored the debate about labor force participation in a recent report.6 In short, there are some policymakers who believe that “maximum employment” cannot be achieved until the labor force participation rate has returned to pre-COVID levels. There are others, however, who think that an aging population and the recent uptick in retirements make such a return impossible. Randy Quarles, for example: I expect that as conditions normalize, [the labor force participation rate] will pick up, but it is unlikely to return to its February 2020 level. One reason is that a disproportionate number of older workers responded to the initial shock of the COVID event by retiring, which may be an area where participation and employment struggle to retrace lost ground.7 In his speech, Governor Waller also mentioned “2 million jobs” that will be lost forever due to retirements.8 While many policymakers cite increased retirements as a reason why the overall labor force participation rate will remain permanently lower, there is much broader agreement that a reasonable definition of “maximum employment” should include the prime-age (25-54) labor force participation rate being much closer to its February 2020 level (Chart 7, bottom panel). We think the Fed will refrain from lifting rates until the prime-age (25-54) labor force participation rate is close to its February 2020 level. 3.  Wage Growth Accelerating wages are a tried-and-true signal that the labor market is running hot. While wage growth is rising quickly right now (Chart 8), there is a strong sense that this is due to pandemic-related labor supply shortages and that wage growth will moderate as pandemic fears (and labor shortages) wane. Chart 8Wage Growth Wage Growth Wage Growth What will be more important is what wage growth looks like when the unemployment rate is close to the Fed’s target of 3.8%. At that point, accelerating wages will give the Fed a strong signal that a 3.8% unemployment rate really does constitute “maximum employment”. 4.  Non-Transitory Inflation Of our five factors, this is admittedly the most difficult to pin down. However, Governor Quarles did a good job of explaining non-transitory inflation in last week’s speech: The fundamental dilemma that we face at the Fed now is this: Demand, augmented by unprecedented fiscal stimulus, has been outstripping a temporarily disrupted supply, leading to high inflation. But the fundamental productive capacity of our economy as it existed just before COVID – and, thus, the ability to satisfy that demand without inflation – remains largely as it was, constraining demand now, to bring it into line with a transiently interrupted supply, would be premature. Essentially, Quarles is saying that the Fed does not want to respond to a pandemic-related supply shock by lifting rates and curtailing aggregate demand. The Fed only wants to tighten policy if it sees an increase in broad-based inflationary pressures that will not be contained naturally by a return to more normal aggregate supply conditions. Accelerating wages would be one signal of such broad-based inflationary pressures, as would be measures of core inflation excluding those sectors that have been most impacted by the pandemic supply disruptions (Chart 9). Lastly, we could also look at indicators of inflation’s breadth across its different components, which have recently spiked to concerning levels (Chart 10). Chart 9Non-Covid Inflation Non-Covid Inflation Non-Covid Inflation Chart 10CPI Breadth Has Spiked CPI Breadth Has Spiked CPI Breadth Has Spiked 5.  Inflation Expectations Inflation expectations are also critical to monitor. While all Fed participants seem to agree that inflation will fall during the next year, there is also widespread agreement that if high inflation causes inflation expectations to rise to uncomfortably high levels, then the Fed will be forced to act. Chris Waller: A critical aspect of our new framework is to allow inflation to run above our 2 percent target (so that it averages 2 percent), but we should do this only if inflation expectations are consistent with our 2 percent target. If inflation expectations become unanchored, the credibility of our inflation target is at risk, and we likely would need to take action to re-anchor expectations at our 2 percent target. At present, inflation expectations remain well-anchored near levels consistent with the Fed’s target (Chart 11). In particular, we like to track the 5-year/5-year forward TIPS breakeven inflation rate targeting a range of 2.3% to 2.5% as consistent with the Fed’s target. Incidentally, Governor Waller also flagged TIPS breakeven inflation rates as his “preferred” measure of inflation expectations in last week’s speech.  Chart 11Inflation Expectations Remain Well-Anchored Inflation Expectations Remain Well-Anchored Inflation Expectations Remain Well-Anchored The Fed will move much more quickly toward rate hikes if the 5-year/5-year forward TIPS breakeven inflation rate moves above 2.5%. Bottom Line: Given our view that inflation will fall during the next 12 months, we still view December 2022 as the most likely liftoff date. However, we will continue to monitor our Five Factors For Fed Liftoff to see if our forecast needs to be revised.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 All of our scenarios use a 12-month investment horizon and assume a term premium of 0 bps. 2 In both cases we assume that the fed funds rate trades 8 bps above its lower-bound, as is currently the case. 3 Please see US Bond Strategy Weekly Report, “Right Price, Wrong Reason”, dated October 19, 2021. 4 We adjust the unemployment rate for distortions in the number of people employed but absent from work. Please see US Bond Strategy Weekly Report, “Overreaction”, dated July 13, 2021 for further details. 5 Please see US Bond Strategy Weekly Report, “Damage Assessment”, dated September 28, 2021. 6 Please see US Bond Strategy Weekly Report, “2022 Will Be All About Inflation”, dated September 14, 2021. 7 https://www.federalreserve.gov/newsevents/speech/quarles20211020a.htm 8 https://www.federalreserve.gov/newsevents/speech/waller20211019a.htm Recommended Portfolio Specification Other Recommendations Treasury Index Returns Spread Product Returns

Highlights Duration: We recommend that investors run below-benchmark portfolio duration in US bond portfolios on the expectation that the Treasury curve will bear-flatten between now and Fed liftoff in December 2022. Nominal Treasury Curve: We recommend positioning for curve flattening by going short the 5-year Treasury note versus a duration-matched barbell consisting of the 2-year and 10-year notes. TIPS: Investors should position for higher short-maturity real yields. This can be done through an outright short position in 2-year TIPS, an inflation curve steepener or a real yield curve flattener. The Long And Short Of It Chart 1Short-End Joins The Selloff Short-End Joins The Selloff Short-End Joins The Selloff It’s still a bit early for a 2021 retrospective, but unless something dramatic happens during the next 2 ½ months it’s likely that the year will go into the books as a bad one for US bonds. Looking back, we can identify three phases of bond market performance in 2021. First, a selloff in long-dated bonds early in the year driven by economic re-opening and fiscal stimulus. Second, a partial reversal of this long-end selloff that lasted through the spring and early summer. Finally, a renewed selloff involving both the long and short ends of the yield curve (Chart 1). The Long End Looking first at the long end of the curve, we don’t see any immediate signs that yields have risen too far. Estimates of the 10-year term premium created by taking the difference between the spot 10-year Treasury yield and survey estimates of the future 10-year average fed funds rate show that the term premium is not as elevated as it was when yields peaked last March or when they peaked in 2018 (Chart 2). The 25-delta risk reversal on 30-year Treasury futures – a technical indicator with a strong track record of calling turning points in the 30-year yield – also remains below the 1.5 level that has historically signaled a peak in the 30-year yield (Chart 3). Table 1 shows that while it is rare for the risk reversal to rise above 1.5, such a move usually indicates that yields have risen too far, too fast Chart 210-Year Term Premium Still Low 10-Year Term Premium Still Low 10-Year Term Premium Still Low Chart 3Technicals Not Stretched Technicals Not Stretched Technicals Not Stretched Table 1Track Record Of Risk Reversal Indicator Right Price, Wrong Reason Right Price, Wrong Reason Finally, we look at the 5-year/5-year forward Treasury yield relative to a range of survey estimates of the long-run neutral fed funds rate (Chart 4). At 2.09%, the 5-year/5-year yield is close to median survey estimates of the long-run neutral fed funds rate.1 We take this to mean that the 5y5y yield has limited upside. Further increases in yields will take the form of the rest of the curve catching up to the 5y5y. Put differently, further increases in yields are more likely to coincide with curve flattening, not steepening.2  Chart 45y5y Is At Its Fair Value 5y5y Is At Its Fair Value 5y5y Is At Its Fair Value The Short End While long-maturity bond yields have moved up during the past few months, it is the breakout in short-maturity Treasury yields that has been the most notable feature of the recent bond selloff (Chart 1, bottom panel). In particular, near-term interest rate expectations have adjusted sharply higher since the September FOMC meeting (Chart 5). Prior to the September FOMC meeting, the overnight index swap (OIS) market was priced for Fed liftoff in February 2023 and for a total of 80 bps of rate hikes by the end of 2023. Now, the OIS curve is priced for Fed liftoff in September 2022 and for a total of 113 bps of rate hikes by the end of 2023. Chart 5Fed Funds Rate Expectations Fed Funds Rate Expectations Fed Funds Rate Expectations We continue to view the December 2022 FOMC meeting as the most likely date for the first rate hike. We also think it’s reasonable to expect the Fed to lift rates at a pace of 75-100 bps per year once tightening begins. In other words, we view fair pricing at the front-end of the curve as consistent with liftoff in December 2022 and a total of 100-125 bps of rate hikes by the end of 2023. The recent selloff has made front-end pricing more consistent with our assessment of fair value. Therefore, we don’t see any huge opportunities for directional bets on short-dated nominal yields. That said, we also contend that the bond market has arrived at the correct conclusion about the near-term pace of Fed tightening, but for the wrong reason. As is discussed in the next section of this report (see section titled “Massive Upside In Short-Maturity Real Yields”), this presents some attractive opportunities to trade short-maturity real yields and short-maturity inflation breakevens. One other observation from Chart 5 is that the market’s expected pace of Fed tightening flattens off considerably in 2024 and beyond. The market is priced for a mere 34 bps of tightening in 2024 and 2025 and the fed funds rate is still expected to be below 1.6% by the end of 2025. This highlights that, while pricing at the front-end of the yield curve looks reasonable, yields with slightly longer maturities remain too low. Bottom Line: We recommend that investors run below-benchmark portfolio duration in US bond portfolios on the expectation that the Treasury curve will bear-flatten between now and Fed liftoff in December 2022. We recommend positioning for curve flattening by going short the 5-year Treasury note versus a duration-matched barbell consisting of the 2-year and 10-year notes. Massive Upside In Short-Maturity Real Yields Table 2Yield Changes Since September FOMC (BPs) Right Price, Wrong Reason Right Price, Wrong Reason The prior section noted that the market’s near-term rate expectations have risen considerably during the past few weeks. While we think that pricing looks reasonable compared to our own monetary policy expectations, we alluded to the idea that the market has brought forward its rate hike expectations for the wrong reason. Table 2 illustrates what we mean. Practically all the increase in nominal Treasury yields since the September FOMC meeting has been driven by a rising cost of inflation compensation. Real yields, on the other hand, have either been relatively stable (for long maturities) or have fallen massively (at the short-end of the curve). In fact, the 2-year real yield has declined 34 bps since the September FOMC meeting even as the 2-year nominal yield has increased by 16 bps. What the stark divergence between real yields and the cost of inflation compensation tells us is that the market is concerned that inflation may not fall as much as was previously assumed and the Fed may be forced to tighten more quickly in response. First off, we think concerns about persistently high inflation are a tad overblown. It’s certainly true that 12-month headline and core CPI inflation remain extremely high, at 5.4% and 4.0% respectively, but 3-month rates of change have moderated during the past few months and the 12-month figures will soon follow suit (Chart 6). Second, even if inflation is slow to moderate, the composition of what is driving that high inflation has implications for how the Fed will respond. Specifically, if elevated inflation continues to be driven by extreme readings from a few sectors that have been inordinately impacted by the pandemic, the Fed will be inclined to write-off that inflation as “transitory” while it awaits more broad-based inflationary pressures driven by tight labor markets and accelerating wages. It continues to be worth noting that after stripping out COVID-impacted services and cars, core inflation remains well contained near levels consistent with the Fed’s target (Chart 7). Chart 6Inflation Is Falling Inflation Is Falling Inflation Is Falling Chart 7Inflation Pressures Are Narrow Inflation Pressures Are Narrow Inflation Pressures Are Narrow In a speech last week, Atlanta Fed President Raphael Bostic said that the Fed should use the word “episodic” instead of “transitory” to describe the nature of the current inflationary shock.3 The problem with the word “transitory” is that it is linked to a notion of time. It implies that inflation pressures are expected to fade quickly, but this is not the message that the Fed meant to convey with that word. Rather, in Bostic’s words, the Fed meant to convey that “these price changes are tied specifically to the presence of the pandemic and, once the pandemic is behind us, will eventually unwind, by themselves, without necessarily threatening longer-run price stability.” In other words, the Fed will not tighten policy to lean against narrow inflationary pressures driven by a few sectors that can easily be traced back to the pandemic. Rather, the Fed will only respond if inflationary pressures are sufficiently broad and/or if long-run inflation expectations become un-anchored to the upside. On the first point, there is some evidence that inflation pressures are broadening. As of September, 49% of the CPI index was growing at a 12-month rate above 3%, up from a 2021 low of 22% (Chart 8). However, long-run inflation expectations remain well-anchored near the Fed’s target levels (Chart 9). Chart 8CPI Breadth Indicator CPI Breadth Indicator CPI Breadth Indicator Chart 9Long-Term Inflation Expectations Long-Term Inflation Expectations Long-Term Inflation Expectations Our sense is that inflationary pressures will fade during the next 12 months as pandemic fears abate. Long-dated inflation expectations will remain close to current levels, but short-dated inflation expectations will fall. The Fed will start to lift rates in December 2022 as broad-based inflationary pressures emerge, but inflation will be only slightly above the Fed’s target by then. The best way to position for this outcome is to go short 2-year TIPS. The cost of 2-year inflation compensation will fall as inflation moderates during the next 12 months, but the nominal 2-year yield will rise modestly as we advance toward a Fed tightening cycle. These two factors will combine to drive the 2-year real yield sharply higher (Chart 10). If you prefer not to put on an outright short 2-year TIPS position, there are a few other ways to position for the same trend. First, investors could position for a steeper inflation curve. Chart 11 shows that the cost of short-maturity inflation compensation is much further above the Fed’s target level than the cost of long-maturity inflation compensation. Further, Table 3 shows that monthly changes in the cost of short-maturity inflation compensation are more sensitive to CPI than are changes in the long-maturity cost of inflation compensation. This means that the inflation curve will steepen during the next 12 months as inflation moderates and the short-term cost of inflation compensation falls. Chart 10Short 2-Year TIPS Short 2-Year TIPS Short 2-Year TIPS Chart 11Position For Inflation Curve Steepening... Position For Inflation Curve Steepening... Position For Inflation Curve Steepening... Table 3Regression of Monthly Changes In CPI Swap Rate Versus Monthly Changes In 12-Month Headline CPI Inflation (2010 - Present) Right Price, Wrong Reason Right Price, Wrong Reason   Second, you could also position for a flatter TIPS yield curve (Chart 12). The combination of inflation curve steepening and nominal curve flattening will lead to a supercharged flattening of the real yield curve during the next 12 months. Chart 12... And Real Yield Curve Flattening ... And Real Yield Curve Flattening ... And Real Yield Curve Flattening Bottom Line: Investors should position for higher short-maturity real yields. This can be done through an outright short position in 2-year TIPS, an inflation curve steepener or a real yield curve flattener.