先進国

Over the past few days, we have received several questions regarding the decrease in US savings we highlighted on Chart 4 in the most recent Strategy Report. US personal savings have decreased by roughly $4 trillion (SAAR) since their recent peak in March 2021, which raises the question of where all that money went? We must start by looking at how personal savings are computed. Broadly speaking, the savings number is a “stock,” which has income as an inflow and spending outlays as an outflow. Change in either of them changes the level of savings in the economy. The recent decrease in personal savings was due to the decrease in the income component, while spending outlays have remained unchanged. Specifically, the $4 trillion SAAR decrease in savings matches the $4 trillion SAAR decrease in government transfers (Chart 1). The implication is that dissaving merely maintained the same level of consumer spending. Chart 1 In fact, US retail sales have contracted since March 2021, further validating our conclusion that savings did not decrease because of additional spending (Chart 2, top panel). The breakdown of the US retail sales release (Charts 2- 4) shows that the categories which increased are restaurant, miscellaneous retail, clothing stores, food & beverage retail, and gas stations. Spending on durable goods has declined. Chart 2 Chart 3 Chart 4 Bottom Line: The fall in US personal savings from the recent March 2021 peak was driven by a decline in government transfers, and consumers dipping into savings to pay for services and everyday needs.  

Today we take a close look at the historical GICS1 level performance following the taper event in 2013.  Chart 1 provides an overview of a price action of the 10-year US Treasury yield, the US dollar, and gold to provide context, while Charts 2 - 4 summarize performance of the S&P sectors.   Chart 1 Chart 2 Chart 3 Chart 4 The Fed’s decision to modestly reduce the pace of its asset purchases in December of 2013 was a risk-off event which triggered a decline in Treasury yields and put upward pressure on the dollar. S&P 500 sectors followed the script from a risk-off “playbook” with Technology outperforming on the back of falling Treasury rates, while Financials underperformed.  A spike in USD also led to underperformance of the Energy sector. The Consumer Discretionary sector was a notable outlier underperforming the S&P 500 by 6%.  However, empirical analysis is hardly helpful in this case as in 2013 Amazon constituted 7.05% of the sector weight compared to 40% today. Finally, the performance of the defensive sectors was mixed as while tapering was perceived by the market as a clear risk-off event, it was also a sign that the economy is strong, and the Fed is comfortable with withdrawing the liquidity crutch. Bottom Line: Investors should not worry about the Fed and tapering as in the US its effect was short-lived and many more years of the bull market have ensued after it.    

Highlights Investment Grade: Investment grade corporate bond total returns will be close to zero or negative during the next 12 months. The bonds are also likely to outperform duration-matched Treasuries during that period, but excess returns are probably capped at 85 bps. High-Yield: High-yield total returns will fall between -0.29% and +1.80% during the next 12 months, but with a much higher likelihood of being positive than investment grade corporates. Junk will outperform duration-matched Treasuries by between 0.94% and 1.84%, besting the excess returns earned in investment grade. Inflation & The Fed: The Fed will announce asset purchase tapering before the end of this year, and tapering will proceed at a pace that opens the door to a potential rate hike before the end of 2022. Ultimately, whether the Fed lifts rates in 2022 will depend on trends in core CPI excluding COVID-impacted services and autos, along with wage growth and inflation expectations. Feature Chart 1Valuations Are Stretched There are two broad factors that must be considered when deciding whether to favor corporate bonds over Treasuries in a US bond portfolio: (i) The cyclical macroeconomic environment and (ii) valuation. The problem is that, as it stands today, these two factors are sending contrasting signals. The cyclical macroeconomic environment is consistent with strong positive excess returns for spread product versus Treasuries. However, corporate bond spreads and yields are extremely low relative to history (Chart 1). We view the slope of the yield curve as the single best indicator of the cyclical macro environment and have shown in prior research that corporate bonds tend to deliver positive excess returns versus Treasuries when the 3-year/10-year Treasury slope is above 50 bps, even when corporate spreads are tight.1 At present, the 3-year/10-year slope sits at 90 bps and