Global

Special Report

Highlights We introduce our rotation graphs to assess the evolution of the relative trend and momentum of various assets. US equities remain on firm footing, but their weakening relative momentum suggests that investors may soon begin to rotate away from this market in favor of the Eurozone and EM. Cyclicals continue to dominate defensives, globally and in Europe. European value stocks are experiencing improving momentum, which suggests that a rotation out of growth equities is afoot. While European small-cap equities sport attractive fundamentals, rotational dynamics indicate it is still too early to overweight them aggressively. The energy crisis is a dominant driver of the relative sector performance in Europe and resulted in a massive shift in leadership from industrials to energy. As long as oil and natural gas act as a drag, industrials will lag. Financials are well supported. Swedish stocks have borne the brunt of the energy price spike, while Norwegian equities have been its main beneficiaries. The improvement in momentum of German stocks suggests that their relative underperformance will soon end. Spanish shares look attractive from a tactical perspective. Swiss industrials will need a recovery in EUR/CHF to outperform other European industrials. UK industrials will continue to outperform their continental competitors, while Spanish industrials have a window through which to shine. A rotation into UK financials may soon begin as their momentum improves. The darkest days for German financials are ending, while Spanish, Italian, and Swedish financials may soon witness a wave of underperformance. Spanish consumer discretionary equities are becoming more attractive compared to their European counterparts. While Dutch names continue to outperform other European tech equities, their softening momentum suggests investors are beginning to rotate out of this country. Spanish and German tech names offer an attractive diversification opportunity within the industry. Feature Methodology The combination of excess liquidity, large pools of fast money, elevated valuations across most securities, and the existence of the near-term momentum reversal effect encourage investors to rotate from one asset to the next in the hope of rapid profits. Measures to assess where each market stands in this rotational pattern can be useful for investors to catch these swings. In this optic, we introduce our rotation screener focused on equities. It is a simple tool that looks at whether a sector or a country is strengthening relative to its benchmark and whether this strength is happening at a faster or slower momentum. To measure each dimension, we use proprietary indicators of relative strength and momentum. Once each asset’s relative strength and relative momentum are established, we can position them in quadrants. We follow traditional terminology. The upper right quadrant denotes “Leading” assets, or securities that are outperforming their benchmark with strengthening momentum. The bottom right quadrant denotes “Weakening” assets, or securities that are outperforming their benchmark but with a deteriorating momentum. The bottom left quadrant denotes “Lagging” assets, or securities that are underperforming with decreasing momentum. Finally, the top left quadrant indicates “Improving” assets, or securities that are underperforming but with increasing momentum. Investors should move to overweight assets that are in the Improving quadrants and to underweight assets that are inching toward the “Lagging” from the “Weakening” quadrants. This method is very flexible and can be applied to sectors, countries, styles, and so on, as long as a benchmark is available to generate comparisons. In this report, we will analyze the following from a rotational perspective: global national markets, global cyclicals vs global defensive’s, European cyclicals vs European defensives, European sectors, European national markets, European financials, European consumer discretionaries, and European tech stocks. Global National Markets US equities have moved from the Leading quadrant to the Weakening one as they continue to outperform the global benchmark but with a decelerating momentum (Chart 1). This locates the US market in a risky position that could herald a period of underperformance, especially if global economic surprises accelerate. From a rotational perspective, US stocks could still experience another wave of outperformance over the coming weeks, as momentum has been firming over the past four weeks.   The Euro Area benchmark has fully moved from the Weakening quadrant in August to the Lagging one today. Investors should monitor Europe’s relative momentum closely, because a pick-up from here would push the Eurozone into Improving territory, a warning of an imminent trend change in European relative stock prices. Emerging markets have exited the Lagging zone and moved into the Improving quadrant. The move is far from decisive and remains at risk with Chinese credit growth still decelerating. The recent decline in steel prices in China suggests that construction activity in that economy continues to slow. Thus, as long as Chinese credit flows deteriorate, EM stocks will have trouble moving into the Leading quadrant. Cyclicals Vs Defensives Global defensive equities tried to move into the Leading quadrant at the end of the summer, but, ultimately, they plunged back into Lagging territory as global stocks recovered in October (Chart 2). Meanwhile, global cyclicals moved in the opposite trajectory, shifting from the Lagging quadrant to the Leading one over the past three months. Cyclicals continue to benefit from the general uptrend in the market. Even the recent decline in yields is doing little to boost the performance of defensive equities. The biggest risk to these stocks remains the Chinese economic slowdown. For now, this deterioration has not been large enough to compensate for the general vigour in profits and consumption in advanced economies. However, if inflation worries do not abate, then the Chinese slowdown will become more problematic for global cyclicals as it will raise the spectre of stagflation.  