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 The median response from the New York Fed’s Survey of Market Participants pegs the long-run neutral fed funds rate at 2.0%. The same measure from the Survey of Primary Dealers sits at 2.25%. 2 For more details on the relationship between the proximity of the 5-year/5-year yield to its fair value range and the slope of the yield curve please see US Bond Strategy Weekly Report, “A Bump On The Road To Recovery”, dated July 27, 2021. 3 https://www.atlantafed.org/news/speeches/2021/10/12/bostic-the-current-inflation-episode.aspx Recommended Portfolio Specification Other Recommendations Treasury Index Returns Spread Product Returns

Highlights Cross-Atlantic Policy Divergence: A steadily tightening US labor market means that the Fed remains on track to formally announce tapering next month. Meanwhile, the ECB is signaling that they are in no hurry to do the same given scant evidence that surging energy prices are seeping into broader European inflation. This leads us to make the following changes to our tactical trade portfolio – taking profits on the 10-year French inflation breakeven spread widener; while switching out of the long December 2023 Euribor futures trade into a 10-year US Treasury-German Bund spread widening trade. Surging Antipodean Inflation: Australia and New Zealand are both seeing higher realized inflation, but market-based inflation expectations are falling in the former and rising in the latter. This leads us to make the following changes to our tactical trades: taking profits on the Australia-US 10-year spread widener; entering a new 10-year Australia inflation breakeven spread widener; and closing the underwater 2-year/5-year New Zealand curve flattening trade. Feature This week, we present a review of the shorter-term recommendations currently in our list of Tactical Overlay trades. These are positions that are intended to complement our strategic Model Bond Portfolio, with shorter holding periods – our goal is no longer than six months - and sometimes in smaller markets that are outside our usual core bond market coverage. As can be seen in the table on page 17, we typically organize these ideas by the type of trade (i.e. yield curve flatteners or cross-country spread wideners). Yet for the purposes of this review, we see two interesting themes that better organize the current trades and help guide our decision to keep them or enter new ones. Playing A Hawkish Fed Versus A Dovish ECB Federal Reserve officials have spent the past few months signaling that a tapering of bond purchases was increasingly likely to begin before year-end given the steadily improving US labor market. The September payrolls report released last Friday, even with the headline employment growth number below expectations for the second consecutive month, does not change that trajectory. Chart of the WeekCyclical UST Curve Flattening Pressures Cyclical UST Curve Flattening Pressures Cyclical UST Curve Flattening Pressures The US unemployment rate fell to 4.8% in September, continuing the uninterrupted decline from the April 2020 peak of 14.8% (Chart of the Week). The pace of that decline has accelerated in recent months, although the Delta variant surge in the US has created distortions in both the numerator and denominator of the unemployment rate. Now that the US Delta wave has crested and case numbers are falling, growth in both employment and the labor force should start to accelerate in the next few payrolls reports. This will result in a faster pace of US job growth, albeit with a slower decline in the unemployment rate, likely starting as soon as the October jobs report. The US Treasury curve has already been reshaping in preparation for a less accommodative Fed, with flattening seen beyond the 5-year point (middle panel). We have positioned for a more hawkish Fed, and a flatter Treasury curve, in our Tactical Overlay via a butterfly trade. Specifically, we are short a 5-year Treasury bullet versus a long position in a 2-year/10-year barbell, all using on-the-run cash Treasuries. That trade was initiated on June 22, 2021 and has so far generated a small profit of +0.27%. Our butterfly spread valuation model for that 2/5/10 Treasury butterfly shows that the 5-year bullet has not yet reached an undervalued extreme versus the 2/10 barbell (Chart 2). We are keeping this trade in our Tactical Overlay, as the current 2/5/10 butterfly spread of 23bps is still 6bps below the +1 standard deviation level implied by our model. Chart 2Stay In Our 2/5/10 UST Butterfly Trade Stay In Our 2/5/10 UST Butterfly Trade Stay In Our 2/5/10 UST Butterfly Trade Moving across the Atlantic, our trades have been the mirror image of our Fed recommendations, positioning for a continued dovish, reflationary ECB policy bias. We have expressed that via two trades: long 10-year French inflation breakevens and long December 2021 Euribor futures. We continue to see no reason for the ECB to follow the Fed’s path towards imminent tapering and signaling future rate hikes. Growth momentum has cooled in the euro area, with both the Markit composite PMI and the ZEW growth expectations index having peaked in June (Chart 3). At the same time, inflation expectations have picked up. The 5-year/5-year forward CPI swap rate has risen to 1.8%, still below the ECB’s 2% inflation target but well above the 2020 low of 0.7% (middle panel). Markets are focusing on the higher inflation and not the slowing growth, with the EUR overnight index swap (OIS) curve now pricing in 12bps of rate hikes in 2022 (bottom panel). We see that as a highly improbable outcome. There is little evidence that the latest pickup in euro area realized inflation is broadening out beyond surging energy price inflation and supply-constrained goods inflation (Chart 4). Euro area headline CPI inflation hit a 13-year high of 3.0% in August, with the “flash” estimate for September showing a further acceleration to 3.4%. Yet core inflation only reached 1.6% in August - a month when the trimmed mean euro area CPI inflation rate calculated by our colleagues at BCA Research European Investment Strategy was a scant 0.2%. Chart 3ECB Will Not React To This Cyclical Bout Of Inflation ECB Will Not React To This Cyclical Bout Of Inflation ECB Will Not React To This Cyclical Bout Of Inflation Chart 4Euro Area Inflation Upturn Is Not Broad-Based Euro Area Inflation Upturn Is Not Broad-Based Euro Area Inflation Upturn Is Not Broad-Based While the September flash estimate of core inflation did perk up to 1.9%, the trimmed mean measure shows that the rise in euro area inflation to date has not been broad based. Like the Fed, ECB officials have indicated that they view this pick-up in inflation as “transitory”, fueled by soaring energy costs and base effect comparisons to low inflation in 2020. Signs that higher inflation was feeding into “second round” effects like rising wage growth might change the ECB’s thinking. From that perspective, the recent increase in labor strike activity in Germany is a potentially worrisome sign, but the starting point is one of low wage growth – the latest available data on euro area wage costs showed a -0.1% decline during Q2/2021. Chart 5Close Our Long Dec/23 Euribor Futures Trade Close Our Long Dec/23 Euribor Futures Trade Close Our Long Dec/23 Euribor Futures Trade We have been trying to fade ECB rate hike expectations via our long December 2023 Euribor futures trade. That position, initiated on May 18, 2021 has generated a small loss of -0.11% (Chart 5). We still expect the ECB to keep rates on hold in 2022, and most likely 2023, so there is the potential for that trade to recover that underperformance. However, that position has now reached the six-month holding period “re-evaluation” limit that we have imposed on our Tactical Overlay trades. Thus, we are closing that trade this week. In its place, we are initiating a new tactical trade to position for not only persistent ECB dovishness but a more hawkish Fed – a US Treasury-German Bund spread widening trade using 10-year bond futures. The specific details of the trade (futures contracts, duration-neutral weightings on each leg of the trade) can be found in the table on page 17. This new UST-Bund trade is attractive for three reasons: Our valuation model for the Treasury-Bund spread - which uses relative policy interest rates, relative unemployment, relative inflation and the relative size of the Fed and ECB balance sheets as inputs – shows that the spread is currently undervalued by more than one full standard deviation, and fair value is rising (Chart 6). The technical backdrop for the Treasury-Bund spread has turned more favorable for wideners, with the spread having fallen back to its 200-day moving average and the 26-week change in the spread now down to levels that preceded past turning points in the spread (Chart 7). Chart 6Enter A New 10yr UST-Bund Spread Widening Trade Enter A New 10yr UST-Bund Spread Widening Trade Enter A New 10yr UST-Bund Spread Widening Trade Relative data surprises are pointing to relatively higher US yields and a wider Treasury-Bund spread, with the Citigroup Data Surprise Index for the US now rising and the euro area equivalent measure falling (Chart 8). Chart 7UST-Bund Technical Backdrop Positioned For Widening UST-Bund Technical Backdrop Positioned For Widening UST-Bund Technical Backdrop Positioned For Widening Chart 8Relative Data Surprises Favor Wider UST-Bund Spread Relative Data Surprises Favor Wider UST-Bund Spread Relative Data Surprises Favor Wider UST-Bund Spread While we are entering a new trade to play for a relatively dovish ECB, we are also choosing to take the substantial profit in our tactical trade in French inflation breakevens. Specifically, we are closing our 10-year French inflation breakeven spread widening position – long a 10-year cash OATi bond, short 10-year French bond futures – with a solid gain of +6.3%. Chart 9Take Profits On Our Long 10yr French Breakevens Trade Take Profits On Our Long 10yr French Breakevens Trade Take Profits On Our Long 10yr French Breakevens Trade We have held this trade for nine months, a bit longer than our typical tactical trade holding period. We did so because French 10-year breakevens continued to look cheap on our valuation model. Now, the breakeven spread has risen to fair value (Chart 9), prompting us to take our gains and move on. Diverging Inflation Expectations In Australia & New Zealand Playing Fed/ECB policy divergence was the first main theme of this Tactical Overlay trade review. The second broad theme is also a divergence, between inflation expectations in New Zealand (which are rising) and Australia (which are falling). This trend leads us to close two existing trades and enter a new position. Chart 10An Inflation-Induced Bear Steepening Of Yield Curves An Inflation-Induced Bear Steepening Of Yield Curves An Inflation-Induced Bear Steepening Of Yield Curves In New Zealand, we are closing out our 2-year/5-year government bond yield curve flattener trade, initiated on July 21, for a loss of -0.32%. While we were correct in our expectation of ramped-up hawkishness from the Reserve Bank of New Zealand (RBNZ), we were caught offside by persistently sticky inflation which has become a headache for global central bankers. With supply squeezes and high commodity prices not going away anytime soon, sovereign curves have bear-steepened across developed markets, driven by rising long-dated inflation expectations (Chart 10). This global steepening pressure also hit the New Zealand curve, to the detriment of our domestic RBNZ-focused flattener trade. There was also a technical component to the steepening in the New Zealand 2-year/5-year curve (Chart 11). With the 2-year/5-year curve having dipped far below its 200-day moving average and the 26-week rate of change at stretched levels, the flattener was already “overbought” when we entered the trade. Despite a steady stream of hawkish messaging from the RBNZ, leading to an actual rate hike last week, technicals did win out in the short term as the 2-year/5-year spread steepened back up towards the 200-day moving average. Chart 11The NZ 2s/5s Curve Has Also Steepened Due To Technical Factors The NZ 2s/5s Curve Has Also Steepened Due To Technical Factors The NZ 2s/5s Curve Has Also Steepened Due To Technical Factors On the positive side, our decision to implement this trade as a duration-neutral “butterfly”, selling a 2-year bond, and using the proceeds to buy a weighted combination of a 5-year bond and a 3-month treasury bill with an equivalent duration to the 2-year bond, worked as intended with the butterfly underperforming as the underlying 2-year/5-year curve steepened. Looking forward, technicals are still some distance from turning favorable and will remain a headwind for the flattener trade. Implied forward rates are also not in our favor, with markets already pricing in some flattening, making this a negative carry trade. Over a cyclical horizon – i.e. beyond our normal six-month holding period for tactical trades - we still expect the shorter-end of the New Zealand to flatten. The experience of past hiking cycles shows that the 2-year/5-year curve tends to continue flattening during policy tightening, usually leveling out at 0bps before re-steepening (Chart 12). Considering that we have already been in this trade for three months, however, we do not believe our initial curve flattening bias will play out successfully over the remainder of our six-month tactical horizon. While we are closing out our flattener trade, we will investigate ways to better express our bearish cyclical view on New Zealand sovereign debt in a future report. Turning to Australia, we are closing out