Fixed Income

Special Report

Highlights There is no evidence of a decline in US corporate credit or bank lending spreads over the past few decades, meaning that any excess savings effect structurally depressing interest rates is occurring in the Treasury market. We note the possible mechanisms of action for excess savings to lower government bond yields, by lowering the current policy rate, expectations for the policy rate in the future, or the term premium on long-maturity bonds. To investigate the impact that excess savings may be having on bond yields, we define historical periods of abnormal yields based on the gap between long-maturity Treasury yields and the potential rate of economic growth. This reflects our view that potential growth is the equilibrium interest rate under normal economic conditions. Since 1960, there have been three major episodes when the difference between bond yields and economic growth was large and persistent, but the first two seem to be easily explained by the stance of US monetary policy rather than by a savings/investment imbalance. The excess savings story better fits the facts after 2000. We do find evidence that a global savings glut lowered bond yields during the early-2000s, and it may have even modestly contributed to the excessive household credit demand that ultimately caused the global financial crisis. But as a deviation from equilibrium, the effect of the global savings glut was relatively insignificant compared to what has prevailed over the past decade. Excess savings did certainly play a role in lowering long-term investor expectations for the Federal funds rate during the last economic cycle, but it did so for cyclical reasons that spanned several years rather than as a result of demographic effects or other structural factors unrelated to the business cycle. That is an important distinction, as long-term investor expectations for the Fed funds rate remained low in the second half of the last economic expansion despite a reduction in savings and significantly stronger growth. The historical impact of FOMC meetings on the structural decline in long-maturity US Treasury yields strongly implies that fixed-income investors have been guided by the Fed to expect a lower average Fed funds rate. It is our view that the Fed has a backward-looking neutral rate outlook, informed by an incomplete understanding of the economic circumstances of the latter half of the last expansion. A low neutral rate narrative has become entrenched in the minds of investors and the Fed itself, and we regard this as the primary factor anchoring yields at the long-end of the maturity spectrum. This phenomenon is only likely to dissipate once short-term interest rates rise and a recession does not materialize. While the nearer-term outlook more likely favors a neutral or at best modestly short duration stance within a fixed-income portfolio, investors should remain structurally short duration in response to a potentially rapid shift in long-term interest rate expectations from the Fed and fixed-income investors over the coming few years. Feature Chart II-110-Year US Treasury Yields Are The Lowest Relative To Headline Inflation In Over 60 Years For many investors, one of the most striking features of the pandemic, especially over the past year, is how low US long-maturity government bond yields have remained in the face of the highest headline consumer price inflation in four decades (Chart II-1). To many investors, this has provided even further evidence of a structural “excess savings” effect that has kept interest rates well below the prevailing rate of economic activity. The theory of secular stagnation, revived by Larry Summers in late 2013, is a related concept, but many investors believe that interest rates will remain low even in a world in which the US economy is growing at or even above its trend. The fundamental basis for this view is the idea that over the longer term, the real rate of interest is determined by the balance (or imbalance) between desired savings and investment, and that advanced economies have and will continue to experience excess savings – defined as a chronically high level of desired savings relative to the investment opportunities available. According to this view, in order for the actual level of savings to equal investment, interest rates must fall. Chart II-2Do Excess Savings Explain This Gap? (Spoiler: No) This report challenges the view that excess savings are mostly responsible for the current level of long-term bond yields in the US. We agree that excess savings have played a role in explaining changes in long-term bond yields at different points over the past 20 years; we also agree that it is normal for interest rates in advanced economies to trend down over time in response to a demographically-driven decline in potential growth. But our goal is not to explain the downtrend in interest rates over time. Instead, we aim to explain the gap between the level of long-term bond yields today and the prevailing rate of economic activity, or consensus forecasts of the trend rate of growth (Chart II-2). We do not believe that this gap is economically justified, nor do we believe that it is driven by excess savings. We conclude that the Fed’s backward-looking neutral rate outlook is the primary factor anchoring US Treasury yields at the long-end of the maturity spectrum. This is only likely to change once short-term interest rates rise and a recession does not materialize; it suggests that investors should remain structurally short duration in response to a potentially rapid shift in long-term interest rate expectations from the Fed and fixed-income investors over the coming few years. Excess Savings And Interest Rates: Defining A “Mechanism Of Action” Households, businesses, and governments can directly purchase debt securities in capital markets, but they do not typically provide loans directly to borrowers. Direct lending usually occurs through the banking system, which means that excess savings would only lower interest rates in the economy through one of the following ways: By lowering the Fed funds rate By lowering long-maturity government bond yields relative to the Fed funds rate, by reducing either the term premium or investors’ expectations for the average Fed funds rate in the future By lowering corporate bond yields relative to duration-matched government bond yields By lowering lending rates on bank loans relative to banks’ cost of borrowing Charts II-3-II-5 highlight that there is no evidence of a structural decline in corporate credit spreads or bank lending rates relative to the Fed funds rate, so we can rule out this effect as a mechanism of action for excess savings to have structurally lowered interest rates. Chart II-6 highlights that interest paid on bank deposits lags the Fed funds rate, so we can also rule out the idea that excess deposits force the Fed to keep the effective Fed funds rate low. Chart II-3No Evidence Of A Structural Decline In Corporate Credit Spreads… Chart II-4…Or Auto Loan Rate Spreads… Chart II-5…Or Personal Loan Rate Spreads… Chart II-6...Or Bank Deposit Rate Spreads This means that if excess savings are depressing interest rates in the US, that the effect is truly occurring in the Treasury market. As noted, this could occur by lowering the current policy rate, expectations for the policy rate in the future, or the term premium on long-maturity bonds. Related Report  The Bank Credit AnalystR-star, And The Structural Risk To Stocks All of these effects are certainly possible. Keynes’ paradox of thrift highlights that excess savings can manifest itself as a chronic shortfall in aggregate demand, which would persistently lower the Fed funds rate as the Fed responds to a long period of high unemployment. This could also lower the term premium on long-maturity bond yields in a scenario in which the Fed repeatedly engages in asset purchases to help stabilize aggregate demand. As well, domestic excess savings could lower the term premium on long-maturity bond yields, as aging savers directly purchase government securities as part of their retirement portfolios. Finally, foreign capital inflows could also cause this effect, especially if they originate from countries with chronic current account surpluses that use an increase in US dollar reserves to purchase long-maturity US government securities. Table II-1 summarizes these possible mechanisms of action for excess savings to lower US government bond yields. With these mechanisms in mind, we review the past 60 years to identify periods of “abnormal” bond yields, with the goal of understanding whether excess savings appear to explain major gaps. Table II-1Possible Mechanisms Of Action For Excess Savings To Lower Long-Term Government Bond Yields Identifying Periods Of “Abnormal” Long-Maturity Bond Yields Chart II-7There Have Been Three Distinct Periods Of Abnormal Long-Maturity Bond Yields Chart II-7 shows the difference between nominal 10-year US Treasury yields and nominal potential GDP growth. Panel 2 shows an alternative version of this series using the ten-year median annualized quarterly growth rate of nominal GDP in lieu of estimates of potential growth, which highlights a generally similar relationship. This approach to defining “abnormal” long-maturity bond yields reflects our view that the potential rate of economic growth is the equilibrium interest rate under normal economic conditions. To see why, given that GDP also effectively represents gross domestic income, an interest rate that is persistently below the potential growth rate of the economy would create a strong incentive to borrow on the part of households and especially firms. Chart II-7 makes it clear that the relationship has been mean-reverting over time, but that there have been three major episodes when the difference between bond yields and economic growth was large and persistent. The first episode occurred from 1960 to the late 1970s, and saw government bond yields average well below the prevailing rate of economic growth. We do not see this period as having been caused by an excess of desired savings relative to investment. As we discussed in our November Special Report,1 this gap represented a period of persistently easy monetary policy which contributed to excessive aggregate demand and a structural rise in inflation. The second major episode is also easily explained, as it occurred in response to the first. Following a decade of high inflation, Fed chair Paul Volcker raised interest rates aggressively beginning in 1979 to combat inflationary expectations, which led to a two-decade period of generally tight monetary policy. Like the first period, this was not caused by an imbalance between desired savings and investment. The third episode has prevailed since the late-1990s, and has seen a negative yield/growth gap on average – albeit one that has been smaller than what occurred in the 1960s and 1970s. From 2000 to 2007, the gap was generally