our bias will be toward an overweight allocation to corporates until the slope breaks below 50 bps. A flatter yield curve is negative for corporate bond performance because it suggests that monetary conditions are less accommodative. It also makes it more likely that an unforeseen shock will lead to yield curve inversion, a highly reliable recession indicator. While the macro environment is consistent with continued corporate bond outperformance versus Treasuries, valuation suggests that we should anticipate lower returns than usual from corporate bonds. Table 1 shows annualized corporate bond excess returns during each of the past six cycles. Additionally, it splits each cycle into three phases based on the slope of the 3-year/10-year Treasury curve. Phase 1 of the cycle lasts from the end of the prior recession until the slope breaks below 50 bps. Phase 2 encompasses the period when the slope is between 0 bps and 50 bps. Phase 3 lasts from when the yield curve inverts until the start of the next recession. Table 1Corporate Bond Excess Returns In Different Phases Of The Cycle The first conclusion to draw from Table 1 is that excess returns tend to be lower in Phase 2 than in Phase 1 and lower in Phase 3 than in Phase 2. Second, we see that investment grade corporates have returned an annualized 7.55% in excess of duration-matched Treasuries so far this cycle and high-yield corporates have delivered 15.15% of outperformance. These figures are well above even those seen in prior Phase 1 periods. Based on this, an expectation for lower – but still positive – excess corporate bond returns seems like a reasonable base case for the next 6-12 months. Table 2 is identical to Table 1 except that it shows total returns instead of excess returns. We observe that, so far this cycle, junk bond total returns have outpaced prior Phase 1 periods. Investment grade total returns have been slightly lower given the greater exposure to interest rate risk of those securities. Table 2Corporate Bond Total Returns In Different Phases Of The Cycle As noted above, our expectation is that corporate bonds will outperform Treasuries during the next 6-12 months, but that both excess returns and total returns will take a step down. The next section of this report presents a scenario analysis that puts some more specific numbers on the sorts of excess and total corporate bond returns investors might expect to earn during the next year. Corporate Bond Returns: Scenario Analysis Methodology To run our scenario analysis for investment grade corporate bond returns we use the following equations: Excess Return = OAS0  – D0 (dOAS) Total Return = OAS0+ TSY0 – D0 (dOAS+dTSY) Where: Excess Return = The expected corporate index excess return versus duration-matched Treasuries during the next 12 months Total Return = The expected corporate index total return during the next 12 months OAS0 = Today’s average index option-adjusted spread D0 = Today’s average index duration TSY0 = Today’s Treasury yield that matches the duration of the corporate index dOAS = The expected change in the index option-adjusted spread during the next 12 months dTSY = The expected change in the duration-matched Treasury yield during the next 12 months These equations are obviously simplifications. For example, the impact of convexity is ignored. However, Chart 2 shows that our proxies track actual index returns very closely over time, assuming the estimated yield and spread changes are accurate. Chart 2Estimating IG Returns We use similar equations for assessing high-yield corporate returns, with the additional complication that we must include an assumption for default losses. Excess Return= OAS0 – (DR × (1 - RR)) –D0(dOAS) Total Return= OAS0 + TSY0 – (DR × (1 – RR)) –D0 (dOAS + dTSY) In these equations: DR = The expected issuer-weighted default rate for the next 12 months RR = The expected average recovery rate on defaulted debt for the next 12 months Once again, though these equations are relatively simple, they do a good job of capturing actual returns over time (Chart 3). Chart 3Estimating HY Returns Scenarios With the above equations in hand, we can easily make some educated guesses about future yields, spreads and default losses and translate those assumptions into expected return forecasts. Specifically, we test three different scenarios (bullish, neutral and bearish) for corporate spreads, Treasury yields and default losses. For corporate index spreads, both investment grade and high-yield, our bullish scenario assumes that spreads reach the all-time tight levels seen in the mid-1990s. For investment grade bonds this spread level is 58 bps, 27 bps below the current level. For high-yield bonds this spread level is 233 bps, 41 bps below the current level. Our neutral scenario assumes that index spreads remain at their current levels (85 