our long Australia/short US spread trade, implemented using 10-year bond futures, taking a healthy profit of +2.1%. We have held this trade for longer than our typical six-month holding period (the trade was initiated on January 26, 2021) because our Australia-US 10-year spread valuation model has continued to flash that the spread was too wide to its fair value (Chart 13). The model has been signaling that the spread should be negative, yet Australian yields have been unable to trade below US yields for any sustained length of time in 2021. Furthermore, the model-implied fair value is now starting to bottom out, suggesting a diminishing tailwind from the relative fundamental drivers of the spread embedded in our model. Chart 12The NZ 2s/5s Curve Will Flatten Over A Cyclical Horizon The NZ 2s/5s Curve Will Flatten Over A Cyclical Horizon The NZ 2s/5s Curve Will Flatten Over A Cyclical Horizon Chart 13Take Profits On Our 10-Yr Australia-US Spread Narrowing Trade Take Profits On Our 10-Yr Australia-US Spread Narrowing Trade Take Profits On Our 10-Yr Australia-US Spread Narrowing Trade Chart 14Inputs Into Our Australia-US Spread Model Inputs Into Our Australia-US Spread Model Inputs Into Our Australia-US Spread Model The inputs into our 10-year spread model are relative policy interest rates, core inflation, unemployment and the size of central bank balance sheets (to incorporate QE effects) for Australia and the US. Of these variables, the biggest drivers of the decline in the fair value since the start of the COVID pandemic in 2020 have been relative inflation and the relative size of the Fed and Reserve Bank of Australia (RBA) balance sheets as a percentage of GDP (Chart 14). Both of those trends are related. Persistently underwhelming Australian inflation – despite accelerating inflation in the US and other developed economies over the past year – has forced the RBA into a pace of asset purchases relative to GDP that exceeded even what the Fed has done since the pandemic started (bottom panel). However, Australian inflation finally began catching up to the rising trends seen elsewhere in the spring of this year, with headline CPI inflation jumping from 1.1% to 3.8% on a year-over-year basis during Q2. Australian bond yields have traded more in line with US yields since that mid-year pop in inflation, preventing the Australia-US spread from narrowing below zero and converging to our model-implied fair value. This is despite a severe COVID wave that forced much of Australia into the kind of severe lockdowns that the nation avoided during the worst of the global pandemic in 2020. With Australian inflation now moving higher and converging towards US levels, economic restrictions starting to be lifted thanks to a rapid vaccination campaign, and the RBA having already done some tapering of its asset purchases before the Fed, the fundamental rationale for holding our Australia-US trade is no longer valid, leading us to take profits. The convergence to fair value in our spread model is now more likely to come from fair value rising rather than the actual spread falling. The pickup in Australian inflation also leads us to enter a new trade Down Under. This week, we are initiating a new trade, going long 10-year Australia inflation breakevens, implemented by going long a 10-year cash inflation-linked bond and selling 10-year bond futures. The details of the new trade are shown in the table on page 17. Despite the uptick in realized Australian inflation, breakevens have actually been declining over the past several months, falling from a peak of 247bps on May 13 to the current 208bps. That move has accelerated more recently due to a rise in Australian real yields that has coincided with markets pricing in more future RBA rate hikes. Our 24-month Australia discounter, which measures the total amount of tightening over the next two years discounted in the AUD OIS curve, now shows that 104bps of rate hikes are expected by the fourth quarter of 2023 (Chart 15, bottom panel). This has occurred despite Australian wage growth remaining well below the 3-4% range that the RBA believes is consistent with underlying Australian inflation returning sustainably to the RBA’s 2-3% target band (top two panels). Chart 15Market Expectations For The RBA Are Too Hawkish Market Expectations For The RBA Are Too Hawkish Market Expectations For The RBA Are Too Hawkish Chart 16Go Long 10-Yr Australian Inflation Breakevens Go Long 10-Yr Australian Inflation Breakevens Go Long 10-Yr Australian Inflation Breakevens Australian real bond yields have begun to move higher in response to this more hawkish market policy expectation that seems overdone, helping push breakeven inflation even lower more recently. This has helped unwind some of the overvaluation of 10-year inflation breakevens from earlier in 2021. Our fundamental model for the 10-year Australian breakeven showed that the spread was over two standard deviations above fair value to start 2020 (Chart 16). The decline in the spread since that has largely eliminated that overvaluation, providing a better entry point for a new breakeven spread widening trade. With survey-based measures of inflation expectations rising even as breakevens fall back to fair value (bottom panel), we see a strong case for adding a new Australian inflation trade to our Tactical Overlay.   Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com Shakti Sharma Senior Analyst ShaktiS@bcaresearch.com Recommendations The GFIS Recommended Portfolio Vs. The Custom Benchmark Index A Thematic Update Of Our Tactical Trades A Thematic Update Of Our Tactical Trades Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns

  Highlights Chart 1Bond Yields Still Track The "Re-Opening" Trade Bond Yields Still Track The "Re-Opening" Trade Bond Yields Still Track The "Re-Opening" Trade Bond yields rose notably in September, with the bulk of the move coming in the days after the Fed teased an upcoming tapering of its asset purchases and revealed slightly hawkish revisions to its interest rate projections. Interestingly, some of the details of the bond market move don’t mesh nicely with the mildly hawkish policy surprise that the Fed delivered. For example, the Treasury curve steepened on the month and long-maturity TIPS breakeven inflation rates rose. Our sense is that September’s market moves were less driven by the Fed and more by a revival of the reflation (or re-opening) trade from earlier this year. The daily new US COVID case count ticked down and, while overall S&P 500 returns were negative on the month, a basket of equities designed to profit from the end of the pandemic soundly beat a basket of “COVID winners” (Chart 1). With the delta COVID wave receding, we remain confident that economic growth will be sufficiently strong for the Fed to launch a new rate hike cycle in December 2022. The Treasury curve will bear-flatten as that outcome gets priced in.   Feature Table 1Recommended Portfolio Specification A Bout Of Reflation A Bout Of Reflation Table 2Fixed Income Sector Performance A Bout Of Reflation A Bout Of Reflation Investment Grade: Neutral Chart 2Investment Grade Market Overview Investment Grade Market Overview Investment Grade Market Overview Investment grade corporate bonds outperformed the duration-equivalent Treasury index by 26 basis points in September, bringing year-to-date excess returns up to +193 bps. The combination of above-trend economic growth and accommodative monetary policy supports continued positive excess returns for spread product versus Treasuries. At 99 bps, the 3-year/10-year Treasury slope remains steep. This is a strong signal that monetary conditions are accommodative. But despite the positive macro back-drop, investment grade valuations are extremely tight (Chart 2). A recent report presented the results of a scenario analysis for investment grade corporate bond returns during the next 12 months.1 We concluded that investment grade corporate bond total returns will be close to zero or negative during the next 12 months and that excess returns versus duration-matched Treasuries are capped at 85 bps. With that in mind, we advise investors to seek out higher returns in junk bonds, municipal bonds and USD-denominated Emerging Market sovereign and corporate bonds. We also recommend favoring long-maturity corporate bonds and those corporate sectors with elevated Duration-Times-Spread.2 Table 3ACorporate Sector Relative Valuation And Recommended Allocation* A Bout Of Reflation A Bout Of Reflation Table 3BCorporate Sector Risk Vs. Reward* A Bout Of Reflation A Bout Of Reflation High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 53 basis points in September, bringing year-to-date excess returns up to 558 bps. A recent report looked at the default expectations that are currently priced into the junk index and considered whether they are likely to be met.3 If we demand an excess spread of 100 bps and assume a 40% recovery rate on defaulted debt, then the High-Yield index embeds an expected default rate of 3.2% (Chart 3). Using a model of the 12-month trailing speculative grade default rate that is based on gross corporate leverage (pre-tax profits over total debt) and C&I lending standards, we estimate that the 12-month default rate will fall to between 2.3% and 2.8%, below what the market currently discounts. Notably, the corporate default rate is tracking at an annualized rate of roughly 1.7% through the first eight months of the year, well below the estimate generated by our macro model. Another recent report considered different plausible scenarios for junk bond returns during the next 12 months.4 We concluded that junk bond total returns will fall into a range of -0.29% to +1.80% during the next 12 months and that excess returns versus duration-matched Treasuries will be between +0.94% and +1.84%. MBS: Underweight Chart 4MBS Market Overview MBS Market Overview MBS Market Overview Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 24 basis points in September, bringing year-to-date excess returns up to -43 bps. The nominal spread between conventional 30-year MBS and equivalent-duration Treasuries tightened 19 bps in September. The spread is wide compared to recent history, but it remains tight compared to the recent pace of mortgage refinancings (Chart 4). The conventional 30-year MBS option-adjusted spread (OAS) tightened 6 bps in September to reach 31 bps (panel 3). This is above the 22 bps offered by Aaa-rated consumer ABS but below the 52 bps offered by Aa-rated corporate bonds and the 33 bps offered by Agency CMBS. In a recent report we looked at MBS performance and valuation across the coupon stack.5 We noted that the higher convexity of high-coupon MBS makes them likely to outperform lower-coupon MBS in a rising yield environment. Higher coupon MBS also have greater OAS than lower coupons. This makes the high-coupon MBS more likely to outperform in a flat bond yield environment as well. Given our view that bond yields will be higher in 6-12 months, we recommend favoring high coupons (4%, 4.5%) over low coupons (2%, 2.5%, 3%) within an overall underweight allocation to Agency MBS. Government-Related: Neutral Chart 5Government-Related Market Overview Government-Related Market Overview Government-Related Market Overview The Government-Related index underperformed the duration-equivalent Treasury index by 15 basis points in September, dragging year-to-date excess returns down to +69 bps (Chart 5). Sovereign debt underperformed duration-equivalent Treasuries by 95 bps in September, dragging year-to-date excess returns down to -87 bps. Foreign Agencies outperformed the Treasury benchmark by 5 bps on the month, bringing year-to-date excess returns up to +49 bps. Local Authority bonds outperformed by 24 bps in September, bringing year-to-date excess returns up to +406 bps. Domestic Agency bonds underperformed by 7 bps, dragging year-to-date excess returns down to +24 bps. Supranationals underperformed by 4 bps, dragging year-to-date excess returns down to +27 bps. Last week’s report looked at performance and valuation trends for Emerging Market sovereign and corporate bonds relative to US corporates.6 The recent underperformance of EM bonds versus US corporates has led to attractive relative valuations in the sector. We see investment grade EM sovereign and corporate bonds both outperforming investment grade US corporates during the next 12 months. The outperformance will be the result of better starting valuations and an acceleration of EM growth in 2022. The bonds of Colombia, Russia, Mexico, Indonesia, Saudi Arabia, UAE and Qatar look particularly attractive within the USD-denominated EM sovereign space. Municipal Bonds: Overweight Chart 6Municipal Market Overview Municipal Market Overview Municipal Market Overview Municipal bonds outperformed the duration-equivalent Treasury index by 29 basis points in September, bringing year-to-date excess returns up to +292 bps (before adjusting for the tax advantage). The economic and policy back-drop remains favorable for municipal bond performance. Trailing 4-quarter net state & local government savings were already positive through the end of Q1 2021 and they received another significant boost in Q2 as funds from the American Rescue Plan were doled out (Chart 6). With state & local government balance sheets in such good shape, we are comfortable moving down in quality within municipal bonds. A move down in quality is especially compelling because of tight Aaa muni valuations relative to Treasuries (top panel). Valuation is more compelling in the lower investment grade credit tiers, especially at the long-end of the curve.7 Both General Obligation (GO) and Revenue munis in the 12-17 year maturity bucket offer a before-tax yield pick-up versus corporate bonds with the same credit rating and duration (panel 2). Finally, high-yield muni spreads are reasonably attractive relative to high-yield corporates, offering a breakeven tax rate of 25% (panel 4). But despite the attractive spread, we recommend only a neutral allocation to high-yield munis versus high-yield corporates as the deep negative convexity of high-yield munis makes them susceptible to extension risk if bond yields rise. Treasury Curve: Buy 2/10 Barbell Versus 5-Year Bullet Chart 7Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury Yield Curve Overview The Treasury curve bear-steepened in September, with yields moving sharply higher – especially in the 5-10 year maturity space. The 2-year/10-year Treasury slope steepened 14 bps to end the month at 124 bps. The 5-year/30-year slope flattened 5 bps to end the month at 110 bps. We expect bond yields to be higher in 6-12 months, but we also anticipate that the next significant move higher in bond yields will coincide with curve flattening, not steepening. At 2.08%, the 5-year/5-year forward Treasury yield is already within our target fair value range of 2% - 2.25%. In a recent report, we demonstrated that yield curve steepening only occurs when either the Fed is cutting rates or the 5-year/5-year forward yield rises.8 This means that the 2/10 Treasury curve is more likely to flatten than steepen during the next 6-12 months, even as bond yields move higher. Similarly, we observe that the overnight index swap (OIS) curve is priced for the fed funds rate to be 0.30% in one year’s time and 1.62% in five years (Chart 7). The latter rate has 131 bps of upside if it converges all the way back to its 2018 high, but this pales in comparison to the 256 bps of upside in the 12-month forward rate. The yield curve will flatten as the 12-month forward OIS rate converges with the 5-year forward rate (panel 3). Investors should position in yield curve flatteners on a 6-12 month horizon. Specifically, we recommend shorting the 5-year bullet versus a duration-matched 2/10 barbell. TIPS: Neutral Chart 8TIPS Market Overview TIPS Market Overview TIPS Market Overview TIPS outperformed the duration-equivalent nominal Treasury index by 47 basis points in September, bringing year-to-date excess returns up to +627 bps. The 10-year TIPS breakeven inflation rate rose 3 bps on the month, while the 5-year/5-year forward TIPS breakeven inflation rate rose 5 bps. At 2.41%, the 10-year TIPS breakeven inflation rate is near the middle of the 2.3% to 2.5% range that is consistent with inflation expectations being well anchored around the Fed’s target (Chart 8). Meanwhile, at 2.26%, the 5-year/5-year forward TIPS breakeven inflation rate is only just below target (panel 3). With long-dated inflation expectations close to the Fed’s target levels, we see limited upside on a 6-12 month horizon. We also see the cost of short-maturity inflation protection falling during the next few months as realized inflation moderates from its extremely high level. This will lead to a steepening of the inflation curve (bottom panel). We recommend that investors position for a steeper 2/10 inflation curve, or alternatively for a flatter 2/10 real Treasury curve. We noted in last week’s report that the combination of nominal curve flattening and inflation curve steepening will lead to a large flattening of the 2/10 real curve during the next 6-12 months.9The 2-year TIPS yield, in particular, has a lot of upside.                         