negative, although it turned positive by the end of the economic cycle. It was modestly negative on average from 2008 to 2010, and only became persistently negative starting in 2011. The gap fell to a new low during the COVID-19 pandemic, and remains wider today than at any point during the last economic recovery. It is these post-2000 periods of a persistently negative yield/growth gap that should be closely investigated for evidence of an excess savings effect. The Global Savings Glut As noted, prior to 2000, the yield/growth gap in the US seems clearly explained by the Fed’s monetary policy stance, not by an excess savings effect. So the question is whether there is any evidence of excess savings having caused this negative gap since 2000. In our view, the answer is yes, but the effect was relatively small compared to what prevails today. We do find evidence of a global savings glut during the early-2000s. Chart II-8 highlights that the private and external sector savings/investment balances in China and emerging markets more generally were persistently positive during the 2000s. Chart II-9 highlights that multiple estimates of the term premium declined around that time – especially during Greenspan’s “conundrum” period of between 2004 and 2005. Chart II-8There Was A Global Savings Glut Prior To The Global Financial Crisis Chart II-9The Global Savings Glut Does Seem To Have Lowered The Term Premium On US 10-Year TreasurysChart II-10 breaks down the components of the 10-year yield into the 5-year yield and the 5-year/5-year forward yield, and highlights that the negative correlation between the two components lasted for only one year. Overall, the 10-year Treasury yield was lower than potential growth for roughly two years as a result of the global savings glut effect.       Chart II-10Still, The Global Savings Glut Effect Did Not Last Long And Was Not Especially Large In Magnitude This was a significant event, and it may even have modestly contributed to the excessive household credit demand that ultimately caused the global financial crisis. But as a deviation from equilibrium, it was relatively insignificant compared to what has prevailed over the past decade. Excess Savings And US Household Deleveraging Chart II-11Most Of The Post-2007 Decline In 10-Year Yields Is Attributable To Lower Long-Term Fed Funds Rate Expectations Chart II-11 highlights that, relative to June 2007 levels, the vast majority of the cumulative decline in the 10-year Treasury yield has occurred because of a decline in implied long-term expectations for the Fed funds rate, rather than a major decline in the term premium. The chart also shows that almost all the decline in implied long-term interest rate expectations since 2007 occurred during the 2008/2009 recession. This normally occurs during a recession as investors price in a low average Fed funds rate at the short end of the curve; the anomaly is that these expectations remained permanently low even as the economy recovered and as the Fed raised interest rates from 2015 to 2018. To us, Chart II-11 also underscores that the Fed’s asset purchases are not the main culprit behind low long-maturity bond yields today, given that the decline in long-term expectations for the Fed funds rate persisted even as the Fed stopped purchasing assets in 2014. It is not difficult to see why investors lowered their long-term Fed funds rate expectations in the immediate aftermath of the global financial crisis, even as economic recovery took hold. Chart II-12 highlights that the “balance sheet” nature of the 2008/2009 recession unleashed the longest period of US household deleveraging in the post-WWII period, and Chart II-13 highlights that this occurred despite extremely low interest rates – and in contrast to other countries like Canada that did not experience the same loss in household net worth. Chart II-12Household Deleveraging Did Lower The Neutral Rate For Several Years Following The Global Financial Crisis Chart II-13The US Balance Sheet Recession Structurally Impaired Credit Demand For Several Years After 2008     Given that interest rates represent the price of borrowing, it is entirely unsurprising that a US balance sheet recession led to a persistent period in which credit growth was essentially unresponsive to interest rates, as households struggled to rebuild wealth lost during the recession and were unable to, or uninterested in, releveraging. This is another way of saying that the neutral rate of interest fell during that period, which we agree did occur. It is also accurate to characterize the US as having experienced a sharp increase in desired savings over that period, as highlighted by the explosion in the US private sector financial balance in the initial years of the last economic recovery (Chart II-14). Chart II-14Excess Savings Surged After 2008, But Eventually Normalized. Long-Term Rate Expectations Ignored The Normalization. So excess savings did certainly play a role in lowering long-term investor expectations for the Federal funds rate during the last economic cycle, but it did so because of cyclical reasons that spanned several years rather than because of demographic effects or other structural factors unrelated to the business cycle. That is an important distinction, because while Chart II-14 shows that this excess savings effect eventually waned in importance, long-term investor expectations for the Fed funds rate remained low in the second half of the last economic expansion. Chart II-15Growth Was Historically Weak Last Cycle, But Only Because Of The First Few Years Of The Expansion Chart II-15 highlights that the cumulative annualized growth in real per capita GDP during the last economic cycle was significantly below that of the average of previous expansions, but this was only the case because of the very slow growth period between 2008 and 2014. Per capita growth during the latter half of the expansion was comparable to that of previous expansions, and this occurred while the Fed was raising interest rates. And yet, investors only modestly raised their long-term interest rate expectations during that period. In our view, it is this fact that holds the key to understanding why investors’ long-term rate expectations are still low today. An Alternative Explanation For Today’s Extremely Low Long-Maturity Bond Yields Chart II-16Fixed-Income Investors Have Been Guided By The Fed To Expect A Low Average Fed Funds Rate Chart II-16 highlights that, since 1990, all of the structural decline in US 10-year Treasury yields has occurred within a three-day window on either side of FOMC meetings. This strongly suggests that fixed-income investors have been guided by the Fed to expect a low average Fed funds rate, which is consistent with how similar 5-year/5-year forward US Treasury yields are in relation to published FOMC and market participant estimates of the average longer-run Fed funds rate (as shown in Chart II-2). This raises the important question of why the Fed did not revise up its expectation for the neutral rate during or following the second half of the last economic expansion, when growth was much stronger than during the first half. In our view, one of the clearest articulations of the Federal Reserve’s understanding of the neutral rate of interest was presented in a 2015 speech by Lael Brainard at the Stanford Institute for Economic Policy Research. Brainard noted the following: “The neutral rate of interest is not directly observable, but we can back out an estimate of the neutral rate by relying on the observation that output should grow faster relative to potential growth the lower the federal funds rate is relative to the nominal neutral rate. In today’s circumstances, the fact that the US economy is growing at a pace only modestly above potential while core inflation remains restrained suggests that the nominal neutral rate may not be far above the nominal federal funds rate, even now. In fact, various econometric estimates of the level of the neutral rate, or similar concepts, are consistent with the low levels suggested by this simple heuristic approach.”2 Chart II-17The Fed, Wrongly, Sees The 2019 Experience As Having Confirmed A Low Neutral Rate... Given how the Fed determines the neutral rate is, two factors explain why the Fed’s estimates of the neutral rate have not increased (and, in fact, fell modestly in March). First, core inflation remained below 2% from 2015-2019, despite the fact that the economy was clearly growing at an above-trend pace during this period in the face of Fed rate hikes. We have noted in previous reports the role that the 2014 collapse in oil prices had on household inflation expectations. The latter were already vulnerable to a disinflationary shock, given how negative the output gap had been in the first half of the expansion.3 We do not think that the decline in inflation expectations that occurred following the 2014 collapse in oil prices reflects a low neutral rate, but rather we believe that the Fed saw this as a conundrum that supported the expectation of a low average Fed funds rate. The second event explaining the Fed’s persistently low long-term rate expectations is the fact that the Fed was forced to cut interest rates in 2019, which we believe it saw as confirmation that the stance of monetary policy had become either meaningfully less easy or openly tight. From the Fed’s point of view, this perspective was also supported by recessionary indicators, such as the inversion of the 2-10 yield curve (Chart II-17), and popular (but now discontinued) econometric estimates of the real neutral rate of interest, such as those calculated by the Laubach-Williams model (panel 3). Chart II-18...Without Appreciating The Damaging Impact The China-US Trade War Had On Global Activity However, this view entirely ignores the fact that the US and global economies were negatively impacted in 2018 and 2019 by a politically-motivated nonmonetary shock to aggregate demand: the China-US trade war, which also impacted or targeted several major advanced economies. Chart II-18 highlights that global trade uncertainty exploded during this period, which severely damaged business confidence around the world and caused a slowdown in global industrial production. Tighter Chinese policy also likely contributed to the slowdown in global activity, but the bottom line is that factors other than US monetary policy contributed to economic weakness during this period, and that it is incorrect to infer from the 2018/2019 experience that interest rates rose to or exceeded the neutral rate of interest. In short, it is our view that the Fed has simply become backward-looking in how it perceives the neutral rate of interest; it has not yet observed a period when the Fed funds rate has risen to its estimate of neutral but is unambiguously still easy. Fixed-income investors, having demonstrably anchored their own assessments to those of the Fed over the past 30 years, have had no basis to come to a meaningfully