アメリカ合衆国

Highlights US Vs. Europe: Growth and inflation momentum remains stronger in the US versus Europe. The latter is taking the bigger economic hit from more severe Omicron economic restrictions and a greater exposure to slowing Chinese demand. European inflation has accelerated, but remains slower and less broad-based than elevated US inflation. The backdrop remains more negative for US fixed income compared to Europe. UST-Bund Spread: With markets already priced for multiple Fed rate hikes in 2022, it is now harder to earn significant returns shorting US Treasuries outright compared to 2021. We prefer positioning for higher US bond yields through less-volatile US Treasury-German Bund spread widening positions, with the ECB unlikely to deliver even the single discounted 2022 rate hike. We recommend the position both as a structural allocation in bond portfolios (underweight the US versus Germany) and as a tactical trade (selling US Treasury futures versus Bund futures). Feature Chart of the WeekUS Bond Yields & Bond Volatility Are Both Rising Global fixed income markets are off to a volatile start in 2022, on the back of significant repricing of US interest rate expectations. The 10-year US Treasury yield now sits at 1.85%, up +34bps so far in January and is up +72bps from the August 4/2021 intraday low of 1.13%. The 2-year US yield, which is even more sensitive to changes in Fed expectations, is 1.04%, up +31bps so far this month and up +87bps since early August 2021. Yields are rising in other countries as well, with the 10-year benchmark government bond yield up year-to-date in the UK (+24bps), Canada (+45bps) and even Germany (+18bps) where the Bund yield is threatening to return to positive territory. US Treasuries are selling off as markets have heeded the hawkish shift in the Fed’s interest rate guidance. The US overnight index swap (OIS) curve now discounting 89bps of Fed rate hikes in 2022. Bond volatility further out the Treasury curve has increased as yields have moved higher, with the realized volatility of the Bloomberg 7-10 US Treasury index now at an 19-month high (Chart of the Week). We continue to recommend a defensive strategic posture towards direct US Treasuries with below-benchmark exposure on both duration and country allocations in global bond portfolios. However, we prefer a more efficient way to position for the same theme of rising US yields – betting on a wider 10-year US Treasury-German Bund spread. US Growth & Inflation Fundamentals Support A More Hawkish Fed The rise in global bond yields seen in recent weeks has inflicted damage on risk assets, but not in a consistent fashion. Equity markets have taken the brunt of the hit, with the S&P 500 down around -3% so far in January with the tech-heavy NASDAQ down -6%. Yet the MSCI emerging market equity index is up around +1%, European equities are flat and global high-yield corporate bond spreads are essentially unchanged so far this month. While higher bond yields are reflecting expectations of more global monetary tightening over the next year, medium-term interest rate expectations remain subdued. Our proxy for the market pricing of terminal interest rate expectations – 5-year OIS rates, 5-years forward – remains at or below pre-pandemic levels in the US, the UK, Canada and the euro area (Chart 2). Risk assets are performing relatively well in the face of higher bond yields because markets still do not believe that a major increase in interest rates will be needed in the current global tightening cycle. We see this – the likelihood that interest rates will have to rise much more than markets expect - as the biggest vulnerability for global bond markets over the next couple of years. The US remains the “poster child” for this view. In the US, core CPI inflation accelerated to an 31-year high of 5.5% in December. The pickup in US inflation continues to be broad-based, with the Cleveland Fed median CPI and trimmed mean CPI inflation measures reaching 3.8% and 4.8%, respectively (Chart 3). This massive run-up in US inflation has filtered through to medium-term household inflation expectations; the preliminary University of Michigan consumer survey for January showed that inflation 5-10 years out is expected to be 3.1% - the highest level in 13 years. Chart 2Rising Yields Are Not A Threat To Risk Assets ... Yet​​​​​ Chart 3The Fed Cannot Ignore Elevated Inflation Expectations​​​​​​ Chart 4US Demand Steadily Normalizing From The Pandemic Shock While much of the run-up in US inflation over the past year has been fueled by supply chain disruption and high energy prices, there is still a robust demand component to the high inflation. Consumer spending on goods remains elevated versus its pre-pandemic trend, while services spending is steadily returning back to the pre-pandemic pace (Chart 4). The overall US