ABS: Overweight Chart 9ABS Market Overview ABS Market Overview ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent nominal Treasury index by 3 basis points in September, bringing year-to-date excess returns up to +43 bps. Aaa-rated ABS outperformed by 2 bps on the month, bringing year-to-date excess returns up to +32 bps. Non-Aaa ABS outperformed by 7 bps, bringing year-to-date excess returns up to +99 bps. The stimulus from last year’s CARES Act led to a significant increase in household savings when individual checks were mailed in April 2020. That excess savings has still not been spent and the most recent round of stimulus checks has only added to the stockpile (Chart 9). The extraordinarily large stock of household savings means that the collateral quality of consumer ABS is also extraordinarily high. Indeed, many households have been using their windfalls to pay down consumer debt (bottom panel). Investors should remain overweight consumer ABS and should also take advantage of the high quality of household balance sheets by moving down the quality spectrum.     Non-Agency CMBS: Neutral Chart 10CMBS Market Overview CMBS Market Overview CMBS Market Overview Non-Agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 2 basis points in September, bringing year-to-date excess returns up to +195 bps. Aaa Non-Agency CMBS outperformed Treasuries by 4 bps in September, bringing year-to-date excess returns up to +96 bps. Non-Aaa Non-Agency CMBS underperformed Treasuries by 4 bps on the month, dragging year-to-date excess returns down to +525 bps (Chart 10). Though returns have been strong and spreads remain attractive, particularly for lower-rated CMBS, we continue to recommend only a neutral allocation to the sector because of the structurally challenging environment for commercial real estate. Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 3 basis points in September, bringing year-to-date excess returns up to +94 bps. The average index option-adjusted spread tightened 1 bp on the month. It currently sits at 33 bps (bottom panel). Though Agency CMBS spreads have recovered to well below pre-COVID levels, they still look attractive compared to other similarly risky spread products. Stay overweight.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Appendix A: 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 September 30th, 2021) A Bout Of Reflation A Bout Of Reflation 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 September 30th, 2021) A Bout Of Reflation A Bout Of Reflation 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 -17 bps in the 5 over 2/10 cell means that we would expect the 5-year to outperform the 2/10 if the 2/10 flattens by less than 17 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) A Bout Of Reflation A Bout Of Reflation Appendix B: 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 September 30th, 2021) A Bout Of Reflation A Bout Of Reflation Footnotes 1  Please see US Bond Strategy Weekly Report, “Expected Returns In Corporate Bonds”, dated September 21, 2021. 2  Please see US Bond Strategy Weekly Report, “The Collapsing Credit Risk Premium”, dated July 20, 2021. 3  Please see US Bond Strategy Weekly Report, “The Post-FOMC Credit Environment”, dated June 29, 2021. 4  Please see US Bond Strategy Weekly Report, “Expected Returns In Corporate Bonds”, dated September 21, 2021. 5  Please see US Bond Strategy Weekly Report, “A New Conundrum”, dated April 20, 2021. 6  Please see US Bond Strategy Weekly Report, “Damage Assessment”, dated September 28, 2021. 7  Please see US Bond Strategy Weekly Report, “The Collapsing Credit Risk Premium”, dated July 20, 2021. 8  Please see US Bond Strategy / Global Fixed Income Strategy Weekly Report, “A Bump On The Road To Recovery”, dated July 27, 2021. 9  Please see US Bond Strategy Weekly Report, “Damage Assessment”, dated September 28, 2021.

Highlights Monetary Policy: It’s all but certain that tapering will begin next month and conclude by the middle of next year, but the FOMC is currently split right down the middle on whether it will be appropriate to lift rates in H2 2022. We present five factors to track to decide when the Fed will hike rates. Yield Curve: A bear-flattening of the nominal yield curve remains the most likely scenario for the next 6-12 months. Maintain a position short the 5-year bullet versus a duration-matched 2/10 barbell. TIPS: Investors should enter 2/10 steepeners on the inflation compensation curve and/or 2/10 flatteners on the real (TIPS) curve. The 2-year real yield, in particular, has a lot of upside during the next 6-12 months. EM Bonds: US bond investors should shift some allocation out of expensive US investment grade corporate bonds and into USD-denominated investment grade EM Sovereigns and Corporates where valuation is more compelling. Feature This past week was quite eventful for US bond investors. First, the Fed presented its September update on Wednesday, offering some hints about the timing and pace of asset purchase tapering alongside its updated economic and interest rate forecasts. Then, bonds sold off sharply on Thursday, Friday and Monday with the 10-year Treasury yield rising all the way to 1.49%. The first section of this week’s report looks at what we learned from last week’s FOMC meeting and opines on some monetary policy questions that remain unanswered. The second section updates our views on portfolio duration, the yield curve and TIPS in light of last week’s large market moves. Finally, we conclude with an update on the outlook for USD-denominated Emerging Market bonds. Powell Answers Some Small Questions, Avoids The Big One The Small Questions The start date for asset purchase tapering is the first small question that the Fed answered last week. A new sentence was added to the post-meeting statement saying that “a moderation in the pace of asset purchases may soon be warranted” and, in his press conference, Chair Powell clarified that the purpose of the new language is “to put notice out there that [tapering] could come as soon as the next meeting”. After this statement from the Fed, we expect asset purchase tapering to be announced at the next FOMC meeting on November 3rd. A delay until December is possible if September’s employment report is a massive disappointment, but the bar for delaying tapering beyond November appears high. Chair Powell also shed some light on a second small question related to the Fed’s balance sheet. Specifically, the pace of asset purchase tapering. In his press conference the Chair said that “participants generally view that, so long as the recovery remains on track, a gradual tapering process that concludes around the middle of next year is likely to be appropriate.” The information about the pace of tapering is slightly more interesting than the start date, if only because Fed policymakers have expressed a desire for net purchases to reach zero before the first interest rate increase. The fact that the Fed is planning to reach net zero purchases by the middle of next year suggests that it wants to leave the door open for a possible rate hike in the second half of 2022. This message was confirmed by the Fed’s updated interest rate projections (Chart 1). The projections revealed an even split (9 vs. 9) on the Committee between those that expect at least one 25 basis point rate increase before the end of 2022 and those that do not expect to lift rates next year. It’s also notable that, out of the 9 participants that don’t expect a rate hike in 2022, 8 anticipate lifting rates in 2023. Chart 1Rate Expectations Rate Expectations Rate Expectations Of course, as Chair Powell often points out, the Fed’s interest rate projections are contingent on the future state of the economy and will almost certainly be revised as the outlook evolves. What’s more important from a forecasting perspective is knowing how the Fed will react to different economic conditions. The Big Question With that in mind, we know that the Fed has promised not to lift rates until the labor market reaches “maximum employment”. However, as we noted in a recent report, the Fed hasn’t provided much detail on what “maximum employment” actually means.1 Powell’s comments last week didn’t add much clarity, but we can hunt for clues in the Summary of Economic Projections (SEP). In the SEP, we first notice that 13 FOMC participants expect the unemployment rate to be 3.8% - 3.9% or lower by the end of 2022 (Chart 2A). It is logical to assume that this group includes the 9 participants who expect to lift rates in 2022 and 4 additional ones. We also observe that 17 out of 18 participants anticipate an unemployment rate of 3.8% - 3.9% or lower by the end of 2023 (Chart 2B), exactly matching the number of participants who expect to lift rates by then. We can therefore infer that the Fed views an unemployment rate of roughly 3.8% as consistent with “maximum employment”. That is, FOMC participants tend to have rate hikes penciled into their forecasts once the unemployment rate is below 3.8%. Chart 2AFOMC 2022 Unemployment Rate Projection Distribution Damage Assessment Damage Assessment Chart 2BFOMC 2023 Unemployment Rate Projection Distribution Damage Assessment Damage Assessment The SEP’s inflation forecasts are less illuminating than the unemployment ones. All but 4 participants expect core PCE inflation to be above the Fed’s 2% target at the end of 2022 (Chart 3A) and no FOMC participant is forecasting below-2% core PCE at the end of 2023 (Chart 3B).2 The most we can take away from these forecasts is that 14 FOMC participants expect inflation to be above target in 2022, but five of those participants don’t see the labor market as being tight enough to lift rates by then. Chart 3AFOMC 2022 Core PCE Inflation Rate Projection Distribution Damage Assessment Damage Assessment Chart 3BFOMC 2023 Core PCE Inflation Rate Projection Distribution Damage Assessment Damage Assessment The revelation that FOMC participants view a 3.8% unemployment rate as consistent with “maximum employment” is illuminating, but it doesn’t tell the entire story. We don’t know, for example, what assumptions about labor force participation the different Fed officials are using. Our sense is that the following five criteria will ultimately determine when the Fed starts to lift interest rates: An unemployment rate in the neighborhood of 3.8%. Prime-age (25-54) labor force participation close to its pre-pandemic level (Chart 4, top panel). Accelerating wage growth (Chart 4, bottom 2 panels). Long-dated inflation expectations at or above target levels (Chart 5, top 2 panels). Non-transitory inflation at or above target levels (Chart 5, bottom panel). Bottom Line: The FOMC is currently split right down the middle on whether it will be appropriate to lift rates in 2022, but these forecasts aren’t set in stone and next year’s economic data – particularly the five factors listed above – will determine which group is correct. Chart 4Part Rate & Wage Growth Part Rate & Wage Growth Part Rate & Wage Growth Chart 5Inflation & Inflation Expectations Inflation & Inflation Expectations Inflation & Inflation Expectations Our own view is that the labor market will be sufficiently tight for the Fed to deliver its first rate hike in December 2022, and that rate hikes will proceed more quickly than what is currently priced in the yield curve. US bond investors should maintain below-benchmark portfolio duration. The Fallout In Bond Markets Bonds sold off sharply last Thursday and again yesterday morning. At the time of publication, the 30-year Treasury yield had risen from 1.81% on Wednesday to 2.02% (Chart 6). The 10-year Treasury yield had risen from 1.30% to 1.49% (Chart 6, panel 2). The 5-year yield is up to 0.98% from 0.85% on Wednesday (Chart 6, panel 3), and the 2-year yield is up to 0.28% from 0.24% (Chart 6, panel 4). The yield curve has also steepened since Wednesday, though the 2/10 slope remains well below its March peak (Chart 7). The trend during the past few months has been one of curve steepening out to the 5-year maturity point (Chart 7, panel 2) and flattening beyond that point (Chart 7, bottom 2 panels). Recent market action has led to some steepening beyond the 5-year maturity, but so far that steepening is minor compared to the flattening move we’ve witnessed since the spring. Chart 6Treasury Yield Trends Treasury Yield Trends Treasury Yield Trends Chart 7Treasury Curve Trends Treasury Curve Trends Treasury Curve Trends Nominal Curve We have been recommending a position short the 5-year bullet and long a duration-matched barbell consisting of the 2-year and 10-year notes, and we think this trade will continue to outperform. First, the 5-year/5-year forward Treasury yield is back above 2%, consistent with median estimates of the long-run neutral fed funds rate from the New York Fed’s Surveys of Market Participants (2%) and Primary Dealers (2.25%) (Chart 6, bottom panel). We think it will be difficult for the 5-year/5-year forward yield to rise much above these levels, which makes curve flattening more likely than steepening going forward.3 Second, we updated our work on yield curve scenarios to incorporate recent market action and some new information gleaned from last week’s SEP. In this analysis we estimate fair value levels for different parts of the yield curve as of the end of 2022 based on a scenario for the path of the fed funds rate. For example, if we assume that Fed liftoff occurs in December 2022 and rate hikes proceed at a pace of 75 bps per year (the median pace in the SEP) until the fed funds rate levels-off at 2.08% (consistent with a 2%-2.25% range for the long-run neutral fed funds rate), we calculate that the 2-year yield has 74 bps of upside between now and the end of 2022. This is slightly more than the 65 bps of upside in the 5-year yield and much more than the 37 bps of upside in the 10-year yield (Table 1). Comparing this expected change to what is already discounted in the forward curve, we see that both the 2-year and 10-year yields are expected to exceed their forwards by 6 bps. The 5-year yield is expected to exceed its forward by 16 bps. Table 1Treasury Curve Scenario Analysis Damage Assessment Damage Assessment To us, the message is clear. We want to remain short the 5-year note versus a duration-matched 2/10 barbell. Real & Inflation Curves Chart 8Real & Inflation Curves Real & Inflation Curves Real & Inflation Curves