different conclusion. We believe that the Fed’s backward-looking low neutral rate outlook has now become entrenched in the minds of investors and the Fed itself, and is the primary factor anchoring yields at the long-end of the maturity spectrum. This will probably only change once short-term interest rates rise and a recession does not materialize. As a final point, we clearly acknowledge that private savings increased massively during the pandemic. Investors who are inclined to see excess savings as the primary driver of low bond yields will point to this fact. But this was a forced increase in savings, rather than a desired one. The rise in household sector savings occurred mostly because of a substantial reduction in services spending, as pandemic restrictions and forced changes in behavior prevented the consumption of many services. The household savings rate has already returned to its pre-pandemic level in the US, and 5-year/5-year forward Treasury yields have risen to a higher point than they were prior to the onset of the COVID-19 pandemic. US households are likely to deploy a portion of their enormous stock of excess savings, as the pandemic continues to recede in importance, which is one of the main reasons to expect that the US economy will not succumb to a recession over the coming 12-18 months – and why investors and the Fed may soon be presented with evidence that warrants an increase in their long-term interest rate expectations. Investment Conclusions There are two important investment implications of the view that the Fed’s backward-looking neutral rate projection is the primary factor anchoring yields at the long end of the maturity spectrum. As we noted in Section 1 of our report, the first implication is that investors will likely be faced with a recession scare as the 2-10 yield curve durably inverts and as rate sensitive sectors of the economy, such as housing, inevitably slow in response to the extremely sharp rise in mortgage rates that has occurred over the past three months. We believe that it is ultimately the level of interest rates that matters for economic activity, rather than the change in interest rates. Large changes over short periods of time, however, create a degree of uncertainty about the trajectory of rates that temporarily impacts economic activity. This underscores that investors should not maintain an aggressively overweight stance toward global equities in a multi-asset portfolio, as it is likely that concerns about corporate profits will increase significantly at some point this year. The second investment implication is that US long-maturity bond yields could increase to much higher levels over the coming 12-24 months than many investors expect, in a scenario in which pandemic-driven price pressure dissipates, real wages recover, and no major politically-driven nonmonetary policy shocks emerge. We acknowledge that long-term interest rate expectations are unlikely to change until hard evidence of the economy’s capacity to tolerate interest rates above the Fed’s implied current estimate of the neutral rate emerges. This is a case, however, when we believe that investors should heed the now-famous words of Rüdiger Dornbusch: “In economics, things take longer to happen than you think they will, and then they happen faster than you thought they could.” As such, while the nearer-term outlook more likely favors a neutral or at best modestly short duration stance within a fixed-income portfolio, investors should remain structurally short duration in response to a potentially rapid shift in long-term interest rate expectations from the Fed and fixed-income investors over the coming few years. Jonathan LaBerge, CFA Vice President The Bank Credit Analyst   Footnotes 1 Please see The Bank Credit Analyst "Gauging The Risk Of Stagflation," dated October 29, 2021, available at bca.bcaresearch.com 2 Lael Brainard, Normalizing Monetary Policy When The Neutral Rate Is Low, December 2015 3 Please see The Bank Credit Analyst "The Modern-Day Phillips Curve, Future Inflation, And What To Do About It," dated December 18, 2020, available at bca.bcaresearch.com

Executive Summary Refreshing Our Tactical Trade List Our current list of tactical trade recommendations centers around two broad themes that predate the Ukraine conflict – rising global inflation expectations and relatively stronger upward pressure on US interest rates. Both themes have been strengthened by the spillovers from the war in Eastern Europe, most notably the link between soaring commodity prices and rising inflation. We still see value in holding our recommended cross-country spread trades that will benefit from continued US bond underperformance (short US Treasuries versus government bonds in Germany, Canada and New Zealand, all at the 10-year maturity). We also maintain our bias to lean against the yield curve flattening trend in the US, but we now prefer to do it solely via our existing SOFR futures calendar spread position. Finding attractively valued inflation breakeven spread trades is more difficult after the latest oil-fueled run-up in developed market inflation expectations. Canadian breakevens, however, stand out as having the greatest upside potential according to our Comprehensive Breakeven Indicators. Bottom Line: Remain in US-Germany, US-Canada an US-New Zealand 10-year government bond yield spread widening trades. Maintain our recommended position in the US SOFR futures curve (long Dec/22 futures, short Dec/24 futures). Add a new inflation-linked bond trade, going long 10-year Canadian breakevens. Feature One month has passed since Russia invaded Ukraine, and investors are still struggling to sort out the financial market implications. Equity markets in the US and Europe have recovered the losses incurred immediately after the conflict began. Equity market volatility has also fallen back to pre-invasion levels according to the VIX index (and its European counterpart, the VStoxx index). That decline in equity volatility has also coincided with a narrowing of corporate credit spreads in both the US and Europe, with the former now fully back to pre-invasion levels. Yet while credit spread volatility has calmed down, government bond yield volatility remains elevated thanks to rising commodity prices putting upward pressure on expectations for inflation and monetary policy (Chart 1). Chart 1Global Bond Yields Are Above Pre-Invasion Levels​​​​​​ Table 1Refreshing Our Tactical Trade List We have already made some “wartime” adjustments to our global bond market cyclical recommendations, with those changes reflected in our model bond portfolio. This week, we review our shorter-term tactical trade recommendations. Our current list of tactical trades revolves around two broad themes that predate the Ukraine conflict – rising global inflation expectations and relatively stronger cyclical upward pressure on US interest rates. Both themes have been strengthened by the spillovers from the war in Eastern Europe, most notably the link between soaring commodity prices and rising inflation. We continue to see the value in holding on to most of our existing tactical trades, with only a couple of adjustments to be made to our US yield curve and global inflation-linked bond positions (Table 1). US Yield Curve Tactical Trades: Shift Focus To SOFR Steepeners We have recommended trades that lean against the aggressive flattening of the US Treasury curve discounted in forward rates since late 2021. Our view has been that markets were discounting too rapid a pace of Fed rate increases in 2022. With the Fed likely delivering fewer hikes than expected, Treasury curve steepening trades would benefit as the spot Treasury curve would flatten by less than implied by the forwards. Related Report  Global Fixed Income StrategyFive Reasons To Tactically Increase US Duration Exposure Now Needless to say, that view has not panned out as we anticipated. The spread between 10-year and 2-year US Treasury yields now sits at a mere +13bps, down from +104bps when we initiated our 2-year/10-year steepener trade last November. The forwards now discount an inversion of that curve starting in June of this year, which would be an extraordinary outcome by historical standards. Typically, the US Treasury curve inverts only after the Fed has delivered an extended monetary tightening cycle that delivers multiple rate hikes over at least a 1-2 year period (Chart 2). Today, the curve has nearly inverted with the Fed having only delivered only a single 25bp rate increase earlier this month. Chart 2The UST Curve Is Unusually Flat Right Now Of course, the Fed’s reaction function in the current cycle is different compared to the past. The Fed now follows an average inflation targeting framework that tolerates temporary inflation overshoots after periods when US inflation ran below the Fed’s 2% target. Now, however, the Fed has no choice but to respond to surging US inflation, which has been accelerating since September and is now at levels last seen in 1982. Chart 3Our SOFR Trade Is Similar To Our UST Curve Trade We still see the market pricing in too much Fed tightening this year and too few rate hikes in 2023/24. The US overnight index swap (OIS) curve now discounts 218bps of rate hikes in 2022, but 44bps of rate cuts between June 2023 and December 2024. We think a more likely scenario is the Fed doing less than discounted this year, as US inflation should show some deceleration in the latter half of 2022, but then continuing to raise rates in 2023 into 2024. We have expressed this view more specifically through an additional tactical trade that was initiated last month, going long the December 2022 3-month SOFR futures contract versus shorting the December 2024 3-month SOFR futures contract. This new trade is essentially a calendar spread trade between two futures contracts, but with a return profile that has looked quite similar to our 2-year/10-year US Treasury curve flattening trade (Chart 3). Having two tactical trades that are highly correlated, and which both are driven by the same theme of the Fed doing less this year and more over the next two years, is inefficient. We see the SOFR calendar spread trade as a more precise expression of our Fed policy view compared to the 2-year/10-year Treasury curve steepener. In addition, the SOFR trade now offers slightly better value after it has lagged the performance of the Treasury curve trade over the past couple of weeks. Thus, we are keeping this trade in our Tactical Overlay portfolio (see the table on page 15), while closing out our 2-year/10-year steepener at a loss of -92bps.1 Cross-Country Spread Trades: Keeping Betting On Relatively Higher US Yields In our Tactical Overlay portfolio, we currently have three recommended cross-country government bond spread trades that all have one thing in common – a sale of 10-year US Treasuries. The long side of the three trades are different (Germany, New Zealand and Canada), but the logic underlying all three trades is the same. The Fed will deliver more rate hikes than the central banks in the other countries. 