unemployment rate is now down to 3.9%, the lowest level since February 2020, with broad-based strength in the US labor market across most industries (bottom panel). The rise in consumer inflation expectations has to be most worrisome to Fed officials. Yes, market-based inflation expectations have already seen a significant run-up since the mid-2020 lows, and have even drifted down a bit of late on the back of the more hawkish rhetoric from the Fed. However, survey-based measures of inflation expectations tend to be less volatile than market-based measures, and typically follow trends in realized inflation, which is not slowing down in the US. In other words, rising household inflation expectations are a more reliable indication that an inflationary mindset is becoming entrenched in consumer behavior. US inflation dynamics are transitioning away from supply-driven goods inflation toward more lasting domestically driven forces like tight labor markets, faster wage growth and rising housing costs (Chart 5). Measures of supply chain disruption like global shipping costs are showing signs of peaking (top panel), while commodity price momentum has clearly rolled over – both should eventually feed into slower goods inflation this year. At the same time, tight labor markets will continue to boost US employment costs, which historically have been strongly correlated to US services inflation (middle panel). Chart 5US Inflation Pressures Remain Intense Meanwhile, shelter costs, which represents 32% of the US CPI index, were up 4.2% on a year-over-year basis in December and are likely to continue accelerating given a dearth of housing supply versus demand that is pushing up both house prices and rents (bottom panel). Tying it all together, there are good reasons why the Fed has ramped up the hawkish rhetoric over the past couple of months. However, with the US OIS curve now discounting between 3-4 rate hikes in 2022, it will be harder to generate a second consecutive year of negative returns in the US Treasury market this year. Dating back to the early 1970s, there have only been five calendar years where the Bloomberg US Treasury index delivered an outright negative total return: 1994, 1999, 2009, 2013 and 2021 (Chart 6). None of the four cases prior to last year saw negative returns in the following year, as Treasury yields fell in 1995, 2000, 2010, 2014. Yet even the episodes that saw consecutive years of US yield increases – 1974-75, 1977-81, 1987-88, 2005-06 and 2015-16 – did not see outright negative returns from the Bloomberg US Treasury index. Chart 6Negative Return Years For US Treasuries Are Rare Given the starting point of deeply negative real US bond yields, and interest rate expectations that remain too low beyond 2022, we still see value in staying below-benchmark on US duration exposure on a medium-term basis. However, we see a more efficient way to play for higher Treasury yields this year by positioning US Treasury underweights/shorts versus overweights/longs in government bonds in a region where discounted rate hikes will not happen – Europe. The ECB Is In No Hurry To Hike Rates The same supply driven factors that have pushed up US inflation over the past year have also lifted inflation in the euro area. Headline HICP inflation reached an 30-year high of 5.0% in December, while core HICP inflation hit an all-time high of 2.6%. The European Central Bank (ECB), however, is unlikely to deliver any rate hikes in 2022 even with the high inflation, for several reasons (Chart 7): Growth momentum entering 2022 was soft, thanks to Omicron related economic restrictions at the end of 2021 and also weak demand for European exports from China. It will take time for both of those factors to reverse, thus reducing any growth related pressure to tighten monetary policy. Inflation expectations are not exceeding the ECB 2% inflation target, with the 5-year/5-year forward EUR CPI swap now at 1.9% even with headline inflation of 5.0%. The surge in European energy prices will eventually subside in the first half of 2022, which will reduce inflationary pressure on the ECB to tighten. The ECB is ending its pandemic emergency bond buying program (PEPP) in March, and is only partially replacing that buying activity by upsizing its existing pre-pandemic asset purchase program (APP). The ECB will not want to compound the effect of this “tapering” of bond buying by also hiking interest rates, which would surely tighten financial conditions further through higher Italian government bond yields, rising corporate bond yields and a firmer euro. There is little evidence to date showing any pass-through of higher energy-fueled inflation into more domestically-driven inflation. Euro area wage growth was only 1.3% as of the latest available data in Q3/2021 (which is still well after realized inflation had started to