The increase in the 10-year nominal yield since last Wednesday was roughly evenly split between the real and inflation compensation components. Interestingly, inflation drove a greater proportion of the increase at the front-end of the curve, and the 2-year real yield actually fell. We maintain a neutral recommendation on TIPS versus nominal Treasuries, but this week we recommend taking advantage of recent market action by implementing some curve trades across the real and inflation curves. Specifically, we note that the cost of short-maturity inflation compensation remains well above the cost of long-maturity inflation compensation (Chart 8). Going forward, the cost of front-end inflation compensation will fall as inflation moderates from its current extremely high level. Meanwhile, the cost of long-maturity inflation compensation will remain sticky near the Fed’s target levels. This will lead to a steepening of the inflation curve (Chart 8, panel 2). The combination of a steeper inflation curve and a flatter nominal curve will lead to a much flatter real yield curve (Chart 8, bottom 2 panels). The 2-year real yield, in particular, has a lot of upside if inflation moderates, as we expect, and the 2-year nominal yield rises in line with the projections shown in Table 1. Bottom Line: Investors should remain short the 5-year bullet versus a duration-matched 2/10 barbell on the nominal Treasury curve. Investors should also enter 2/10 steepeners on the inflation compensation curve and/or 2/10 flatteners on the real (TIPS) curve. A Quick Update On Emerging Market Bonds The collapse of Chinese real estate behemoth Evergrande dominated headlines for most of the past week, though so far there has been little contagion into USD-denominated credit markets. Unsurprisingly, the spread on Chinese corporate and quasi-sovereign bonds has widened, and there has been some passthrough into both the High-Yield EM USD Corporate & Quasi-Sovereign Index and the High-Yield EM USD Sovereign Index (Chart 9). However, investment grade EM credit spreads have been relatively unaffected by the turmoil, as have US corporate bond spreads (both IG and HY) (Chart 9, bottom 2 panels). While the Evergrande drama – and weakening Chinese economic growth in general – could weigh on USD-denominated EM bond performance in the near-term, we see an excellent buying opportunity in investment grade EM bonds on a 6-12 month horizon. Investment grade EM Sovereigns outperformed credit rating and duration-matched US corporate bonds in the early months of the recovery but have lagged during the past few months (Chart 10). The same is true for the investment grade EM Corporate and Quasi-Sovereign index (Chart 10, panel 3). This is not to say that EM bond performance has been poor – the EM Sovereign index is still up 97 bps versus duration-matched Treasuries on the year – it has simply failed to keep pace with the stellar performance of US corporate bonds. Chart 9No Contagion No Contagion No Contagion Chart 10EM Bonds Versus US Corporates EM Bonds Versus US Corporates EM Bonds Versus US Corporates But that recent stellar performance of US investment grade corporate bonds has left spreads in the sector near historically tight levels. In last week’s report we ran some scenarios for US investment grade corporate bond returns during the next 12 months and concluded that excess returns versus Treasuries are probably capped at 85 bps.4 This makes the current spread advantage in EM Sovereigns (Chart 10, panel 2) and EM Corporates & Quasi-Sovereigns (Chart 10, bottom panel) very compelling. This will be especially true if Emerging Market growth accelerates in 2022 on the back of increased Chinese stimulus and a COVID vaccination campaign that is already picking up steam.5 Bottom Line: US bond investors should shift some allocation out of expensive US investment grade corporate bonds and into USD-denominated investment grade EM Sovereigns and Corporates where valuation is more compelling.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Weekly Report, “2022 Will Be All About Inflation”, dated September 14, 2021. 2 Chart 3B shows that 4 participants expect core PCE inflation of 1.9%-2.0% in 2023, but the SEP also reveals that the minimum forecast for core PCE in 2023 is 2.0%. In other words, all 4 participants in that range are forecasting 2.0%, not 1.9%. 3 For more details on the relationship between our 5-year/5-year forward yield target and the yield curve please see US Bond Strategy Weekly Report, “A Bump On The Road To Recovery”, dated July 27, 2021. 4  Please see US Bond Strategy Weekly Report, “Expected Returns In Corporate Bonds”, dated September 21, 2021. 5 For a recent debate about the outlook for Emerging Market assets please see last Friday’s Webcast: https://www.bcaresearch.com/webcasts/detail/458 Recommended Portfolio Specification Other Recommendations Treasury Index Returns Spread Product Returns

Highlights Fed: The Fed will be forced to clarify its definition of “maximum employment” in 2022, and the path of inflation will ultimately dictate how far the Fed tries to push the labor market. We expect Fed rate hikes to start in December 2022 and that the pace of hikes will proceed more quickly than is currently priced in the yield curve. Duration: Investors should maintain below-benchmark portfolio duration in anticipation of a rate hike cycle starting in December 2022. Yield Curve: Investors should position in Treasury curve flatteners. Specifically, we recommend shorting the 5-year Treasury note versus a duration-matched barbell consisting of the 2-year and 10-year notes. Feature Chart 1Bonds De-Coupled From Inflation In 2021 Bonds De-Coupled From Inflation In 2021 Bonds De-Coupled From Inflation In 2021 One of our themes this year is that US bond investors should pay more attention to the employment data than the inflation data.1 This is because the Fed has successfully convinced markets that it will not lift rates until “maximum employment” is achieved, even if inflation is strong.2 This story has played out during the past few months as bond yields have remained low despite surging prices (Chart 1). Our view is that the muted reaction in bonds is due to the widespread belief that the labor market remains far from “maximum employment” and that rate hikes are therefore a long way off. In this environment, only surprisingly strong employment prints can upset the market’s narrative and send bond yields higher. This playbook for the bond market will continue to function for the next few months. Strong employment data will pull bond yields higher and disappointing employment data will push them down. Inflation prints will be largely irrelevant for the market. But this will change next year. In fact, we see the employment data taking a back seat to the inflation data in the minds of bond investors in 2022. A More Explicit Definition of “Maximum Employment” Must Emerge In 2022 Almost everyone agrees that the US labor market is far from “maximum employment” today, but that will no longer be the case in 2022. The Appendix to this report shows the average monthly nonfarm payroll growth that is required to reach different possible definitions of “maximum employment” by a few specific future dates. For example, we calculate that average monthly nonfarm payroll growth of 414 thousand would cause the unemployment rate to reach 3.8% and the labor force participation rate to reach 63% by the end of 2022. Our sense is that the US economy will be able to add more than 414 thousand jobs per month between now and December 2022. This means that if Fed officials believe that an unemployment rate of 3.8% and a participation rate of 63% meet the definition of “maximum employment”, then they will start to lift interest rates by then. This example sets the scene for what will become next year’s most important monetary policy debate. What constitutes “maximum employment”? Does our example of a 3.8% unemployment rate and a 63% participation rate meet the definition? Or does the Fed have different targets in mind? The Fed will be forced to clarify its position on the topic as the labor market gets closer to reasonable definitions of “maximum employment”. Our sense is that, as of now, there are a range of views on the committee with some FOMC participants taking a more hawkish view of how much slack is left in the labor market and some adopting a more dovish posture. We outline the differences between the hawkish and dovish positions below, but ultimately the path of inflation in 2022 will determine which camp wins out. If inflation remains high next year, then the Fed will be quicker to declare that the labor market is at “maximum employment”, and vice-versa. The Fed’s reliance on the inflation data to settle the argument of what constitutes “maximum employment” will make inflation the most important economic indicator for bond yields in 2022. Labor Market Slack: The Hawkish Case Chart 2The Unemployment Rate Is Falling Fast The Unemployment Rate Is Falling Fast The Unemployment Rate Is Falling Fast The hawkish case for the US labor market reaching “maximum employment” sooner rather than later was outlined nicely last month by our own Bank Credit Analyst.3 First, the Bank Credit Analyst points out that the US labor market was likely beyond “maximum employment” before COVID-19 struck. The implication being that the Fed may move to lift interest rates before the unemployment and participation rates fully recover their pre-pandemic levels. Notice that the unemployment rate (adjusted for the post-COVID surge in people employed but absent from work) was 3.5% in February 2020, well below the Congressional Budget Office’s 4.5% estimate of the natural rate of unemployment (Chart 2).4 Today, the adjusted unemployment rate is 5.5%, not that far above the 3.5%-4.5% range of FOMC participant estimates of the natural rate. If this year’s rate of decline continues, the unemployment rate will hit 4.5% by January 2022 and 3.5% by May 2022. Of course, we know that the Fed takes a broader view of labor market utilization than just the unemployment rate. In particular, we observed sharp declines in labor force participation rates across a wide range of demographic groups when the pandemic struck last year (Chart 3). While the Fed will want to see some improvement in labor force participation, it might be unrealistic to expect the overall labor force participation rate to return to its pre-pandemic level. This is because the aging of the US population imparts a structural downtrend to the participation rate. The dashed line in Chart 4 shows where the participation rate would be if the rate of labor force participation of every individual age cohort remained constant at its February 2020 level. Even in this case, the greater flow of people into the older age groups causes the part rate to fall over time. The message from Chart 4 is that even if the participation rates of every age cohort tracked by the Bureau of Labor Statistics rebound to their February 2020 levels, we would still only expect an overall participation rate of 62.8% by the end of 2022, significantly below the 63.3% seen in February 2020. Chart 3Labor Force Participation By Age Cohort Labor Force Participation By Age Cohort Labor Force Participation By Age Cohort Chart 4The Demographic Downtrend In Participation The Demographic Downtrend In Participation The Demographic Downtrend In Participation On top of the demographic argument, we also notice that the pandemic led to a surge in the number of retired people last year, a number that continues to rise quickly (Chart 5). While we should probably expect some increase in the flow of people coming out of retirement to re-join the labor force as the economy recovers, it’s also logical to assume that there will be at least some hysteresis among the retired population. That is, the longer someone is retired, the less likely they are to re-enter the labor force at all. To the extent that the increase in retired people is sticky, it may be ambitious to expect a full convergence of the 55-year+ part rate back to February 2020 levels (Chart 3, bottom panel). All else equal, this will cause the labor market to reach “maximum employment” more quickly than even our demographic trendline for participation suggests. Chart 5A Surge In Retirees A Surge In Retirees A Surge In Retirees The question of how many FOMC participants agree with the above arguments remains open, but our sense is that there are some who will be eager to declare that “maximum employment” has been achieved before we see a full rebound in the unemployment and participation rates back to pre-COVID levels. For example, Fed Vice-Chair Richard Clarida mentioned the “demographic trend” in labor force participation in his most recent speech.5 Also, Dallas Fed President Robert Kaplan said the following in a recent interview: We’ve had 3 million retirements since February 2020. […] Some of these workers will come back into the workforce, but some of these workers are 55 and older and they’re in reasonably good financial shape and COVID has caused them to re-think whether they really want to re-enter the workforce.6 Labor Market Slack: The Dovish Case There are also good arguments on the side of those who think that an appropriate definition of “maximum employment” involves an unemployment rate closer to 3.5% than 4.5% and a participation rate that does return to pre-COVID levels, and maybe even moves higher. First, a study from the Federal Reserve Bank of Kansas City noted that the bulk of the recent increase in the number of retired people is explained, not by an increase in the number of retirements, but by a reduction in the flow of people from retirement back into the workforce (Chart 6).7 This suggests that pandemic-related health risks are the likely culprit behind the increase in the number of retired people, casting doubt on the idea that the increase in retired people will be sticky. Chart 6Increased Retirees: A Closer Look 2022 Will Be All About Inflation 2022 Will Be All About Inflation Second, there is a strong case to be made that even the February 2020 labor force participation rate is not high enough to meet the definition of “maximum employment”. If we look at the participation rates for 25-54 year old men and women, we see that both were in strong uptrends prior to the pandemic (Chart 7), and there is every reason to believe that they would have continued to move higher if COVID hadn’t cut the recovery short. Chart 7Part Rates Were Rising Pre-Pandemic Part Rates Were Rising Pre-Pandemic Part Rates Were Rising Pre-Pandemic Consider what some FOMC participants were saying prior to the pandemic: The strong labor market is also encouraging more people in their prime working years – ages 25 to 54 – to rejoin or remain in the labor force, […] So far, we have made up more than half the loss in the Great Recession, which translates to almost 2 million more people in the labor force. But prime age participation could still be higher. - Jerome Powell, November 20198 Whether participation will continue to increase in a tight labor market remains to be seen. But I note that male prime-age participation still remains below levels seen in previous business cycle expansions. - Richard Clarida, November 20199 In a more recent interview, Minneapolis Fed President Neel Kashkari expressed skepticism about the idea that labor force participation is destined to remain in a long-run structural downtrend and said that he’s “not convinced we were actually at maximum employment before the COVID shock hit us.” He also said: Getting [labor force participation] and employment-to-population at least back to where they were before [the pandemic], but not necessarily even declaring victory when we do that. I think that’s a reasonable thing for us to try to achieve.10   Inflation: The Ultimate Argument Settler What the above arguments make clear is that there are good reasons to think that the US labor market will reach some policymakers’ definitions of “maximum employment” perhaps by as early as the middle of next year. However, there are also some policymakers who will adopt a more dovish view of what constitutes “maximum employment”. Ultimately, the path of inflation will determine which camp wins out. This is because the entire concept of “maximum employment” is only meaningful when viewed alongside inflation. If employment is pushed beyond its “maximum”, it definitionally means that labor market tightness is leading to unwanted inflationary pressures. With that in mind, the Fed will increasingly refer to the inflation data next year as it tries to make its definition of “maximum employment” more precise. Crucially, what will matter for the Fed (and for the bond market) is where inflation is next year, not where it is right now. Right now, core inflation is well above the Fed’s price stability target, but it is well known that the recent increase in inflation is concentrated in a few sectors – COVID-impacted services and autos – where prices will decelerate as post-pandemic bottlenecks ease (Chart 8). Just as the Fed ignored surging prices in those sectors this year, it will ignore plunging prices in those sectors next year. What will matter for monetary policy is whether core inflation excluding COVID-impacted services and autos remains contained or rises above levels consistent with the Fed’s target (Chart 8, bottom panel). The Fed will also be inclined to declare that “maximum employment” has been achieved if wage growth is accelerating. Currently, there is some evidence of rising wages but also some major supply bottlenecks in the labor market, as evidenced by the all-time high in job openings (Chart 9). Labor supply constraints should ease next year, but the Fed will be watching closely to see if wage growth moderates in kind or continues to increase. Chart 8Watch CPI (ex. COVID-Impacted Services And Autos) In 2022 Watch CPI (ex. COVID-Impacted Services And Autos) In 2022 Watch CPI (ex. COVID-Impacted Services And Autos) In 2022 Chart 9Watch Wages In 2022 Watch Wages In 2022 Watch Wages In 2022 Finally, the Fed will keep a close eye on inflation expectations next year. In particular, it will monitor the Common Inflation Expectations Index and the 5-year/5-year forward TIPS breakeven inflation rate (Chart 10). If either of these indicators break above levels consistent with the Fed’s 2% inflation target, then policymakers will be more inclined to think that “maximum employment” has been attained. Chart 10Watch Inflation Expectations In 2022 Watch Inflation Expectations In 2022 Watch Inflation Expectations In 2022 Bottom Line: The Fed will be forced to clarify its definition of “maximum employment” in 2022, and the path of inflation will ultimately dictate how far the Fed tries to push the labor market. The key indicators to monitor to decide when the Fed will declare that “maximum employment” has been attained are: core inflation excluding COVID-impacted services and autos, wage growth, inflation expectations and the prime-age (25-54) labor force participation rate (Chart 3, panel 2). Investment Implications For bond markets, the question of when the Fed decides that the labor market has reached “maximum employment” is crucial because it will determine the start of the next rate hike cycle. At present, the overnight index swap curve is priced for Fed liftoff in January 2023 and for a total of 78 bps of rate hikes by the end of 2023 (Chart 11). Chart 11Rate Hike Expectations Rate Hike Expectations Rate Hike Expectations Our expectation is that the Fed will start lifting rates in December 2022 and that rate hikes will proceed more quickly than what is currently priced in the market. The unemployment rate will be close to 3.5% by December 2022 and inflation will be sufficiently above the Fed’s target that policymakers will be inclined to view the labor market as at “maximum employment”. Investors should run below-benchmark duration in US bond portfolios to profit from this outcome. We also recommend that investors position for a flatter yield curve by the end of 2022. Specifically, we recommend shorting the 5-year Treasury note versus a duration-matched barbell consisting of the 2-year and 10-year notes. Table 1A shows fair value estimates for the 2-year, 5-year and 10-year yields as of the end of 2022 assuming the market moves to price-in the following path for the fed funds rate: The first 25 bps rate hike occurs in December 2022 Rate hikes proceed at a pace of 100 bps per year The fed funds rate levels-off at a terminal rate of 2.08%11 Table 1ATreasury Curve Fair Value Estimates: December 2022 Liftoff Scenario 2022 Will Be All About Inflation 2022 Will Be All About Inflation In that example, the 2-year and 5-year yields both rise by much more than the 10-year yield and both exceed the change that is priced into the forward curve by more than the 10-year yield. Table 1B shows the results from a similar scenario, the only difference is that the liftoff date is pushed back to March 2023. Both the 2-year and 5-year yields also rise by more than the 10-year yield in this scenario, though the delayed liftoff dampens the relative upside in the 2-year yield. Table 1BTreasury Curve Fair Value Estimates: March 2023 Liftoff Scenario 2022 Will Be All About Inflation 2022 Will Be All About Inflation Bottom Line: Investors should maintain below-benchmark portfolio duration and position in Treasury curve flatteners in anticipation of a rate hike cycle that will start in December 2022. Appendix: How Far From “Maximum Employment” And Fed Liftoff? Chart A1Defining “Maximum Employment” Defining "Maximum Employment" Defining "Maximum Employment" The Federal Reserve has promised that the funds rate will stay pinned at zero until the labor market returns to “maximum employment”. The Fed has not provided explicit guidance on the definition of “maximum employment”, but we deduce that “maximum employment” means that the Fed wants to see the U3 unemployment rate within a range consistent with its estimates of the natural rate of unemployment, currently 3.5% to 4.5%, and that it wants to see a more or less complete recovery of the labor force participation rate back to February 2020 levels (Chart A1). Alternatively, we can infer definitions of “maximum employment” from the New York Fed’s Surveys of Primary Dealers and Market Participants. These surveys ask respondents what they think the unemployment and labor force participation rates will be at the time of Fed liftoff. Currently, the median respondent from the Survey of Market Participants expects an unemployment rate of 3.5% and a participation rate of 63%. The median respondent from the Survey of Primary Dealers expects an unemployment rate of 3.8% and a participation rate of 62.8%. Tables A1-A4 present the average monthly nonfarm payroll growth required to reach different combinations of unemployment rate and participation rate by specific future dates. For example, if we use the definition of “maximum employment” from the Survey of Market Participants, then we need to see average monthly nonfarm payroll growth of +414k in order to hit “maximum employment” by the end of 2022. Table A1Average Monthly Nonfarm Payroll Growth Required For The Unemployment To Reach 4.5% By The Given Date 2022 Will Be All About Inflation 2022 Will Be All About Inflation Table A2Average Monthly Nonfarm Payroll Growth Required For The Unemployment To Reach 4% By The Given Date 2022 Will Be All About Inflation 2022 Will Be All About Inflation Table A3Average Monthly Nonfarm Payroll Growth Required For The Unemployment To Reach 3.5% By The Given Date 2022 Will Be All About Inflation 2022 Will Be All About Inflation Table A4Average Monthly Nonfarm Payroll Growth Required To Reach “Maximum Employment” As Defined By Survey Respondents 2022 Will Be All About Inflation 2022 Will Be All About Inflation Chart A2 presents recent monthly nonfarm payroll growth along with target levels based on the Survey of Market Participants’ definition of “maximum employment”. This chart is to help us track progress toward specific liftoff dates. For example, if monthly nonfarm payroll growth continues to print at the same level as last month, then we could anticipate a Fed rate hike by June 2022. Chart A2Tracking Toward Fed Liftoff Tracking Toward Fed Liftoff Tracking Toward Fed Liftoff We will continue to track these charts and tables in the coming months, and will publish updates after the release of each monthly employment report. Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Weekly Report, “Watch Employment, Not Inflation”, dated June 15, 2021. 2 Specifically, the Fed’s forward guidance states that it will not lift interest rates until (i) inflation is above 2%, (ii) inflation is expected to remain above 2% for some time and (iii) the labor market has reached “maximum employment”. 3 Please see Bank Credit Analyst Special Report, “The Return To Maximum Employment: It May Be Faster Than You Think”, dated August 26, 2021. 4 For details on the adjustment we make to the unemployment rate please see US Bond Strategy Weekly Report, “Overreaction”, dated July 13, 2021. 5 https://www.federalreserve.gov/newsevents/speech/clarida20210804a.htm 6  https://www.bloomberg.com/news/articles/2021-08-09/dallas-fed-president-rob-kaplan-on-the-economy-and-monetary-policy-right-now?sref=Ij5V3tFi 7 https://www.kansascityfed.org/research/economic-bulletin/what-has-driven-the-recent-increase-in-retirements/ 8 https://www.federalreserve.gov/newsevents/speech/powell20191125a.htm 9 https://www.federalreserve.gov/newsevents/speech/clarida20191114a.htm 10 https://www.bloomberg.com/news/articles/2021-08-16/neel-kashkari-on-the-fed-s-quest-to-get-to-full-employment?srnd=oddlots-podcast&sref=Ij5V3tFi 11 We assume a target range of 2% to 2.25% for the terminal fed funds rate. We also assume that the effective fed funds rate trades 8 bps above the lower-end of its target band, as is presently the case. Recommended Portfolio Specification Other Recommendations Treasury Index Returns Spread Product Returns

Highlights Chart 1Employment Growth Will Rebound Employment Growth Will Rebound Employment Growth Will Rebound August’s weak employment growth reflects the surge of Delta variant COVID cases in the United States. This is evidenced by the fact that Leisure & Hospitality sector payrolls held flat in August after having grown by 415k in July and 397k in June (Chart 1). While Delta could still be a drag on employment growth for another month or two, there is mounting evidence that the daily new case count is close to its peak. Leisure & Hospitality employment growth will regain its prior pace as new Delta cases trend down. This will lead to a resumption of strong monthly payroll reports (500k – 1000k) as we head into the new year. For monetary policy, we calculate that average monthly nonfarm payroll growth of 414k will be sufficient for the Fed to start rate hikes before the end of 2022 (bottom panel). We anticipate that this threshold will easily be met. The Treasury curve will bear-flatten as employment growth improves and the market prices-in an earlier start and quicker pace of Fed rate hikes. Investors should maintain below-benchmark portfolio duration and stay short the 5-year Treasury note versus a duration-matched 2/10 barbell. Feature Table 1Recommended Portfolio Specification The Delta Drag The Delta Drag Table 2Fixed Income Sector Performance The Delta Drag The Delta Drag Investment Grade: Neutral Chart 2Investment Grade Market Overview Investment Grade Market Overview Investment Grade Market Overview Investment grade corporate bonds underperformed the duration-equivalent Treasury index by 5 basis points in August, dragging year-to-date excess returns down to +166 bps. The combination of above-trend economic growth and accommodative monetary policy supports continued positive excess returns for spread product versus Treasuries. At 91 bps, the 3-year/10-year Treasury slope remains steep. This is a strong signal that monetary conditions are accommodative. But despite the positive macro back-drop, investment grade valuations are extremely tight (Chart 2). A recent report looked at what different combinations of Treasury slope and corporate spreads have historically signaled for corporate bond excess returns.1 It shows that tight corporate spreads only correlate with negative excess returns once the 3-year/10-year Treasury slope is below 50 bps. Though we retain a positive view of spread product as a whole, better value can be found outside of the investment grade corporate sector. Specifically, we recommend that investors shift into high-yield corporates, municipal bonds and USD-denominated Emerging Market sovereigns and corporates. We also advise investors to favor long-maturity corporate bonds and those corporate sectors with elevated Duration-Times-Spread.2  Table 3ACorporate Sector Relative Valuation And Recommended Allocation* The Delta Drag The Delta Drag Table 3BCorporate Sector Risk Vs. Reward* The Delta Drag The Delta Drag High-Yield: Overweight Chart 3High-Yield Market Overview High-Yield Market Overview High-Yield Market Overview High-Yield outperformed the duration-equivalent Treasury index by 66 basis points in August, bringing year-to-date excess returns up to +502 bps. A recent report looked at the default expectations that are currently priced into the junk index and considered whether they are likely to be met.3 If we demand an excess spread of 100 bps and assume a 40% recovery rate on defaulted debt, then the High-Yield index embeds an expected default rate of 3.0% (Chart 3). Using a model of the 12-month trailing speculative grade default rate that is based on gross corporate leverage (pre-tax profits over total debt) and C&I lending standards, we estimate that the 12-month default rate will fall to between 2.3% and 2.8%, below what the market currently discounts. Notably, the corporate default rate is tracking at an annualized rate of roughly 1.7% through the first seven months of the year, well below the estimate generated by our macro model. Another recent report looked at the incremental spread pick-up investors can earn by moving out of investment grade corporates and into junk.4 It concluded that the extra spread available in high-yield is worth grabbing and that B-rated bonds look particularly attractive in risk-adjusted terms.   