10-year US Treasury-German Bund spread Chart 4UST-Bund Spread Is Too Low Expecting a wider US Treasury-German Bund spread remains our highest conviction view in G-10 government bond markets. This is a trade we have described as a more efficient way to position for rising US bond yields than a pure below-benchmark US duration stance. We have maintained that recommendation in both our model bond portfolio and our Tactical Overlay portfolio. For the latter, that trade was implemented using 10-year bond futures in both markets and is up 3.9% since initiation back in October 2021. The case for expecting even more Treasury-Bund spread widening remains strong, for several reasons: Underlying inflation remains higher in the US, particularly when looking at domestic sources of inflation like wages and service sector prices. Europe, which relies more heavily on Russia for its energy supplies than the US, is more at risk of a negative growth shock from the Ukraine conflict. Our fundamental model of the 10-year Treasury-Bund spread shows that the current level of the spread (+197bps) is about one full standard deviation below fair value, which itself is rising due to stronger US economic growth, faster US inflation and a more aggressive path for monetary tightening from the Fed relative to the ECB (Chart 4). The spread between our 24-month discounters in the US and Europe, which measure the amount of rate hikes priced into OIS curves for the two regions over the next two years, has proven to be good leading indicator of the 10-year Treasury-Bund spread. That discounter spread is currently at 99bps, levels last seen when the 10-year Treasury-Bund spread climbed to the 250-300bps range in 2017/18 (Chart 5). With the relative forward curves now discounting a slight narrowing of the US-German 10-year spread over the next year, betting on a wider spread does not suffer from negative carry. We are maintaining this trade in our Tactical Overlay portfolio with great conviction. 10-year US Treasury-Canada government bond spread We entered another cross-country spread trade involving a US Treasury short position earlier this month, in this case versus 10-year Canadian government bonds. This trade is a bet on relative monetary policy moves between the Fed and the Bank of Canada (BoC). Like the Fed, the BoC is facing a problem of high inflation and tight labor markets. Canadian core CPI inflation hit a 19-year high of 3.9% in January, while the Canadian unemployment rate is at a 3-year low of 5.5%. The US is facing even higher inflation and even lower unemployment, but one major difference between the two nations is the degree of household sector debt loads. Canada’s household debt/income ratio now stands at 180%, 55 percentage points higher than the equivalent US ratio, thanks to greater residential mortgage borrowing in Canada (Chart 6). Chart 5Stay Positioned For More UST-Bund Spread Widening The Canadian OIS curve is now discounting a peak policy rate of 3.1% in 2023, which is at the high end of the BoC’s estimated 1.75-2.75% range for the neutral policy rate. Chart 6The BoC Will Have Trouble Matching Fed Hawkishness​​​​​ Elevated household debt will limit the BoC’s ability to lift rates that high, as this would trigger a major retrenchment of housing demand and a significant cooling of house prices. While the US is also facing issues with robust housing demand and high house prices, this is less of a factor that would limit Fed tightening relative to the BoC because US household balance sheets are not as levered as their Canadian counterparts. We are keeping our short US/long Canada spread trade (implemented using bond futures) in our Tactical Overlay portfolio, with the BoC unlikely to keep pace with the expected Fed rate increases over the next year (Chart 7). Chart 7Stay Positioned For A Narrower Canada-US Spread 10-year US Treasury-New Zealand government bond spread The third cross-country trade in our Tactical Overlay is 10-year New Zealand-US spread widening trade. Chart 8A Big Gap In NZ-US Relative Interest Rate Expectations​​​​​​ Like the Germany and Canada spread trades, we expect the Fed to deliver more rate hikes than the Reserve Bank of New Zealand (RBNZ) which should push up US Treasury yields versus New Zealand equivalents. In the case of this trade, however, interest rate expectations in New Zealand are far more aggressive. Chart 9Stay Positioned For NZ-US Spread Tightening The RBNZ has already lifted its Official Cash Rate (OCR) by 75bps since starting the tightening cycle in mid-2021. The New Zealand OIS curve is now discounting an additional 253bps of rate hikes in this cycle, eventually reaching a peak OCR of 3.5% in June 2023. This would put the OCR into slightly restrictive territory based on the range of neutral rate estimates from the RBNZ’s various quantitative models (Chart 8). This contrasts to the pricing in the US OIS curve that places the peak in the fed funds rate at 2.8% next year before falling back to the low end of the FOMC’s 2.0-3.0% range of neutral estimates in 2024. Both the US and New Zealand are suffering from similarly high rates of inflation, with New Zealand headline inflation reaching 5.9% in the last available data from Q4/2021. However, while markets are already pricing in restrictive monetary settings in New Zealand, markets are yet to price in a similarly restrictive move in the fed funds rate. We continue to see scope for a narrowing of the New Zealand-US 10-year bond yield spread over at least the next six months. There has already been meaningful compression of the 2-year yield spread as US rate expectations have converged towards New Zealand levels (Chart 9) – we expect the 10-year spread to follow suit. Inflation Breakeven Trades: Swap Canada For Australia We currently have one inflation-linked bond (ILB) trade in our Tactical Overlay portfolio, betting on higher inflation breakevens in Australia. We initiated this trade last October, largely based on the signal from our suite of Comprehensive Breakeven Indicators (CBI) for the major developed economy ILB markets. The CBIs contain three components: the deviation from fair value from our 10-year breakeven spread models, the distance between realized headline inflation and the central bank target, and the gap between the 10-year breakeven and survey-based measures of longer-term inflation expectations. Those three measures are standardized and aggregated to form the CBI. Countries with lower CBIs have more upside potential for breakevens, and their ILBs should be favored over those from nations with higher CBIs. Chart 10Breaking Down Our Comprehensive Breakeven Inflation Indicators Chart 11Favor Canadian Inflation-Linked Bonds Vs. Australia Given the latest run-up in global inflation breakevens on the back of soaring oil prices, there are now no countries in our CBI universe that have a negative CBI (Chart 10). Canada has the lowest CBI, and thus the highest upside potential for breakeven spread widening. We are taking a modest profit of +40bps in our Australian breakeven trade, as we are approaching the self-imposed six-month holding period limit on our tactical trades and our Australian CBI is not indicating major upside for Australian breakevens.2 Based on the message from our indicators, we see a better case for entering a new tactical spread widening position in 10-year Canadian ILBs. A comparison of the CBIs between Canada and Australia shows that the Canadian 10-year inflation breakeven is well below our model-implied fair value, which incorporates both oil prices and currency levels (Chart 11). This contrasts to the Australian breakeven which is now well above fair value. A similar divergence appears when comparing breakeven spreads to survey-based measures of inflation expectations, with Canadian breakevens looking too “undervalued” compared to Australia. While realized headline inflation is above the respective central bank targets, especially in Canada, the valuation cushion makes the ILBs of the latter the better bargain of the two. The details of our new Canadian 10-year breakeven trade, where we go long the cash ILB and sell 10-year Canadian bond futures against it, are shown in our Tactical Overlay table on page 15.   Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com Footnotes 1      The Treasury curve trade is actually a “butterfly” trade, where we have included an allocation to US 3-month Treasury bills (cash) to make the curve steepener duration-neutral. Thus, the trade is more specifically a position where we are long a 2-year US Treasury bullet and short a cash/10-year US Treasury barbell with a duration equal to that of the 2-year. 2     We have recently discovered an error in our how we have calculated the returns on the 10-year Australian futures leg of our Australian 10-year inflation breakeven widening trade. The final total return for our trade shown in the Tactical Overlay table on page 15 corrects for our error, and fortunately shows a significantly higher return than we have published in past reports. 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 Global Fixed Income - Strategic Recommendations* Cyclical Recommendations (6-18 Months) Tactical Overlay Trades

Special Report

Executive Summary An inverted yield curve is a reliable recession indicator. Inversions of the 3-month/10-year Treasury slope and the 3-month/3-month, 18-months forward slope both provide more timely recession signals than inversion of the 2-year/10-year Treasury slope. An inverted yield curve is a reliable equity bear market indicator. Even when it’s not signaling a recession, the yield curve’s movements offer some insight into equity returns as stocks have consistently performed better while it is flattening than they have when it is steepening. The 2-year/10-year Treasury slope embeds useful information for corporate bond excess returns. Corporates perform best when the slope is very steep and worst when it is very flat and/or inverted. Treasury securities generally outperform cash when the yield curve is either very steep or inverted. The one exception is the early-1980s when the Fed continued to tighten aggressively even after an inversion of the yield curve. Different Slopes Are Sending Different Signals Bottom Line: The overall message from the yield curve is that, while the economic recovery is no longer in its early stages, it is premature to talk about a recession. On a 6-to-12 month investment horizon, investors should overweight equities in multi-asset portfolios. Within US bond portfolios, investors should maintain a neutral allocation to investment grade corporate bonds and keep portfolio duration close to benchmark. Feature It’s a well-known maxim in macro-finance that an inverted yield curve signals a recession. While that adage embeds a lot of truth, it is also sufficiently vague that it raises more questions than it answers. How far in advance does an inverted