accelerate), highlighting the lack of visible “second round” effects on euro area inflation from high energy prices that would prompt the ECB to consider rate hikes (Chart 8). Chart 7An ECB Rate Hike In 2022 Is Unlikely​​​​​​ Chart 8Limited 'Second Round' Effects From Energy-Driven European Inflation​​​​​​ The EUR OIS curve is discounting 7bps of rate hikes by year-end. Even that modest amount will not be delivered, which will limit how much further European government bond yields will rise this year. A Better Mousetrap: Playing UST Bearishness Through UST-Bund Spread Widening Trades Combining our view of an increasingly hawkish Fed and a still-dovish ECB produces our highest conviction investment recommendation for 2022: positioning for a wider 10-year US Treasury/Germany Bund spread. This can be done by underweighting the US versus core Europe in global bond portfolios, or shorting US Treasury futures versus German Bund futures as we are already recommending in our Tactical Trade Overlay (see page 15). A Treasury-Bund spread widening view is a more efficient way to play for a more hawkish Fed and higher US Treasury yields, for several reasons: There are many examples over past 30 years where the Treasury-Bund spread widened in consecutive years (Chart 9). This is in contrast to the fewer occurrences of consecutive years of rising Treasury yields shown earlier in this report. Thus, there are better odds that last year’s Treasury-Bund spread widening can be repeated in 2022. Chart 9Consecutive Years Of A Rising UST-Bund Spread Happen Often The realized volatility of Treasury-Bund spread trades is almost always lower than that of an outright short position in US Treasuries, but the direction of returns of the two trades is similar (Chart 10). This shows that there is directionality in the Treasury-Bund spread (i.e. it is driven far more by the movements of US yields), but that is a welcome feature given our more bearish view on US Treasuries. The Treasury-Bund spread remains well below fair value on our fundamental valuation model, with fair value increasing due to widening US-European inflation differentials (Chart 11). Tighter relative monetary policies this year (more tapering and rate hikes from the Fed compared to the ECB) also favor a wider fair value spread on our model. Chart 10UST-Bund Wideners Have Lower Volatility Than Outright UST Shorts​​​​​ Chart 11The UST-Bund Spread Looks Very Cheap On Our Model​​​​​​ The gap between our 24-month discounters, which measure the change in policy interest rates over the next two years discounted in OIS curves, for the US and euro area is a reliable leading indicator of the 10-year Treasury-Bund spread (Chart 12, bottom panel). The “discounter spread” is currently calling for the Treasury-Bund spread to widen by more than the current path discounted in US Treasury and German Bund forward rates. Chart 12Position For More UST-Bund Spread Widening In 2022​​​​​​ Chart 13UST-Bund Spread Is Not Technically Stretched​​​​​ The Treasury-Bund spread is not stretched from a technical perspective (Chart 13). The spread is sitting right at its 200-day moving average and the 26-week change in the spread (a measure of price momentum) is rising but remains well below previous peak levels that have capped past spread increases. Summing it all up, the case is strong for including US-Germany spread widening positions as core holdings in investor portfolios in 2022. The current spread is 185bps and we have a year-end target of 225bps. Bottom Line: With markets already priced for multiple Fed rate hikes in 2022, it is now harder to earn significant returns shorting US Treasuries outright compared to 2021. We prefer positioning for higher US bond yields through less-volatile US Treasury-German Bund spread widening positions, with the ECB unlikely to deliver even the single discounted 2022 rate hike. We recommend the position both as a structural allocation in bond portfolios (underweight the US versus Germany) and as a tactical trade (selling US Treasury futures versus Bund futures).   Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com   Recommendations Duration Regional Allocation Spread Product Tactical Trades GFIS Model Bond Portfolio Recommended Positioning     Active Duration Contribution: GFIS Recommended Portfolio Vs. Custom Performance Benchmark The GFIS Recommended Portfolio Vs. The Custom Benchmark Index