MBS: Underweight Chart 4MBS Market Overview MBS Market Overview MBS Market Overview Mortgage-Backed Securities underperformed the duration-equivalent Treasury index by 3 basis points in August, dragging year-to-date excess returns down to -67 bps. The nominal spread between conventional 30-year MBS and equivalent-duration Treasuries tightened 4 bps in August. The spread is wide compared to recent history, but it remains tight compared to the pace of mortgage refinancings (Chart 4). The conventional 30-year MBS option-adjusted spread (OAS) widened 2 bps in August (panel 2), and it is now starting to look attractive compared to other similarly risky spread sectors. The conventional 30-year MBS OAS sits at 38 bps, below the 56 bps offered by Aa-rated corporate bonds but above the 20 bps offered by Aaa-rated consumer ABS and the 35 bps offered by Agency CMBS. In a recent report we looked at MBS performance and valuation across the coupon stack.5 We noted that the higher convexity of high-coupon MBS makes them likely to outperform lower-coupon MBS in a rising yield environment. Higher coupon MBS also have greater OAS than lower coupons. This makes the high-coupon MBS more likely to outperform in a flat bond yield environment as well. Given our view that bond yields will be higher in 6-12 months, we recommend favoring high coupons (4%, 4.5%) over low coupons (2%, 2.5%, 3%) within an overall underweight allocation to Agency MBS. Government-Related: Neutral Chart 5Government-Related Market Overview Government-Related Market Overview Government-Related Market Overview The Government-Related index outperformed the duration-equivalent Treasury index by 27 basis points in August, bringing year-to-date excess returns up to +84 bps (Chart 5). Sovereign debt outperformed duration-equivalent Treasuries by 122 bps in August, bringing year-to-date excess returns up to +7 bps. Foreign Agencies outperformed the Treasury benchmark by 8 bps on the month, bringing year-to-date excess returns up to +44 bps. Local Authority bonds outperformed by 9 bps in August, bringing year-to-date excess returns up to +382 bps. Domestic Agency bonds outperformed by 3 bps, bringing year-to-date excess returns up to +30 bps. Supranationals outperformed by 5 bps, bringing year-to-date excess returns up to +31 bps. USD-denominated Emerging Market (EM) Sovereign bonds outperformed US corporates in August and relative valuation between the two sectors is starting to equalize (panel 4). That said, we retain a preference for EM sovereigns over US corporates, particularly the bonds of Russia, Mexico, Saudi Arabia, UAE and Qatar where value remains attractive. A recent report looked at valuation within the investment grade USD-denominated EM corporate space.6 It found that EM corporates are attractively priced relative to US corporate bonds across the entire investment grade credit spectrum. It also found that EM corporates are attractive relative to EM sovereigns within the A and Baa credit tiers. EM sovereigns have the edge in the Aa credit tier. Municipal Bonds: Overweight Chart 6Municipal Market Overview Municipal Market Overview Municipal Market Overview Municipal bonds underperformed the duration-equivalent Treasury index by 9 basis points in August, dragging year-to-date excess returns down to +262 bps (before adjusting for the tax advantage). The economic and policy back-drop remains favorable for municipal bond performance. Trailing 4-quarter net state & local government savings were already positive through the end of Q1 2021 and they received another significant boost in Q2 as funds from the American Rescue Plan were doled out (Chart 6). With state & local government balance sheets in such good shape, we are comfortable moving down in quality within municipal bonds. A move down in quality is especially compelling because of tight Aaa muni valuations relative to Treasuries (top panel). Valuation is more compelling in the lower investment grade credit tiers, especially at the long-end of the curve.7 GO munis in the 12-17 year maturity bucket offer a 5% breakeven tax rate versus corporates with the same credit rating and duration. 12-17 year Revenue munis actually offer a before-tax yield pick-up (panel 2). Finally, high-yield muni spreads are reasonably attractive relative to high-yield corporates, offering a breakeven tax rate of 23% (panel 4). But despite the attractive spread, we recommend only a neutral allocation to high-yield munis versus high-yield corporates as the deep negative convexity of high-yield munis makes them susceptible to extension risk if bond yields rise. Treasury Curve: Buy 2/10 Barbell Versus 5-Year Bullet Chart 7Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury Yield Curve Overview Treasury yields moved higher in August, with the 5-year and 7-year maturities bearing the brunt of the sell-off. The 2-year/10-year Treasury slope steepened 5 bps to end the month at 110 bps. The 5-year/30-year slope flattened 5 bps to end the month at 115 bps. We expect bond yields to be higher in 6-12 months, but we also anticipate that the next significant move higher in bond yields will coincide with curve flattening, not steepening. At 1.93%, the 5-year/5-year forward Treasury yield is not that far below our target fair value range of 2% - 2.25%. In a recent report, we demonstrated that yield curve steepening only occurs when either the Fed is cutting rates or the 5-year/5-year forward yield rises.8 This means that the 2/10 Treasury curve is more likely to flatten than steepen during the next 6-12 months, even as bond yields move higher. Similarly, we observe that the overnight index swap (OIS) curve is priced for the fed funds rate to be 0.21% in one year’s time and 1.47% in five years (Chart 7). The latter rate has 146 bps of upside if it converges all the way back to its 2018 high, but this pales in comparison to the 265 bps of upside in the 12-month forward rate. The yield curve will flatten as the 12-month forward OIS rate converges with the 5-year forward rate (panel 3). Investors should position in yield curve flatteners on a 6-12 month horizon. Specifically, we recommend shorting the 5-year bullet versus a duration-matched 2/10 barbell.  TIPS: Neutral Chart 8TIPS Market Overview TIPS Market Overview TIPS Market Overview TIPS performed in line with the duration-equivalent nominal Treasury index in August, leaving year-to-date excess returns unchanged at +578 bps. The 10-year and 5-year/5-year forward TIPS breakeven inflation rates both fell by 7 bps in August. At 2.37%, the 10-year TIPS breakeven inflation rate is near the middle of the 2.3% to 2.5% range that is consistent with inflation expectations being well anchored around the Fed’s target (Chart 8). Meanwhile, at 2.21%, the 5-year/5-year forward TIPS breakeven inflation rate is only just below target (panel 3). With long-dated inflation expectations close to the Fed’s target levels, we see limited upside on a 6-12 month investment horizon. We also see the cost of short-maturity inflation protection falling during the next few months as realized inflation continues to moderate from its current extremely high level. This will lead to some modest steepening of the inflation curve (bottom panel). While the inflation curve has some room to steepen, we don’t see it returning to positive territory. An inverted inflation curve is simply more consistent with the Fed’s Average Inflation Target than a positively sloped one. This is because the Fed’s new framework calls for it to attack its inflation target from above rather than from below. ABS: Overweight Chart 9ABS Market Overview ABS Market Overview ABS Market Overview Asset-Backed Securities outperformed the duration-equivalent Treasury index by 3 basis points in August, bringing year-to-date excess returns up to +40 bps. Aaa-rated ABS outperformed by 2 bps on the month, bringing year-to-date excess returns up to +30 bps. Non-Aaa ABS outperformed by 4 bps, bringing year-to-date excess returns up to +92 bps. The stimulus from last year’s CARES Act led to a significant increase in household savings when individual checks were mailed in April 2020. That excess savings has still not been spent and the most recent round of stimulus checks has only added to the stockpile (Chart 9). The extraordinarily large stock of household savings means that the collateral quality of consumer ABS is also extraordinarily high. Indeed, many households have been using their windfalls to pay down consumer debt (bottom panel). Investors should remain overweight consumer ABS and should also take advantage of the high quality of household balance sheets by moving down the quality spectrum.     Non-Agency CMBS: Neutral Chart 10CMBS Market Overview CMBS Market Overview CMBS Market Overview Non-Agency Commercial Mortgage-Backed Securities outperformed the duration-equivalent Treasury index by 6 basis points in August, bringing year-to-date excess returns up to +193 bps. Aaa Non-Agency CMBS outperformed Treasuries by 10 bps in August, bringing year-to-date excess returns up to +92 bps. Non-Aaa Non-Agency CMBS underperformed Treasuries by 9 bps on the month, dragging year-to-date excess returns down to +529 bps (Chart 10). Though returns have been strong and spreads remain attractive, particularly for lower-rated CMBS, we continue to recommend only a neutral allocation to the sector because of the structurally challenging environment for commercial real estate. Agency CMBS: Overweight Agency CMBS outperformed the duration-equivalent Treasury index by 4 basis points in August, bringing year-to-date excess returns up to +91 bps. The average index option-adjusted spread held flat on the month. It currently sits at 35 bps (bottom panel). Though Agency CMBS spreads have recovered to well below pre-COVID levels, they still look attractive compared to other similarly risky spread products. Stay overweight.    Ryan Swift US Bond Strategist rswift@bcaresearch.com Appendix A: 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 August 31st, 2021) The Delta Drag The Delta Drag 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 August 31st, 2021) The Delta Drag The Delta Drag 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 12 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 12 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) The Delta Drag The Delta Drag Appendix B: 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 August 31st, 2021) The Delta Drag The Delta Drag Footnotes 1 Please see US Bond Strategy Weekly Report, “The Post-FOMC Credit Environment”, dated June 29, 2021. 2 For ideas on how to increase the average spread of a US bond portfolio please see US Bond Strategy Weekly Report, “The Collapsing Credit Risk Premium”, dated July 20, 2021. 3 Please see US Bond Strategy Weekly Report, “The Post-FOMC Credit Environment”, dated June 29, 2021. 4 Please see US Bond Strategy Weekly Report, “The Collapsing Credit Risk Premium”, dated July 20, 2021. 5 Please see US Bond Strategy Weekly Report, “A New Conundrum”, dated April 20, 2021. 6 Please see US Bond Strategy Weekly Report, “The Post-FOMC Credit Environment”, dated June 29, 2021. 7 Please see US Bond Strategy Weekly Report, “The Collapsing Credit Risk Premium”, dated July 20, 2021. 8 Please see US Bond Strategy / Global Fixed Income Strategy Weekly Report, “A Bump On The Road To Recovery”, dated July 27, 2021.

Highlights Jackson Hole: The message from Jackson Hole is that the majority of the FOMC – including Fed Chair Powell - is ready to begin tapering asset purchases before year-end. There is less unanimity within the FOMC over the timing of interest rate increases following the taper. Fed Policy: The Fed is trying to communicate a separation of the balance sheet and interest rate components of its monetary policy, hoping to limit bond volatility stemming from markets pulling forward the timing of rate hikes during the taper. A tightening US labor market will make that separation difficult given the shallow path for interest rates currently discounted in the US yield curve - particularly if the current surge in US inflation proves not to be as transitory as the Fed is expecting. US Treasury Yields: Expect higher Treasury yields over the next 12-18 months as the Fed transitions from talking about tapering to actual tapering and, eventually, to rate hikes starting in H2/2022. A September To Remember? Chart 1The Fed Faces Some Tough Decisions The Fed Faces Some Tough Decisions The Fed Faces Some Tough Decisions The much anticipated Jackson Hole speech from Fed Chair Jerome Powell offered a balanced tone.1 Powell did say that the Fed could begin tapering asset purchases by the end of this year, given the “substantial further progress” on the Fed’s 2% average inflation goal, if the US economy evolved in line with the Fed’s forecasts. However, Powell also noted that rate hikes would not occur without greater improvements in the US labor market, particularly given the Fed’s view that the current surge in US inflation will not prove lasting. Several other Fed officials speaking to the media before Powell’s speech hinted at a much more accelerated timetable, with tapering to begin in September and rate hikes potentially starting as soon as mid-2022. The Fed’s messaging is part of an extended conversation with financial markets to prepare for a withdrawal of pandemic-era policy stimulus from quantitative easing (QE). The FOMC is well aware that valuations on asset prices of all stripes have been boosted by loose monetary settings. Powell’s Jackson Hole comments were more nuanced than those of his FOMC colleagues, but this is no surprise as the words of the Fed Chair carry the greatest weight among investors. The Fed Chair does not want to risk a repeat of the 2013 Taper Tantrum in Treasury yields, or the December 2018 plunge in US equity prices, by sounding unexpectedly hawkish and triggering a market rout that tightens US financial conditions (Chart 1). Our baseline assumption has been that the Fed would signal a tapering at the December FOMC meeting and begin to slow asset purchases in January 2022, leading to an eventual liftoff of the fed funds rate by the end of next year. The comments from Powell and others have raised the risk that the Fed moves a bit faster than our expectations on tapering, and perhaps even for liftoff (Chart 2). This would also be faster than the expectations among bond investors. Chart 2The Fed May Be Set To Move Faster Than Our Expected Timeline The Fed’s Separation Anxiety The Fed’s Separation Anxiety The New York Fed’s Survey of Primary Dealers in July showed that tapering is expected by Q1 of next year but a rate hike was not projected until the latter half of 2023 (Table 1). Current pricing in the US overnight index swap (OIS) forward curve is a bit more hawkish than that, with a full 25bp rate hike discounted by January 2023. Table 1Primary Dealers Expect A Taper, Not Rate Hikes The Fed’s Separation Anxiety The Fed’s Separation Anxiety The Fed’s next move will depend on how the questions regarding the Delta variant, the true state of the US labor market and underlying US inflation momentum are resolved. Dismissing The Delta Threat? There has been a clear hit to US economic confidence from the spread of the variant. The August readings from the University of Michigan consumer sentiment survey, the Philadelphia Fed business outlook survey and the ZEW survey of US growth expectations all showed sharp declines (Chart 3). The August flash estimate of the Markit PMIs fell to 8-month and 4-month lows, respectively, indicating that the pace of US economic activity slowed. Higher frequency data like restaurant reservations and hotel bookings have also dipped in recent weeks, potentially a sign of US consumers turning more cautious on leaving home during the Delta surge. Yet there is some tentative positive news on the spread of the variant. The 7-day moving average of new COVID-19 cases in the US appears to be rolling over (Chart 4). In the more stricken states in the US south like Florida, Texas and Louisiana, the effective reproduction number has fallen below one and cases are clearly peaking, suggesting that the transmission of Delta is slowing. If these trends continue, the full hit to US growth from the variant could prove to be minimal and potentially contained to only August data Chart 3A Hit To US Confidence From The Delta Variant A Hit To US Confidence From The Delta Variant A Hit To US Confidence From The Delta Variant Chart 4Has The US Delta Wave ##br##Peaked? Has The US Delta Wave Peaked? Has The US Delta Wave Peaked? Fed officials have been highlighting Delta as a potential near-term risk to the economy, but some comments made last week suggested only a modest level of concern that would not derail tapering plans. For example: Dallas Fed President Robert Kaplan: “[…] what I'm seeing is, in certain sectors, as you would expect, travel-related, you're seeing weakness in some other sectors but by and large, predominantly, what we're seeing is resilience across the indicators that we look at.”2 Kansas City Fed President Esther George: “[…] by and large, I think, unlike what we experienced