yield curve signal a recession? What specific yield curve segment sends the most helpful signal? And most importantly, does the yield curve tell us anything useful about the future performance of financial assets? These sorts of questions are particularly relevant today as we observe some sections of the yield curve approaching inversion while others make new highs (Chart 1). Chart 1Different Slopes Are Sending Different Signals This Special Report explains how to think about the slope of the US Treasury curve as an indicator for the economy and financial markets. We first examine the yield curve’s empirical track record as a recession indicator. We then consider what the slope of the yield curve tells us about future equity, corporate bond and Treasury returns. The analysis presented in this report focuses on three different measures of the yield curve slope: The 2-year/10-year Treasury slope, the 3-month/10-year Treasury slope and the spread between the 3-month T-bill rate and the 3-month T-bill rate, 18 months forward. That last spread measure is less commonly cited, but Fed research has shown it to be a reliable predictor of recession.1 It was also recently highlighted by Fed Chair Jerome Powell.2 In the remainder of this report we will refer to the 3-month/3-month, 18-month forward spread as the “Fed Slope”. The Yield Curve & Recession Recession forecasting is a tricky business. It is often not so much a question of identifying “good” and “bad” recession indicators, but a question of balancing lead time and reliability. Recession indicators derived from financial market prices tend to offer greater advance warning of recession but also provide more false signals. On the flipside, indicators derived from macroeconomic data tend to give less lead time but with fewer false signals. Typically, the most useful recession indicators involve some combination of financial market and economic data. For example, a 2018 report from our US Investment Strategy service showed that a useful recession indicator can be created by combining the 3-month/10-year Treasury slope and the Conference Board’s Leading Economic Indicator.3 The Treasury slope’s reputation as an excellent recession indicator is justified because, despite it being derived from volatile financial market data, an inversion of the yield curve provides a very reliable recession signal. The 2-year/10-year Treasury slope has inverted in advance of 7 of the past 8 recessions and has not sent a false signal.4  The 3-month/10-year Treasury slope has done even better, calling 8 out of the past 8 recessions without a false signal. The Fed Slope, meanwhile, has also called 8 out of the past 8 recessions, but it sent one false signal in September 1998. There is room to quibble about the usefulness of the yield curve as a recession indicator in terms of lead time. The 2-year/10-year Treasury slope has, on average, inverted 15.9 months before the start of the next recession (Table 1). This inversion has always occurred before the first Fed rate cut of the cycle, and in all but one instance (1973-75), before the peak in the S&P 500. Table 1Lead Times For Yield Curve Segments, Equity Bear Markets And Fed Rate Cuts But while some advance warning is good, the 2-year/10-year slope probably gives too much lead time. For example, the 2-year/10-year slope inverted a full 24 months before the 2007-09 recession, but it would have been unwise to act on that information since the S&P 500 didn’t peak for another 22 months! The historical record shows that the 3-month/10-year Treasury curve and the Fed Slope offer more useful signals than the 2-year/10-year curve. On average, these curves provide less lead time than the 2-year/10-year slope but still generally provide advance warning of recession and stock market peaks. The recession signal from the 3-month/10-year slope has only missed the peak in the S&P 500 twice. The signal from the Fed Slope has only missed the stock market’s peak once, but it also sent one false signal. Synthesizing all this information, we conclude that the 3-month/10-year Treasury curve and the Fed Slope are both highly reliable recession indicators that typically provide more than enough advance warning for equity investors to adjust their positions. The main value of the 2-year/10-year Treasury curve is that its inversion warns that we may soon get a timelier signal from the 3-month/10-year Treasury slope and the Fed Slope. Looking at the present situation, the 2-year/10-year Treasury slope has flattened dramatically during the past few months, but at 18 bps it remains un-inverted. Meanwhile, the 3-month/10-year Treasury slope and the Fed Slope are both elevated at 195 bps and 255 bps, respectively. We can conclude from this that recession warnings are premature. We will become more concerned about an upcoming recession when the 3-month/10-year slope and the Fed Slope approach inversion. The Yield Curve & Equity Returns Identifying a recession and demarcating its beginning and ending dates may seem like a trivial exercise that has little practical import. Celebrated mutual fund manager Peter Lynch has repeatedly offered the opinion that any time an equity investor spends thinking about the economy is wasted time. We beg to differ. Equity bear markets reliably coincide with recessions (Chart 2) – since the late 1960s, only one recession has occurred without a bear market (the first leg of the Volcker double dip from January to July 1980) and only one bear market has occurred without a recession (October 1987’s Black Monday bear market) – and an asset allocator who reduced equity exposure upon receiving advance notice of recessions would have been in a position to generate significant alpha. Chart 2Recessions And Bear Markets Tend To Coincide The relationship between equity returns and the business cycle is not happenstance – variation in stock prices correlates closely with variation in corporate earnings and corporate earnings growth is a function of the business cycle. Equity prices, P, are simply the product of earnings per share, E, and the multiple investors are willing to pay for them, P/E: P = E x (P/E). If we hold somewhat fickle P/E multiples constant, stock prices will rise and fall with earnings. Given that earnings rarely decline outside of recessions (Chart 3), investors can expect equities to rise during expansions and decline during recessions. Chart 3Earnings Grow In Expansions And Fall In Recessions Earnings Declines Outside Of Recessions Are Rare Digging a little more deeply into the empirical record since consensus S&P 500 earnings estimates began to be compiled reinforces the earnings/returns link. With the exception of the first leg of the Volcker double dip recession in 1980, forward four-quarter earnings estimates have fallen in every recession and have contracted in the aggregate at an annualized 16% rate (Table 2). Multiples have expanded at a hearty 9% clip from the beginning to the end of recessions but have always declined, sometimes sharply, during them. Conversely, earnings estimates always grow heartily during expansions, while multiples tend to observe a fairly tight range. Multiples and stocks move ahead of the business cycle, consistently troughing before the end of a recession, but the 20-percentage-point expansion/recession disparity in annualized returns testifies to the yield curve’s utility as an investment leading indicator. Table 2When Earnings Fall, So Do Stocks The yield curve’s usefulness as a predictor of equity returns goes beyond recession signaling. Over the last half-century, the yield curve has tended to steepen and flatten in distinct phases. Defining a phase as a move of at least 200 basis points (bps) between the 3-month/10-year curve slope’s peak and trough, we count ten steepenings and nine flattenings since August 1969 (Chart 4). Chart 450 Years Of Steepening And Flattening After segmenting performance by slope increments in steepening (Table 3, top panel) and flattening (Table 3, middle panel) phases, we find a clear distinction. S&P 500 total returns tend to be much stronger when the yield curve is in a flattening phase than when it is steepening. In general, a steeper curve is better than a flatter (or inverted) one for equity returns but flattening dominates steepening in every segment but the current one (150-200 bps). The cheery news for investors concerned about an inverted yield curve’s effect on stocks is that the upcoming flattening increments between now and inversion have historically been favorable. Though we are tactically neutral equities, we recommend overweighting them in multi-asset portfolios over a cyclical 6-to-12 month timeframe. Table 3Stocks Like A Flattening Yield Curve The Yield Curve & Corporate Bond Returns This section of the report considers investment grade corporate bond returns in excess of a duration-matched position in US Treasuries and whether the slope of the Treasury curve can help us predict their magnitude. First, it’s important to point out that there is a lot of overlap between excess corporate bond returns and equity returns, but it is not complete (Chart 5). Corporates certainly tend to underperform duration-matched Treasuries during recessions and equity bear markets, but there have also been significant bouts of underperformance that fall outside of those periods. For example, corporate bond returns peaked well before equity returns in the late-1990s and corporates also underwent a severe selloff in 2014-15. Chart 5Investment Grade (IG) Corporate Bond Returns By Starting Slope Level And Trend That said, Table 4 shows that, as is the case with stocks, a strategy of reducing corporate bond exposure during recessions will profit over time. Corporate bonds have underperformed Treasuries by a cumulative 3.1% (annualized) during recessions since 1979 and have outperformed by 1.2% (annualized) in non-recessionary periods. Table 4Corporate Bond Performance In And Out Of Recessions The results in Table 4 suggest that investors should remain overweight corporate bonds versus Treasuries at least until the 3-month/10-year Treasury slope inverts. However, we think investors can perform even better if they pay attention to early warning signs from the 2-year/10-year Treasury slope. Table 5 shows historic 12-month corporate bond excess returns given different starting points for the 2-year/10-year slope. The starting points are also split depending on whether the 2-year/10-year slope was in a steepening or flattening trend at the time. Table 512-Month Investment Grade (IG) Corporate Bond Returns By Starting Slope Level And Trend The results presented in Table 5 show that the level of the slope matters much more than whether the curve is in a steepening or flattening trend. They also show that, in general, excess returns tend to be much higher when the slope is steep than when it is flat. We also see that the odds of corporate bonds outperforming duration-matched Treasuries on a 12-month horizon decline markedly for