Highlights Duration: A look at past rate hike cycles shows that Treasury returns are generally low, though not always negative. For the current cycle, we continue to recommend a below-benchmark portfolio duration stance as we don’t think the full extent of Fed rate hikes is adequately priced in the yield curve. Interest Rate Policy: The Fed will deliver its first rate hike in March and will lift rates 2 or 3 more times this year. We see the fed funds rate moving above 2% this cycle, higher than what is currently priced in the market. Fed Balance Sheet: The Fed will start the passive runoff of its securities holdings in the first half of this year, after one or two rate hikes have been delivered. Balance sheet reduction will proceed more quickly than it did last cycle, but the Fed will refrain from outright sales. Feature Chart 1Market Expectations Are Too Low Rate hikes are just around the corner. In fact, there is a growing consensus among FOMC participants that it will be appropriate to deliver the first rate hike in March, as soon as net asset purchases reach zero. Just last week, San Francisco Fed President Mary Daly called a March rate hike “quite reasonable” and Fed Vice-Chair Lael Brainard testified that the Fed will be “in a position” to lift rates as soon as purchases end. Brainard also mentioned that the Fed has discussed shrinking its balance sheet.1 We expect the Fed to follow through with a 25 basis point rate hike in March, and with 2 or 3 more hikes over the course of 2022. We also see the Fed shrinking its balance sheet this year, via the passive runoff of maturing securities. With all that in mind, this week’s report draws on the experience of past rate hike cycles to give us a sense of what Treasury returns to expect as the Fed lifts rates. We also discuss how the Fed’s balance sheet will evolve over the next few years.  Treasury Returns During Rate Hike Cycles Table 1 provides a useful summary of Treasury returns during the prior four rate hike cycles. The table shows excess Treasury returns versus cash for the Bloomberg Barclays Treasury Index as well as its Intermediate Maturity and Long Maturity sub-indexes. Table 1Treasury Returns During Fed Rate Hike Cycles The first conclusion we draw is that Treasury returns are generally poor during Fed tightening cycles. Intermediate maturity Treasuries underperformed cash in all four cycles. Long maturity Treasuries provided only modestly positive returns in two of the four cycles and deeply negative returns in one of them. One important caveat is that our analysis only considers cycles where the Fed lifted rates multiple times in a row. For example, we exclude the 1997-98 period when one rate hike in 1997 was quickly reversed in 1998. We also define the most recent tightening cycle as spanning from 2015 to 2018 even though the Fed kept the policy rate steady from December 2015 to December 2016. Obviously, if the Fed is forced to abandon its tightening cycle after one or two hikes, then Treasury returns will be much stronger than our historical analysis suggests. Next, let’s dig a bit deeper by looking at each rate hike cycle individually. The 2015-2018 Cycle Chart 22015-2018 Cycle The most recent Fed tightening cycle started with a 25 basis point rate hike in December 2015. The Fed then went on hold for 12 months before delivering a string of 8 hikes between December 2016 and December 2018. All in all, the tightening cycle lasted 36 months and the Fed raised the target fed funds rate by 225 bps, from a range of 0% - 0.25% to a range of 2.25% - 2.5% (Chart 2). If we look at the 36-month discounter on the day before the first hike (Chart 2, panel 3), it shows that the market was priced for 159 bps of tightening over the next three years. The fact that the Fed delivered more tightening (225 bps) explains why excess Treasury returns were negative for short and intermediate maturities. The 5-year/5-year forward Treasury yield is another useful metric because it is a good approximation of the market’s expected terminal fed funds rate, i.e. the fed funds rate at the end of the tightening cycle. The 5-year/5-year forward Treasury yield stood at 2.92% in December 2015, slightly above where the fed funds rate peaked in 2018 (Chart 2, bottom panel). This explains why long-maturity excess Treasury returns were slightly positive during the cycle.         The 2004-2006 Cycle Chart 32004-2006 Cycle During this cycle, which spanned from June 2004 to June 2006, the Fed lifted rates by 400 bps (sixteen 25 basis point rate hikes). The fed funds rate rose from 1% to 5.25% during the two-year span (Chart 3). The 24-month fed funds discounter stood at 369 bps the day before the first hike (Chart 3, panel 3), indicating that the market discounted 31 bps less tightening than was ultimately delivered. Once again, this explains why excess Treasury returns were negative for short and intermediate maturities. The 5-year/5-year forward Treasury yield was 5.72% just prior to the first hike in June 2004 (Chart 3, bottom panel). But, as was the case in the 2015-2018 cycle, the fed funds rate never reached this level. It peaked at 5.25% in 2006 and long-maturity excess Treasury yields were somewhat positive as a result.                 