last year, people have mechanisms to continue to interact with the economy in a way that we didn't before. And so that gives me some confidence in the outlook that we see, that we could continue to push through this.”3 Atlanta Fed President Raphael Bostic: “What I have seen is some suggestion that things are slowing down, but they are still just slowing from extremely high levels. I have not seen big changes in the underlying dynamic.”4 Even Powell himself noted in his speech that “while the Delta variant presents a near-term risk, the prospects are good for continued progress toward maximum employment.” If the hit to the domestic US economy from Delta proves to be modest and short-lived, the Fed will want to see confirmation of this in the US employment data. Labor market slack overestimated? It is clear from other comments made last week that FOMC officials will be watching the August payrolls report very closely, especially given the perception that the US job market may be a lot tighter than the headline unemployment rate suggests. For example, Fed Governor Christopher Waller noted that “when you adjust the labor force for early retirements, if we get another million [jobs in August] we will recover about 85% of the jobs that were lost and that took almost seven years after the last recession.”5 Kaplan noted that “we do think that the labor market is much tighter than the headline statistics indicate. We've had 3 million retirements since February 2020.” Our colleagues at BCA Research’s The Bank Credit Analyst came to a similar conclusion on labor market tightness in a report published last week.6 They determined that the single largest factor driving the US labor force participation rate lower since the onset of the pandemic has been individuals choosing to retire (Chart 5). Only some of that decline has been related to early retirement decisions made in response to COVID. There has been a structural trend of a falling participation rate, by an average of 0.3 percentage points per year, since 2008 due to demographic factors. The labor force participation rate does not need to fully return to pre-pandemic levels for the Fed to conclude that its maximum employment goal has been reached, after accounting for retirements and other demographic shifts (Chart 6). This fits with the comments from Waller and Kaplan indicating that there has likely been enough labor market improvement to begin tapering asset purchases. Chart 5Most Of The Pandemic Decline In Labor Force Participation Has Occurred Due To Retirement The Fed’s Separation Anxiety The Fed’s Separation Anxiety Chart 6Full Employment Without A Pre-COVID Participation Rate Full Employment Without A Pre-COVID Participation Rate Full Employment Without A Pre-COVID Participation Rate Transitory or persistent inflation? In his Jackson Hole speech, Fed Chair Powell downplayed many of the factors that have driven US headline inflation higher in 2021 as “[…] the product of a relatively narrow group of goods and services that have been directly affected by the pandemic and the reopening of the economy.” He also noted that the current surge in durable goods inflation, which has contributed “about one percentage point to the 12-month measures of headline and core inflation”, was likely to end once current supply chain disruptions fade. Durables would then return to the deflationary trend of the past 25 years and help cool off current overheated US inflation. Chart 7US Inflation Is Not Slowing Down US Inflation Is Not Slowing Down US Inflation Is Not Slowing Down Powell also noted the absence of significant US wage growth as reason not to be overly worried about a sustained period of high inflation. He also highlighted that “there is little reason to think” that ongoing structural disinflationary forces like technology and globalization “have suddenly reversed or abated” and that “it seems more likely that they will continue to weigh on inflation as the pandemic passes into history.” This is the message that the Fed has consistently communicated over the past several months, that high inflation was merely “transitory” and the inevitable result of year-over-year base effect comparisons and temporary supply squeezes. The problem with this interpretation is that we are now well into the summer months of 2021, past the period where base effects would be expected to boost US year-over-year inflation rates (the level of both the CPI and PCE deflator indices fell between January and May 2020 before starting to climb again in June). The July 2021 readings on annual headline and core PCE inflation were 4.2% and 3.6%, respectively, the highest rates seen since 1991 (Chart 7, top panel). The year-over-year increase appears to have been concentrated in a few components, with the Dallas Fed’s trimmed mean PCE 12-month inflation for July only climbing to 2.0%. However, the 6-month annualized measure was a more rapid 2.6% - the fastest such pace in 13 years - suggesting that the momentum of US inflation is both broadening and accelerating on the margin (second panel). Chart 8A Sustainable, Not Transitory, Rise In Global Inflation A Sustainable, Not Transitory, Rise In Global Inflation A Sustainable, Not Transitory, Rise In Global Inflation Powell, like many other developed market central bankers, is making a big bet that the “transitory” inflation narrative will prove to be correct and the current surge in inflation will soon subside. Yet already, global supply chain disruptions have lingered longer than the Fed has been expecting. There are also deeper underlying trends in inflation that are challenging the “transitory” narrative. The NFIB small business survey showed that a net 52% of respondents reported raising selling prices in July, while a net 44% planned future price hikes (third panel), both readings last seen during the days of double-digit US inflation in the late 1970s. US firms are successfully passing on rising input costs to US consumers, which is influencing US consumer inflation expectations. The University of Michigan consumer survey for August showed that US households expect inflation over the next year of 4.6% and over the next 5-10 years of 2.9%, with both series well above pre-pandemic lows (bottom panel). The trends in higher inflation seen in the US, and elsewhere, are not just limited to commodity prices where supply squeezes were most prevalent earlier this year and where price momentum is peaking (Chart 8). A GDP-weighted average of core inflation rates for 14 developed market economies reached 2.50% in June and 2.4% in July, levels last seen in the mid-1990s. Higher core inflation readings are consistent with intensifying price pressures stemming from diminished economic slack. The broad swings in our global core inflation measure correlate strongly with the IMF’s estimate of the output gap for the advanced economies (bottom panel). The current acceleration in global core inflation is entirely consistent with the rapid narrowing of the global output gap projected by the IMF for 2021 and, more importantly, 2022. This suggests that underlying inflation pressures, both within and outside the US, will linger into next year, providing an offset the expected drag on “non-core” inflation from slowing commodity price momentum. Already, lingering supply squeezes and stubbornly high US inflation are causing concern among some FOMC members, as noted in these comments last week: Robert Kaplan: “[…] headline PCE inflation next year, we think is going to be in the neighborhood of 2.5%, and there's risk that could be higher. And so we think some of these supply-demand imbalances for materials, some of them will not moderate, but some of them are going to persist longer than people think.” Esther George: “[…] if you continue to have supply constraints and strong demand, you might expect that those will persist more through this year or longer than we originally anticipated.” Chris Waller: “I do think it’s going to be more persistent than I may have thought back in May.” Chart 9Fed Tapering To Deal With Financial Stability Risks? Fed Tapering To Deal With Financial Stability Risks? Fed Tapering To Deal With Financial Stability Risks? Importantly, the senior FOMC leadership - Powell, Lael Brainard, Richard Clarida – has been sticking with the “transitory” narrative. However, even Clarida noted in a speech in early August that he would consider core PCE inflation at or above 3% at year-end to be “much more than a “moderate” overshoot” of the Fed’s 2% inflation objective.7 In his role as Fed Chair, Powell must speak on behalf of the entire FOMC, even if those views are not necessarily his own. Given the growing chorus of Fed voices expressing concern that US inflation could remain higher for longer, it will be increasingly difficult for Powell to do what he did at Jackson Hole – sound more dovish than the individual FOMC members with regards to inflation risks. What about financial stability risks from QE? Fed officials have been understandably cautious in their comments about how QE (and a 0% funds rate) could be influencing asset prices (Chart 9). However, with equity markets at record highs, corporate bond yields near record lows despite high levels of corporate leverage, and US house prices soaring – the S&P CoreLogic Case-Shiller national index rose 18.6% on a year-over-year basis in June, the fastest pace in its 35-year history - it is difficult not to see the role of the Fed’s easy money policies in boosting risk seeking, yield chasing activities. Stimulative financial conditions are also creating future upside growth risks, with the Conference Board leading economic indicator now reaccelerating (bottom panel). Robert Kaplan, Boston Fed President Eric Rosengren and St. Louis Fed President James Bullard have voiced concerns that QE, particularly the Fed’s buying of agency mortgage-backed securities (MBS), have played a significant role in the current US housing boom. The senior FOMC leadership has avoided any such comments for obvious reasons – imagine the market reaction if Powell expressed concerns about high house prices or equity market valuations. However, for those at the Fed already looking to begin tapering sooner, booming asset prices are an additional reason to vote that way as soon as the September FOMC meeting. Separating Tapering From Rate Hikes It seems clear that the majority of the FOMC is now leaning towards starting to taper before year-end, if US growth and employment maintain recent strength. The common message of Fed officials, from Powell on down, is that enough progress has been made on the Fed’s 2% average inflation target objective to justify tapering. Market-based inflation expectations from the TIPS and CPI swap markets are consistent with that interpretation, with breakevens and forward inflation rates within the 2.3-2.5% range consistent with the Fed’s 2% inflation mandate (Chart 10). Yet while our Fed Monitor continues to flag the need for tighter US monetary policy, only 100bps of rate hikes are discounted in the US OIS curve by the end of 2024 – and only after a first rate hike not expected to occur until January 2023. Despite the common messaging on the start of the taper, the Fed voices were singing a bit less in harmony about the potential timing of the first interest rate hike post-taper. Powell went out of his way to note in his Jackson Hole speech that “the timing and pace of the coming reduction in asset purchases will not be intended to carry a direct signal regarding the timing of interest rate liftoff, for which we have articulated a different and substantially more stringent test.” That test, of course, is when the Fed deems that its maximum employment objective has been reached. Can the Fed continue to successfully separate guidance on balance sheet decisions from guidance on future interest rate moves? Current pricing from US OIS and CPI swap forward curves indicates that the market is discounting negative real policy rates, with the Fed never raising rates above inflation, for the next decade (Chart 11). This goes a long way to explain the persistence of negative real US Treasury yields at a time of elevated inflation readings. Although a decade of negative real interest rates is also consistent with the market believing the equilibrium real interest rate (i.e. r-star) is negative – a view currently expressed by no one on the FOMC. Chart 10Too Few Rate Hikes Discounted In The US OIS Curve Too Few Rate Hikes Discounted In The US OIS Curve Too Few Rate Hikes Discounted In The US OIS Curve Chart 11Markets Believe The Fed Will Never Raise Rates Above Inflation Markets Believe The Fed Will Never Raise Rates Above Inflation Markets Believe The Fed Will Never Raise Rates Above Inflation That persistent pricing of negative real rates make sense when there is modest headline inflation and ample spare capacity in the US economy and labor markets. However, that complacency on future rate hikes will be shaken if the US economy approaches full employment and inflation remains above the Fed’s 2% target – outcomes that we expect to occur by the second half of next year. That will lead to the first fed rate hike of the next cycle in Q4 2022, but only after the taper that we expect to start in either December 2021 or January 2022 is completed in Q3 2022. Bottom Line: A tightening US labor market will make the Fed’s current guidance on the separation of tapering from rate hikes increasingly unconvincing, given the shallow path for interest rates currently discounted in the US yield curve - particularly if the current surge in US inflation proves not to be as transitory as the Fed is expecting. Jackson Hole Investment Conclusion – Expect Higher US Treasury Yields Chart 12Stay Below-Benchmark On US Duration Stay Below-Benchmark On US Duration Stay Below-Benchmark On US Duration With such a modest path for future rate hikes, and bond yields, discounted in US forward interest rate curves, we continue to advocate positioning for higher US Treasury yields on a strategic (6-18 months) basis (Chart 12). We see the benchmark 10-year Treasury yield eventually reaching a peak in the 2-2.25% range by the end of 2022. We recommend maintaining a below-benchmark duration stance in the US, while staying underweight US Treasuries in US and global bond portfolios. There is even a case to be made for a more tactical (i.e. shorter-term) bearish stance on US Treasuries with the US data surprise cycle set to turn towards upside surprises, especially if the negative impact of the Delta variant on confidence and spending begins to wane as case numbers start to decline in the coming weeks. Bottom Line: Expect higher Treasury yields over the next 12-18 months as the Fed transitions from talking about tapering to actual tapering and, eventually, to rate hikes starting in H2/2022.   Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com Footnotes 1 A transcript of Powell’s speech can be found here: https://www.federalreserve.gov/newsevents/speech/powell20210827a.htm 2 https://finance.yahoo.com/news/dallas-fed-president-robert-kaplan-yahoo-finance-transcript-august-2021-215700082.html 3 https://finance.yahoo.com/news/kansas-city-fed-president-esther-george-yahoo-finance-transcript-august-2021-113024734.html 4 https://www.reuters.com/business/exclusive-feds-bostic-says-reasonable-begin-bond-buying-taper-october-2021-08-27/ 5 https://finance.yahoo.com/news/fed-gov-waller-strong-august-jobs-report-will-be-green-light-for-taper-202340105.html 6 Please see BCA Research The Bank Credit Analyst September 2021 Section II, “The Return To Maximum Employment: It May Be Faster Than You Think”, available at bca.bcaresearch.com 7 https://www.federalreserve.gov/newsevents/speech/clarida20210804a.htm Recommendations Duration Regional Allocation Spread Product Yields & Returns Global Bond Yields Historical Returns