periods when the 2-year/10-year slope starts below 25 bps. At present, the 2-year/10-year Treasury slope is 18 bps, just below the 25-bps cutoff. Though we take this negative signal from the yield curve seriously, we also anticipate that peaking inflation will prevent the Fed from raising rates by 245 bps during the next 12 months, the pace that is currently discounted in the yield curve. If this view pans out it will likely lead to some modest 2/10 curve steepening and a relief rally in corporate spreads. To square the difference between the current negative message from the yield curve and our more optimistic macro view, we recommend a neutral allocation to investment grade corporate bonds within US fixed income portfolios. Though we will likely downgrade our recommended allocation if the 2-year/10-year slope continues to flatten and approaches inversion. The Yield Curve & Treasury Returns This section of the report considers US Treasury index returns in excess of a position in cash, a metric designed to proxy for the returns earned by varying a US bond portfolio’s average duration. Treasury outperformance of cash indicates that long duration positions are profiting while Treasury underperformance of cash indicates that short duration positions are in the green. The historical relationship between Treasury returns and the slope of the yield curve is heavily influenced by the early-1980s period when the Fed plunged the economy into a double-dip recession to contain spiraling inflation. Chart 6 shows that this early-1980s episode is the only one where Treasuries sold off steeply even after all three of our yield curves had inverted. Chart 6A Repeat Of The Early 1980s Episode Remains Unlikely In fact, if we look at the history of 12-month Treasury returns going back to 1973 split by the starting point for the slope (Tables 6A, 6B & 6C), we see that returns are worst after the curve is inverted and best when the curve is very steep. Table 6A12-Month Treasury Excess Returns* Given Different Starting Points For 2-Year / 10-Year Slope Since 1973 Table 6B12-Month Treasury Excess Returns* Given Different Starting Points For 3-Month / 10-Year Slope Since 1973 Table 6C12-Month Treasury Excess Returns* Given Different Starting Points For the Fed Slope** Since 1973 However, that picture changes if we start our historical sample in 1985 to exclude the early-1980s episode. Now, we see that Treasury returns tend to be high when the yield curve is very steep and when it is inverted (Tables 7A, 7B & 7C). The worst 12-month periods for Treasury returns are when the 2-year/10-year slope is between 75 bps and 100 bps, when the 3-month/10-year slope is between 50 bps and 75 bps and when the Fed Curve is between 0 bps and 25 bps. If we apply today’s situation to the post-1985 results shown in Tables 7A, 7B & 7C, we would conclude that the outlook for Treasury returns is very positive. The 3-month/10-year slope and Fed Curve are very steep, and the 2-year/10-year slope is in the 0 bps to 25 bps range. Of course, that message from the 2-year/10-year slope flips if viewed in the context of the post-1973 data shown in Table 6A. Table 7A12-Month Treasury Excess Returns* Given Different Starting Points For 2-Year / 10-Year Slope Since 1985 Table 7B12-Month Treasury Excess Returns* Given Different Starting Points For 3-Month / 10-Year Slope Since 1985 Table 7C12-Month Treasury Excess Returns* Given Different Starting Points For the Fed Slope** Since 1985 Our assessment is that the risk of a repeat of the early-1980s episode is still relatively small. Yes, inflation is extremely high, but it is likely to moderate naturally as we gain more distance from the pandemic. In that environment, the Fed will not feel the need to continue tightening aggressively even after all three segments of the yield curve have inverted. As such, we are inclined to view the message from the yield curve as positive for Treasury returns on a 12-month horizon, and we continue to advocate keeping average bond portfolio duration “at benchmark”.   Ryan Swift US Bond Strategist rswift@bcaresearch.com   Doug Peta, CFA Chief US Investment Strategist dougp@bcaresearch.com Footnotes 1 https://www.federalreserve.gov/econres/notes/feds-notes/dont-fear-the-yield-curve-20180628.htm 2 https://www.bloomberg.com/news/articles/2022-03-21/powell-says-look-at-short-term-yield-curve-for-recession-risk?sref=Ij5V3tFi 3 Please see US Investment Strategy Special Report, “How Much Longer Can The Bull Market Last?”, dated August 13, 2018. 4 We define an instance of “inversion” as a yield curve slope below zero for two consecutive months. Treasury Index Returns Spread Product Returns

Executive Summary Expansion In European Defense European yields have significant upside on a structural basis. European government spending will remain generous, which will boost domestic demand; meanwhile, lower global excess savings will lift the neutral rate of interest and structurally higher inflation will boost term premia. A short-term pullback in yields is nonetheless likely; however, it will not short-circuit the trend toward higher yields on a long-term basis. CYCLICAL INCEPTION DATE RETURN SINCE INCEPTION (%) COMMENT EQUITIES Favor European Aerospace & Defense Over European Benchmark 3/28/2022     Favor European Aerospace & Defense Over Other Industrials 3/28/2022     Bottom Line: Investors should maintain a below-benchmark duration in their European fixed-income portfolios. Higher yields driven by robust domestic demand and strong capex also boost the appeal of industrial, materials, and financials sectors. Aerospace and defense stocks are particularly appealing.     The economic impact of the war in Ukraine continues to drive the day-to-day fluctuations of the market; however, investors cannot ignore the long-term trends in the economy and markets. The direction of bond yields over the coming years is paramount among those questions. Does the recent rise in yields only reflect the current inflationary shock caused by both supply-chain impairments and commodity inflation—that is, is it finite? Or does that rise mirror structural forces and therefore have much further to run? We lean toward yields having more upside over the coming years, propelled higher by structural forces. As a result, we continue to recommend investors structurally overweight sectors that benefit from a rising yield environment, such as financials and industrials, while also favoring value over growth stocks. The defense sector is particularly attractive. Three Structural Forces Behind Higher Yields The current supply-chain disruptions and inflation crises have played a critical role in lifting European yields. However, a broader set of factors underpins our bearish bond view—namely, the lack of fiscal discipline accentuated by the consequences of the Ukrainian war, the likely move higher in the neutral rate of interest generated by lower savings, and the long-term uptrend in inflation. Profligate Governments Chart 1 Larger government deficits will contribute to higher European yields. Europe is not as fiscally conservative as it was before the COVID-19 crisis. Establishment politicians must fend off pressures caused by voters attracted to populist parties willing to spend more. Consequently, IMF estimates published prior to the Ukrainian war already tabulated that, for the next five years, Europe’s average structurally-adjusted budget deficit would be 2.4% of GDP wider than it was last decade (Chart 1). Chart 2Expanding Military Spending The Ukrainian crisis is also prompting a fiscal response that will last many years. Europe does not want to stand still in the face of the Russian threat. Today, Western Europe’s military spending amounts to 1.5% of GDP, or €170 billion. This is below NATO’s threshold of 2% of GDP. Rebuilding military capacity will take large investments. Thus, European nations are likely to move toward that target and even go beyond. Conservatively, if we assume that military spending hits 2% of GDP by the end of the decade, it will rise above €300 billion (Chart 2). Weaning Europe off Russian energy will also prevent a significant fiscal retrenchment. This effort will take two dimensions. The first initiative will be to build infrastructures to receive more LNG from the rest of the world to limit Russian intake. Constructing regasification and storage facilities as well as re-directing pipeline networks be costly and require additional CAPEX over the coming years. The second initiative will be to double-up on green initiatives to decrease the need for fossil fuel. The NGEU funds are already tackling this strategic goal. Nonetheless, the more than €100 billion reserved for renewable energy and energy preservation initiatives was only designed to kick-start hitting the EU’s CO2 emission target for 2050. Accelerating this process not only helps cutting the dependence on Russian energy, but it is also popular with voters. The path of least resistance is to invest in that sphere and to increase such investment beyond the current sums from the NGEU program. The last fiscal push is likely to be more temporary. The UN estimates that four million refugees have left Ukraine, with the vast majority settling in the EU. Accommodating that many individuals will be costly and will add to government spending across the region. Even if mostly transitory, this spending will have an important impact on activity. Larger fiscal deficits push yields higher for two reasons. Greater sovereign issuance that does not reflect a negative shock to the private sector will need to offer higher rates of returns to attract investors. Moreover, greater government spending will boost aggregate demand, which increases money demand. As a result, the price of money will be higher than otherwise, which means that interest rates will rise—as will yields. Decreasing Global Excess Savings Decreasing global excess savings will put upward pressure on the global neutral rate of interest, a phenomenon Peter Berezin recently discussed in BCA’s Global Investment Strategy service. This process will be visible in Europe as well. The US will play an important role in the process of lifting global neutral rates because the dollar remains the foundation of the global financial system. Compared to last decade, the main drag on US savings is that household deleveraging is over. As households decreased their debt load following the global financial crisis, a large absorber of global savings vanished, putting downward pressure on the price of those savings. Today, US households enjoy strong net worth equal to 620% of GDP and have resumed accumulating debt (Chart 3). Consequently, the downward trend in US total private nonfinancial debt loads has ended. The US capex cycle is likely to experience a boost as well. As Peter highlighted, the US capital stock is ageing (Chart 4). Moreover, the past five years have witnessed three events that underscore the fragility of global supply-chains: a disruptive Sino-US trade