The 1999-2000 Cycle Chart 41999-2000 Cycle In this cycle, the Fed lifted rates by 175 bps between June 1999 and May 2000, driving the fed funds rate from 4.75% to 6.5% (Chart 4). The 12-month fed funds discounter stood at 108 bps on the day before the first hike (Chart 4, panel 3). Once again, this was slightly less than the 175 bps of tightening that transpired. Excess returns for short and intermediate maturity Treasuries were negative as a result. The 5-year/5-year forward Treasury yield was 5.99% on the day before the first hike (Chart 4, bottom panel). This time, the market’s assessment proved to be too low compared to the funds rate’s 6.5% peak. This divergence explains why long-maturity Treasury excess returns were worse during this period than they were in the 2015-18 and 2004-06 cycles.                 The 1994-1995 Cycle Chart 51994-1995 Cycle The Fed surprised markets by lifting rates extremely quickly during this cycle. The Fed moved rates from 3% to 6% in the span of only 12 months between February 1994 and February 1995 (Chart 5). The 12-month discounter was only 130 bps at the beginning of the tightening cycle, well short of the 300 bps rate increase that was delivered (Chart 5, panel 3). This large divergence explains why excess Treasury returns were so poor during this period. Interestingly, the 5-year/5-year forward Treasury yield stood at 6.69% just prior to the first hike (Chart 5, bottom panel), not that far from the ultimate peak in the fed funds rate. In other words, while market expectations for the near-term path of interest rates were too low, expectations for the ultimate peak in interest rates were fairly accurate. However, terminal rate expectations became unmoored when the Fed started to tighten, and the 5-year/5-year forward Treasury yield rose all the way to 8.5%, far above the fed funds rate’s ultimate peak. This dramatic shift in terminal rate expectations explains the deeply negative long-maturity Treasury returns observed during the period. Of course, those losses were quickly reversed in H1 1995 once it became clear that the Fed would not lift rates further. The 5-year/5-year forward Treasury yield plummeted back to 6.5%. Investment Implications Let’s apply the above analysis to today’s situation. At present, the 12-month fed funds discounter stands at 93 bps. The 24-month discounter is 151 bps and the 36-month discounter is 159 bps (Chart 1). In other words, the market is discounting that the Fed will deliver between 3 and 4 rate hikes this year, but only 2 more in 2023 before the funds rate stabilizes at roughly 1.5%. Our expectation is that the fed funds rate will rise to at least 2% during the next three years, and we therefore continue to recommend running below-benchmark portfolio duration. For its part, the 5-year/5-year forward Treasury yield is currently 2.03%. This is at the low-end of survey estimates for the long-run neutral fed funds rate (Chart 1, bottom panel). We expect the 5-year/5-year forward Treasury yield to rise closer to the middle of the range of survey estimates (~2.25%) as it becomes clear that the fed funds rate will rise to at least 2%. It’s also possible that, like in the 1994-95 episode, terminal rate expectations will rise dramatically as the Fed lifts rates more quickly than anticipated. This, however, is not our base case outlook given that expectations for a low terminal fed funds rate are very well entrenched. Bottom Line: A look at past rate hike cycles shows that Treasury returns are generally low, though not always negative. For the current cycle, we continue to recommend a below-benchmark portfolio duration stance as we don’t think the full extent of Fed rate hikes is adequately priced in the yield curve. The Balance Sheet Outlook Chart 6Hike First, Then QT We expect the Fed to start shrinking its securities holdings this year. The process will probably begin in the first half of the year after one or two rate hikes have been delivered. To arrive at this conclusion, we first look at how the Fed proceeded during the last tightening cycle. Back then, the Fed waited until the funds rate was around 1% before it started to shrink its balance sheet in September 2017 (Chart 6). Notably, the Fed didn’t immediately move toward the full passive runoff of its portfolio. Rather, it started slowly by permitting only $6 billion of Treasuries and $4 billion of MBS to mature in October 2017. These amounts were gradually increased in the subsequent months. The Fed will move more quickly toward balance sheet reduction this cycle and the pace of said reduction will be faster. Here are the relevant passages from the minutes of the December FOMC meeting: Almost all participants agreed that it would likely be appropriate to initiate balance sheet runoff at some point after the first increase in the target range for the federal funds rate. However, participants judged that the appropriate timing