bps for investment grade and 274 bps for junk). Finally, our bearish scenario assumes that spreads widen back to the average levels seen during the 2017-2019 period (112 bps for investment grade and 369 bps for junk), this implies 27 bps of widening for investment grade and 95 bps of widening for junk. Given our view that bond yields will rise as we approach the next Fed tightening cycle, none of our scenarios assume that Treasury yields will fall during the next 12 months. Our bullish Treasury yield scenario assumes that yields stay flat at current levels. Our neutral Treasury yield scenario assumes that yields follow the path implied by current forward rates, and our bearish Treasury yield scenario assumes that yields rise to levels consistent with fair value estimates assuming the market prices-in a December 2022 Fed liftoff followed by 100 bps of rate hikes per year until the fed funds rate levels-off at 2.08%.2 We use the 7-year and 6-year Treasury yields as our inputs for the investment grade and high-yield scenarios, respectively, as those yields most closely match the interest rate component embedded in the corporate indexes. For default losses, our bullish scenario assumes a 1.8% default rate – consistent with the rate at which defaults are tracking so far this year – and a recovery rate of 50%. Our neutral scenario assumes a 3% default rate and a 40% recovery rate. Our bearish scenario assumes a 4% default rate and 30% recovery rate. Investment Grade Results Table 3 shows the results of our scenario analysis for investment grade corporate bond returns. Table 3Investment Grade Corporate Bond Expected Return Scenarios Starting with excess returns, we think it is most likely that spreads remain near current levels, or perhaps widen a bit, during the next 12 months. We think it’s extremely unlikely that spreads will tighten to the levels seen in the mid-1990s because the average duration of the index is much higher today than it was back then. All else equal, it’s generally true that securities with higher duration also have higher OAS. This means we expect investment grade corporate bond excess returns to be between -153 bps and +85 bps during the next 12 months, probably closer to +85 bps. Obviously, this represents a significant step down from the +550 bps earned during the past year. In fact, even the most bullish scenario where spreads tighten back to all-time lows only implies an excess return of +323 bps, well below the recent rate of outperformance. As for total returns, we estimate that a neutral scenario where the index spread holds steady and Treasury yields follow the forward curve will lead to total returns being close to zero during the next 12 months. In fact, our results suggest that it’s highly likely that investment grade corporate bonds will deliver negative total returns during the next 12 months. Yes, the index is expected to deliver a total return of 1.98% if both the index spread and duration-matched Treasury yield remain at their current levels, but an environment where growth is slow enough to keep Treasury yields flat is much more likely to coincide with spread widening than with steady corporate spreads. For some additional historical perspective, the columns labeled “Historical Percentile Rank” show how the returns in each scenario would rank relative to actual returns earned during the past 31 calendar years. For example, even the most bullish total return scenario of 4.36% ranks at the 27th percentile relative to history. This means that it would only be better than 27% of historical 12-month return observations for that index. High-Yield Results Tables 4A, 4B and 4C summarize the results of our high-yield scenario analysis. Table 4A assumes the bullish scenario for default losses, Table 4B assumes the neutral scenario for default losses and Table 4C assumes the bearish scenario for default losses. Table 4AHigh-Yield Corporate Bond Expected Return Scenarios: Bullish Default Loss Scenario* Table 4BHigh-Yield Corporate Bond Expected Return Scenarios: Neutral Default Loss Scenario* Table 4CHigh-Yield Corporate Bond Expected Return Scenarios: Bearish Default Loss Scenario* Looking at excess returns, the first result that jumps out is that even the most bullish scenario leads to an expected 12-month excess return of +3.43%, this would be equivalent to the median return earned during the past 31 calendar years. In our view, it’s more likely that excess returns will be in the +0.94% to +1.84% range during the next 12 months. This is consistent with flat spreads and a range for default losses between our neutral and bullish scenarios. Our sense is that junk bonds are less likely to deliver negative total returns than investment grade bonds. Though even the most bullish scenario puts expected junk total returns at +4.54%, consistent with the 39th