war, a pandemic, and now a military conflict. This realization is causing firms to move from a “just-in-time” approach to managing supply-chains to a “just-in-case” one. The process of building redundancies and localized supply chains will add to capex for many years, pushing up ex-ante investments relative to savings, and thus, interest rates. Chart 3US Households Are Done Deleveraging Chart 4An Ageing US Capital Stock China’s current account surplus is also likely to decline. For the past two decades, China has been one of the largest providers of savings to the global economy. This is a result of an annual current account surplus that first averaged $150 billion per year from 2000 to 2010 and then $180 billion from 2010 to 2020, and now stands at $316 billion. Looking ahead, China wants to use fiscal policy more aggressively to support demand, which often boosts imports without increasing exports. Also, more domestically-oriented supply chains around the world will limit the growth of Chinese exports. This combination will compress Chinese excess savings, which will place upward pressure on the global neutral rate of interest. Europe is not immune to declining savings. Over the past ten years, the Euro Area current account surplus has averaged €253 billion. Germany’s current account surplus stood at 7.4% of GDP before the pandemic. Those excess savings depressed global rates in general and European ones especially (Chart 5). As in the US, Europe’s capital stock is ageing and needs some upgrade (Chart 6). Moreover, greater government spending boosts aggregate demand. Because investment is a form of derived demand, stronger overall spending promotes capex to a greater extent. Thus, Europe’s public infrastructure push will lift private capex and curtail regional excess savings beyond the original drag from wider fiscal deficits. Additionally, the European population is getting older and will have to tap into their excess savings as they retire. This process will further diminish Europe’s current account surplus, that is, its excess savings. Chart 5Excess Savings Cap Relative Yields Chart 6An Ageing European Capital Stock Too Structurally Higher Inflation BCA believes that the current inflation surge is temporary and mostly reflects a mismatch between demand and supply. However, we also anticipate that, once this inflation climax dissipates, inflation will settle at a level higher than that prior to COVID-19 and will trend higher for the remainder of this decade. Labor markets will tighten going forward because policy rates remain well below neutral interest rates. Output gaps will close because of robust government spending and capex. This will keep wage growth elevated in the US and reanimate moribund salary gains in the Eurozone (Chart 7). This process, especially when combined with less efficient global supply chains and lower excess savings (which may also be thought of as deficient demand), will maintain inflation at a higher level than in the past two decades. Higher inflation will lift yields for two main reasons. First, investors will require both greater long-term inflation compensation and higher policy rates than in the past. Second, higher inflation often generates greater economic volatility and policy uncertainty, which means that today’s minimal term premia will increase over time (Chart 8). Together, these forces will create a lasting upward drift in yields. Chart 7European Wages Will Eventually Revive Chart 8Term Premia Won't Stay This Low Bottom Line: European yields will sport a structural uptrend for the remainder of the decade. Three forces support this assertion. First, European government spending will remain generous, supported by infrastructure and military spending. Second, global excess savings will recede as US consumer deleveraging ends, global capex rises, and the Chinese current account surplus narrows. Europe will mimic this process in response to an ageing population, greater government spending, and capex. Finally, inflation is on a structural uptrend, which will warrant higher term premia across the world. Not A Riskless View There are two main risks to this view, one in the near-term and one more structural. The near-term risk is the most pertinent for investors right now. Global yields may have embarked on a structural upward path, but a temporary pullback is becoming likely. As Chart 9 highlights, the expected twelve-month change in the US policy rate is at the upper limit of its range of the past three decades. Historically, when the discounter attains such a lofty level, a retrenchment in Treasury yields ensues, since investors have already discounted a significant degree of tightening. The same is true in Europe, where the ECB discounter is also consistent with a temporary pullback in German 10-year yields (Chart 10). Chart 9Discounters Point To A Treasury Rally... Chart 10... And A Bund Rally Chart 11A Mixed Message Investor positioning confirms the increasing tactical odds of a yield correction. The BCA Composite Technical Indicator for bonds is massively oversold, which often anticipates a bond rally (Chart 11). This echoes the signals from the JP Morgan surveys that highlight the very low portfolio duration of the bank’s clients. However, the BCA Bond Valuation Index suggests that bonds remain expensive. Together, these divergent messages point toward a temporary bond rally, not a permanent one. The longer-term risk is regularly highlighted by Dhaval Joshi in BCA’s Counterpoint service. Dhaval often shows that the stock of global real estate assets has hit $300 trillion or 330% of global GDP. Real estate is a highly levered asset class and global cap rates have collapsed with global bond yields. With little valuation cushion, real estate prices could become very vulnerable to higher yields. Nevertheless, real estate is also a real asset that produces an inflation hedge. Moreover, rental income follows global household income, and stronger aggregate demand will likely lift median household income especially in an environment in which globalization has reached its apex and populism remains a constant threat. Bottom Line: Global investor positioning has become stretched; therefore, a near-term pullback in yield is very likely, especially as central bank expectations have become aggressive. Nonetheless, a bond rally is unlikely to be durable in an environment in which bonds are expensive and in which growth and inflation will remain more robust than they were last decade. A greater long-term risk stems from expensive global real estate markets. However, real estate is sensitive to global economic activity and inflation, which should allow this asset class ultimately to weather higher yields. Investment Conclusions An environment in which yields rise will inflict additional damage on global bond portfolios. This is especially true in inflation-adjusted terms, since real yields stand at a paltry -0.76% in the US and -2.5% in Germany. Hence, we continue to recommend investors maintain a structural below-benchmark duration bias in their portfolios. Nonetheless, investors with enough flexibility in their investment mandate should take advantage of the expected near-term pullback in yields. Those without this flexibility should use the pullback as an opportunity to shorten their portfolio duration. Higher yields will also prevent strong multiple expansion from taking place; hence, the broad stock market will also offer paltry long-term real returns. Another implication of rising yields, especially if they reflect stronger growth and rising neutral interest rates, is to underweight growth stocks relative to value stocks (Chart 12). Growth stocks are expensive and very vulnerable to the pull on discount rates that follows rising risk-free rates. Meanwhile, stronger economic activity driven by infrastructure spending and capex will help the bottom line of industrial and material firms. Financials will also benefit. Higher yields help this sector and robust capex also boosts loan growth, which will generate a significant tailwind for banking revenues. Hence, rising yields will boost the attractiveness of banks, especially after they have become significantly cheaper because of the Ukrainian war (Chart 13). Chart 12Favor Value Over Growth Chart 13Bank Remain Attractive Related Report  European Investment StrategyFallout From Ukraine Finally, four weeks ago, we highlighted that defense stocks were particularly appealing in today’s context. The re-armament of Europe in response to secular tensions with Russia is an obvious tailwind for this sector. However, it is not the only one. A long-term theme of BCA’s Geopolitical Strategy service is the expanding multipolarity of the world.  The end of an era dominated by a single hegemon (the US) causes a rise in geopolitical instability and tensions. The resulting increase in conflict will invite a pickup in global military spending. Chart 14Defense Will Outshine The Rest European defense and aerospace stocks are expensive, with a forward P/E ratio approaching the top-end of their range relative to the broad market and other industrials. However, their relative earnings are also depressed following the collapse in airplane sales caused by the pandemic. Our bet on the sector is that its earnings will outperform the broad market as well as other industrials because of the global trend toward military buildup. As relative earnings recover their pandemic-induced swoon, so will relative equity prices (Chart 14). Bottom Line: Higher yields warrant a structural below-benchmark duration in European fixed-income portfolios, even if a near-term yield pullback is likely. As a corollary, value stocks will outperform growth stocks while industrials, materials, and financials will also beat a broad market whose long-term real returns will be poor. Within the industrial complex, aerospace and defense equities are particularly appealing because a global military buildup will boost their earnings prospects durably.   Mathieu Savary, Chief European Strategist Mathieu@bcaresearch.com   Tactical Recommendations Cyclical Recommendations Structural Recommendations Closed Trades