of balance sheet runoff would likely be closer to that of policy rate liftoff than in the Committee’s previous experience. […] Many participants judged that the appropriate pace of balance sheet runoff would likely be faster than it was during the previous normalization episode. Many participants also judged that monthly caps on the runoff of securities could help ensure that the pace of runoff would be measured and predictable…2 From these quotes, we surmise that balance sheet runoff will start earlier than last time – after one or two rate hikes instead of four. Also, while the runoff will proceed more quickly than last time, there is still support for maintaining monthly caps on the pace. The Fed will probably not move immediately to the complete passive runoff of its portfolio, and outright bond sales do not appear to be part of the discussion. One concern that investors might have about the Fed’s balance sheet runoff is the extra supply of Treasuries that will hit the market. As an upper-bound, if we assume complete passive runoff starting in April 2022, the Fed’s Treasury holdings will shrink from $5.7 trillion today to $3.5 trillion by the end of 2024, adding an average of $715 billion extra Treasury supply to the market each year (Chart 7). If we exclude T-bills and TIPS to focus only on coupon-paying nominal Treasury securities, then we calculate that Fed holdings will fall from $4.9 trillion to $3 trillion, adding an extra $639 billion of supply to the market on average for the next three years. However, it’s important to note that Fed policy alone doesn’t dictate the supply of Treasury securities. The Treasury department’s issuance plans also need to be considered. When the Fed allows a maturing bond to passively roll off its portfolio it doesn’t dump that bond directly into the market. Rather, the Treasury Department issues new debt to replace the maturing bond. The Treasury could decide, for example, to increase T-bill issuance instead of coupon issuance. In fact, this sort of decision becomes more likely if Treasury officials are concerned about dumping too much coupon supply on the market. Currently, the Treasury Department targets a range of 15% - 20% for the amount of outstanding T-bills as a proportion of the overall funding mix, a target that it is hitting (Chart 8). However, the minutes from the most recent Quarterly Refunding meeting stressed that the Treasury feels the need to maintain “flexibility” when it comes to this target range and noted that “there is likely more leeway at the top of the recommended range than at the bottom.”3 Chart 7The Pace Of ##br##Runoff Chart 8T-bill Issuance Could Rise As The Fed's Portfolio Shrinks Finally, it is important to consider the extent to which the Fed will be able to shrink its balance sheet. The Fed’s goal will be to achieve a reserve supply that allows it to maintain the funds rate within its target band without putting undue pressure on either its Overnight Reverse Repo Facility (ON RRP) or its new Standing Repo Facility (SRF). Chart 9The Fed's Balance Sheet Was Too Small In September 2019 The ON RRP acts as a floor on interest rates and its usage therefore increases when the Fed’s balance sheet is too large. The third panel of Chart 9 shows that this is currently the case. Conversely, the SRF acts as a ceiling on interest rates and its usage will ramp up if the Fed’s balance sheet becomes too small. This last occurred in September 2019 when the Fed briefly lost control of interest rates and was forced to increase repo holdings and reserve supply (Chart 9).​​​​​​​ Going forward, the Fed will continue to run down its balance sheet until ON RRP usage drops close to zero. However, it will want to stop reducing its holdings before SRF usage picks up. It is highly uncertain when this will occur, but we suspect that the Fed won’t be able to get the balance sheet back to September 2019 levels before seeing SRF usage increase. Bottom Line: The Fed will start the passive runoff of its securities holdings in the first half of this year, after one or two rate hikes have been delivered. Balance sheet reduction will proceed more quickly than it did last cycle, but the Fed will refrain from outright sales. While the size of the Fed’s balance sheet will shrink during the next few years, it will remain larger than it was in September 2019.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see https://www.reuters.com/business/exclusive-feds-daly-march-liftoff-is-quite-reasonable-2022-01-13/ and https://www.nbcnews.com/business/economy/interest-rate-hike-come-soon-march-feds-brainard-signals-rcna12112 2 https://www.federalreserve.gov/monetarypolicy/files/fomcminutes20211215.pdf 3 https://home.treasury.gov/news/press-releases/jy0464 Recommended Portfolio Specification Other Recommendations Treasury Index Returns Spread Product Returns