percentile relative to history. Investment Implications To summarize, our expectation is that investment grade corporate bond total returns will be close to zero or negative during the next 12 months. The bonds are also likely to outperform duration-matched Treasuries during that period, but excess returns are probably capped at 85 bps. Our best guess places high-yield total returns at between -0.29% and +1.80%, but with a much higher likelihood of earning positive total returns than a position in investment grade. We estimate that excess junk returns will fall between +0.94% and +1.84%, above returns earned in investment grade. In general, the message is that investors should remain overweight corporate bonds versus Treasuries, but should retain a preference for high-yield over investment grade and should expect to earn far lower returns than were earned during the past year. Given low expected returns, investors should also seek out creative ways of adding additional spread to a bond portfolio. We offered some suggestions in a recent report.3 CPI Update And FOMC Preview This week’s FOMC meeting could be significant for bond markets. First off, there is a possibility that the Fed will announce a timeline for tapering its asset purchases. Our sense is that last month’s weak employment report probably delays this announcement, but we still expect it to come before the end of the year. We expect that the actual tapering of purchases will start in January 2022 and that net Fed purchases will reach zero by Q3 of next year. More broadly, we continue to think that the market is already priced for a tapering announcement in 2021. In other words, any information about asset purchases probably won’t move bond yields that much. What will move bond yields is any hint about when the Fed thinks it may want to start lifting rates. Such news could come in the form of revisions to the Fed’s interest rate forecasts, or in any information that the Fed provides about the pace of asset purchase tapering. Because the Fed has indicated a strong preference for having net purchases at zero prior to liftoff, any pace of tapering that gets net purchases to zero by the middle of next year opens the door to a possible rate hike before the end of 2022. Of course, the economic data between now and the end of 2022 will have a lot to say about whether the Fed actually starts to hike. In particular, last week’s report made the case that next year’s inflation data will determine when rate hikes begin.4 With that in mind, last week’s CPI release showed a significant deceleration in core inflation, driven by the COVID-impacted service and auto sectors that had previously caused inflation to spike (Chart 4). Interestingly, core inflation excluding COVID-impacted services and autos jumped on the month (Chart 4, bottom panel). From the Fed’s perspective, it ignored the transitory rise of COVID-impacted service and auto inflation on the way up, it will also be inclined to ignore its descent. What will ultimately matter for monetary policy is whether underlying inflationary pressures start to build throughout 2022. It is therefore much more important for us to focus on trends in core inflation excluding the COVID-impacted services and autos, along with wage growth and inflation expectations. Our view is that underlying inflationary pressures will be strong enough for the Fed to lift rates before the end of 2022. This will, in large part, be due to an acceleration of shelter inflation (Chart 5). Owner’s Equivalent Rent and Rent of Primary Residence inflation have already jumped, and leading indicators of shelter inflation like the unemployment rate (Chart 5, panel 3) and the Apartment Market Tightness Index (Chart 5, bottom panel) are consistent with further acceleration. Chart 4Looking For Underlying Inflation Chart 5Shelter Inflation Will Keep Rising Bottom Line: The Fed will announce asset purchase tapering before the end of this year, and tapering will proceed at a pace that opens the door to a potential rate hike before the end of 2022. Ultimately, whether the Fed lifts rates in 2022 will depend on trends in core CPI excluding COVID-impacted services and autos, along with wage growth and inflation expectations.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Weekly Report, “The Post-FOMC Credit Environment”, dated June 29, 2021. 2 Last week’s report provides more detail on this fair value analysis. Please see US Bond Strategy Weekly Report, “2022 Will Be All About Inflation”, dated September 14, 2021. 3 Please see US Bond Strategy Weekly Report, “The Collapsing Credit Risk Premium”, dated July 20, 2021. 4 Please see US Bond Strategy Weekly Report, “2022 Will Be All About Inflation”, dated September 14, 2021. Recommended Portfolio Specification Other Recommendations Treasury Index Returns Spread Product Returns