Due to travel commitments, there will be no Counterpoint report next week. Instead, we will send you a timely update and analysis of the Ukraine Crisis written by my colleague Matt Gertken, BCA Chief Geopolitical Strategist. Executive Summary The tight connection between the oil price and inflation expectations is intuitive, appealing… and wrong. The inflation market is tiny, and its principle function is not to predict inflation per se, but to serve as a hedging investment in an inflation scare, such as that which follows an oil price spike. Hence, we should treat inflation expectations and the real bond yield that is derived from them with extreme care – especially after an oil price spike, which will give the illusion that the real bond yield is lower than it really is. In the near term, the Ukraine crisis has added to already elevated fears about inflation, which will pressure both bonds and stocks. However, looking beyond the next few months, the Ukraine crisis triggered supply shock will cause demand destruction, while central banks also choke demand, and the recent massive displacement of demand into goods, and its associated inflationary impulse, reverses. The 12-month asset allocation conclusion is to overweight stocks and bonds, and to underweight TIPS and commodities. Fractal trading watchlist: The sell-off in some T-bonds is approaching capitulation. The Tight Relationship Between The Oil Price And Inflation Expectations Is Intuitive, Appealing... And Wrong Bottom Line: In the near term, an inflationary impulse will dominate, but on a 12-month horizon, a disinflationary impulse will dominate. Feature In his seminal work Thinking Fast And Slow, Nobel Laureate psychologist Daniel Kahneman presented the bat-and-ball puzzle. A bat and ball cost $1.10. The bat costs one dollar more than the ball. How much does the ball cost? “A number came to your mind. The number, of course, is 10: 10 cents. The distinctive mark of this easy puzzle is that it evokes an answer that is intuitive, appealing, and wrong. Do the math, and you will see. If the ball costs 10 cents, then the total cost will be $1.20 (10 cents for the ball and $1.10 for the bat), not $1.10. The correct answer is 5 cents. It is safe to assume that the intuitive answer also came to the mind of those who ended up with the correct number – they somehow managed to resist the intuition.” Kahneman’s crucial finding is that many people are prone to place too much faith in an intuitive answer, an intuitive answer that they could have rejected with a small investment of effort. The Connection Between The Oil Price and Inflation Expectations Is Intuitive, Appealing… And Wrong Today, the financial markets are presenting their very own bat-and-ball puzzle. The surging price of crude oil is driving up the market expectation for inflation over the next ten years (Chart I-1). This tight relationship is intuitive and appealing, because we associate a high oil price with a high inflation rate. But the intuitive and appealing relationship is wrong, and it requires just a small investment of effort to prove the fallacy. Chart I-1The Tight Relationship Between The Oil Price And Inflation Expectations Is Intuitive, Appealing... And Wrong Inflation over the next ten years equals the price in ten years’ time divided by the current price. So, to the extent that there is any relationship between the current price and expected inflation, dividing by a higher price today means a lower prospective inflation rate. Empirically, the last fifty years of evidence confirms this very clear inverse relationship (Chart I-2). Chart I-2A High Oil Price Means Lower Subsequent Inflation This raises an obvious question: while many people accept the intuitive (wrong) relationship between the oil price and expected inflation, how can the market make such a glaring error? The answer is that the inflation market is relatively tiny, and that its principle function is not to predict inflation per se, but to serve as a hedging investment in an inflation scare. Compared to the $25 trillion T-bond market, the Treasury Inflation Protected Securities (TIPS) market is worth just $1.5 trillion, slightly more than the market capitalisation of Tesla. Just as we do not expect Tesla to represent the view of the entire stock market, we should not expect TIPS to represent the view of the entire bond market. A high oil price means lower subsequent inflation. A recent paper by The Oxford Institute For Energy Studies explains: “the tight relationship between the oil price and inflation expectations defies not only the thesis of economics, but the norms of statistics as well, with a correlation that has reached 90 percent over the last ten years and a corresponding r-squared of 82 percent (Chart I-3 and Chart I-4). The root cause of this phenomenon should probably be searched for in the behaviour of another large group of market participants, the systematic portfolio allocators, and factor investors.”1  Chart I-3Inflation Expectations Are Just A Mathematical Function Of The Oil Price... Chart I-4...Therefore 'The Real Bond Yield' Is Just A Mathematical Function Of The Oil Price So, here’s the explanation for the intuitive, appealing, but wrong connection between the oil price and inflation expectations. In the inflation scare that a surging oil price unleashes, the two main asset-classes – bonds and equities – are vulnerable to sharp losses, leaving TIPS as one of the very few assets that can provide a genuine hedge against inflation. But given that bonds and equities dwarf the $1.5 trillion TIPS (and other inflation) markets, the inflation hedger quickly becomes the dominant force in this tiny market. This large volume of hedging demand chasing limited supply drives down the real yields on TIPS to artificial lows, both in absolute terms and relative to T-bond yields. And as the difference between nominal and real yields defines the ‘market’s expected inflation’, it explains the surge in expected inflation. Be Careful How You Use ‘The Real Bond Yield’ It is an unfortunate reality that we often close the stable door after the horse has bolted, meaning that we react after, rather than before, the event. In financial market terms, this means that we demand inflation protection after, rather than before, it happens, and end up overpaying for it. A high oil price unleashes a massive hedging demand for the tiny TIPS market, driving down the real TIPS yield versus the nominal T-bond yield. To repeat, a high oil price unleashes a massive hedging demand for the tiny TIPS market, driving down the real TIPS yield versus the nominal T-bond yield. The upshot is that the performance of TIPS versus T-bonds is nothing more than a play on the oil price (Chart I-5). Chart I-5The Performance Of TIPS Versus T-Bonds Is Just A Play On The Oil Price A bigger message is that we should interpret the oft-quoted ‘real bond yield’ with extreme care. The real bond yield is nothing more than the nominal bond yield less a mathematical function of the oil price. So, when the oil price is high, it will give the illusion that the real bond yield is low. The danger is that if we value equities against the real bond yield when the oil price is high – such as through 2011-14 or now – equities will appear cheaper than they really are (Chart I-6). Chart I-6When The Oil Price Is High, 'The Real Bond Yield' Will Appear Lower Than It Really Is In The Case Against A ‘Super Bubble’ (And The Case For) we explained the much better way to value equities is versus the product of the nominal bond price and current profits. This valuation approach perfectly explains the US stock market’s evolution both over the long term (Chart I-7) and the short term. Specifically, over the past year, the dominant driver of the US stock market has been the 30-year T-bond price (Chart I-8). Chart I-7The US Stock Market = Profits Times The 30-Year T-Bond Price (Long-Term Chart) Chart I-8The US Stock Market = Profits Times The 30-Year T-Bond Price (Short-Term Chart) 12-Month Asset Allocation Conclusion The current inflation scare comes not from an aggregate demand shock, but from a massive displacement of demand (into goods) followed by the more recent supply shock for energy and food triggered by the Ukraine crisis. In response, central banks are trying to douse the inflation in the only way they can – by choking aggregate demand. Hence, there is a dangerous mismatch between the malady and the remedy. In the near term, the Ukraine crisis has added to already elevated fears about inflation – and this will pressure both bonds and stocks. However, looking beyond the next few months, the near-term inflationary impulse will unleash a disinflationary response from three sources. First, a supply shock means higher prices without stronger demand, which causes an inevitable demand destruction that then pulls down prices. Second, central banks are explicitly trying to pull down prices – or at least price inflation – by choking demand. And third, the massive displacement of demand into goods, and its associated inflationary impulse, is reversing. On a 12-month horizon, the disinflationary impulse will outweigh the inflationary impulse. Therefore, on a 12-month horizon, the disinflationary impulse will outweigh the inflationary impulse. The asset allocation conclusion is to overweight stocks and bonds, and to underweight TIPS and commodities. Is The Bond Sell-Off Close To Capitulation? Finally, several clients have asked if the recent sell-off in bonds is close to capitulation, based on the fragility of its fractal structures. The answer is yes, but only for the shorter maturity T-bonds. Specifically, the 5-year T-bond has reached the point of fragility on its composite 130-day/260-day fractal structure that marked the bottom of the sell-off in 2018, as well as the top of the rally in 2020 (Chart I-9). Chart I-9The Sell-Off In Shorter-Dated T-Bonds Is Close To Capitulation Accordingly, this week’s trade recommendation is to buy the 5-year T-bond, setting the profit target and symmetrical stop-loss at 4 percent, and with a maximum holding period of 1 year. Please note that our full fractal trading watchlist is now available on our website:  cpt.bcaresearch.com     Dhaval Joshi Chief Strategist dhaval@bcaresearch.com Footnotes 1 https://www.oxfordenergy.org/wpcms/wp-content/uploads/2021/08/Is-the-Oil-Price-Inflation-Relationship-Transitory.pdf Chart 1The Strong Trend In The 18-Month-Out US Interest Rate Future Is Fragile Chart 2The Strong Trend In The 3 Year T-Bond Is Fragile Chart 3AUD/KRW Is Vulnerable To Reversal Chart 4Canada Versus Japan Is Vulnerable To Reversal Chart 5Canada's TSX-60's Outperformance Might Be Over Chart 6US Healthcare Vs. Software Approaching A Reversal Chart 7The Euro’s Underperformance Could Be Approaching a Resistance Level Chart 8A Potential Switching Point From Tobacco Into Cannabis Chart 9Bitcoin's 65-Day Fractal Support Is Holding For Now Chart 10Biotech Approaching A Major Buy Chart 11CAD/SEK Reversal Has Started Chart 12Financials Versus Industrials Is Reversing Chart 13Norway's Outperformance Could End Chart 14Greece's Brief Outperformance Has Ended Chart 15BRL/NZD At A Resistance Point Fractal Trading System Fractal Trades 6-Month Recommendations Structural Recommendations Closed Fractal Trades Indicators To Watch - Bond Yields Chart II-1Indicators To Watch - Bond Yields - Euro Area Chart II-2Indicators To Watch - Bond Yields - Europe Ex Euro Area Chart II-3Indicators To Watch - Bond Yields - Asia Chart II-4Indicators To Watch - Bond Yields - Other Developed   Indicators To Watch - Interest Rate Expectations Chart II-5Indicators To Watch - Interest Rate Expectations Chart II-6Indicators To Watch - Interest Rate Expectations Chart II-7Indicators To Watch - Interest Rate Expectations Chart II-8Indicators To Watch - Interest Rate Expectations