The rotational pattern for European cyclicals vs defensive stocks mimics that of global equities (Chart 3). However, European cyclicals are somewhat softer than their global equivalents, hurt by Europe’s greater exposure to the Chinese business cycle compared to the US’s exposure. European Investment Styles Over the past three months, European investment styles have begun a major shift. Value has moved from the Lagging quadrant to the Improving one, which suggests that flows could push value into the Leading quadrant (Chart 4). Moreover, growth has moved from the Leading quadrant to the weakening one, which created a similar dynamic as the decline in performance of the quality factor. This confirms that the rise in yields is beginning to favour a shift in style from growth to value. Meanwhile, small-cap stocks have tumbled into the Lagging quarter. We do expect attractive returns for European small-cap names over an 18- to 24-month investment horizon. However, we have not moved yet to overweight this sector of the market and rotational patterns confirm it is too early to do so safely. European Sectors Sectors have begun to make some important shifts in European markets (Chart 5). Tech has moved from the Leading quadrant to the weakening one. While the sector continues to outperform, it is doing so with a declining momentum, and it could soon move to the Lagging quadrant. This deteriorating price action must be monitored closely. Consumer discretionary names, which were strong performers that have become increasingly weak, have moved from the Weakening quadrant to the lagging one. However, their momentum is not deteriorating as much as it did nine weeks ago, which suggests a move to the Improving quadrant could soon be in the offing. Financials have greatly enjoyed the uptick in global yields. After a short passage through the Lagging quadrant, they have shifted into the Leading one. This suggests that the winds remain behind this sector, which we continue to overweight. Industrials and energy have become mirror images of one another, highlighting the negative impact on European economic activity and profitability of the recent surge in energy prices. The industrials have moved from the Leading quadrant to the lagging one, as the energy sector experienced the opposite direction of travel. This suggests that industrials will only recover their shine once the energy crisis abates, which will also hurt energy stocks. European National Markets The rotational pattern exhibited by European national markets bears their respective sectoral footprints (Chart 6). The tech-heavy Dutch market has moved from the Leading quadrant to the Weakening one, the industrials-focused Swedish market has fallen into the Lagging quadrant from the Weakening one and the Norwegian market has leapt out of Lagging into Leading territory. Hence, if the rotation out of tech deepens, The Netherlands will tumble directly into the Lagging zone, while an easing in energy prices will force Norway and Sweden to switch places on the back of a rotation out of energy into industrials. Germany is of particular interest. It is a well-diversified market that has become oversold. Moreover, as we wrote in September, its relative performance exhibits a significant discount to relative earnings. From a rotational perspective, Germany is moving to leave the Lagging quadrant; a durable shift into the Improving quadrant will sufficiently assuage traders into buying this market. This process will support our overweight position in German equities. Spain is another market we like on a tactical basis. Over the course of the past three months, it moved out of Lagging territory into the Improving zone. This price action supports our thesis that the large country-discount embedded across Spanish equity sectors is excessive and should soon dissipate. The main risk to this view would be another down leg in bond yields, which would hurt financials—a major weight in this market. Italy, too, is in the process of executing a full rotation, having exited the Weakening quadrant and moved into the Lagging one. Italian stocks have tried to punch their way into the Improving zone but have failed to do so. They will require higher yields to move out into the Improving zone durably because of the heavy financials weighting of Italian stocks. European Industrials Within European industrials, a rotational pattern is also evident (Chart 7). Swiss industrials have moved out of the Leading quadrant into the Lagging one as the Swiss franc continues to appreciate against the euro. The rising CHF imparts deflationary pressures into Switzerland and the SNB continues to build up its reserves. As a result, EUR/CHF will appreciate once EUR/USD finds a firmer footing. Thus, while it is too early to overweight Swiss industrials relative to those of the Eurozone, their oversold nature suggests that a rotation in favour of Swiss manufacturing businesses will soon take place. At the current juncture, Spanish industrials look appealing. They have moved out of the Lagging quadrant into the Improving one as the momentum of their relative performance improves. Additionally, they are close to moving into the leading territory. This picture is consistent with a narrowing of the discount embedded in all Spanish sectors since the pandemic broke out. Swedish industrials are also trying to exit the Lagging territory; their elevated RoE, and heavy sensitivity to the DM capex cycle indicate that they should move into the Leading quadrant in the coming weeks. UK industrials have remained in the Leading zone for the past three months, but their relative momentum is softening, which risks them being placed in the Weakening zone. The recent deterioration in GBP/EUR could provide a breath of fresh air, as it will improve the competitiveness of UK industrials compared to continental firms. Even then, for now, rotational dynamics do not flag an imminent problem for UK industrials. European Financials The clearest rotational pattern within European financials may be found in Sweden and the UK (Chart 8). Over the past three months, Swedish financials have fallen out of the Leading quadrant into the Weakening one, and they are inching closer toward the Lagging zone. This suggests that they could soon begin to underperform. Meanwhile, UK financials offer a mirror image as they exited the Lagging quadrant and moved into the improving one. They have yet to enter Leading territory, but seem close to doing so. The pessimism toward the UK is overdone right now. BCA’s Global Fixed-Income Strategy team expects the UK yield curve to steepen anew. UK financials would be prime beneficiaries of this dynamic. Italian and Spanish financials are also exhibiting some concerning moves lately. Both were in the Leading quadrant, but they have since shifted to the lagging one as peripheral spreads widened. Meanwhile, money seems to be moving into German financials, which have advanced from the Lagging quadrant to the Improving quadrant. While they are not as close to the Leading quadrant as their UK competitors, this shift warrants monitoring. European Consumer Discretionary Within the consumer discretionary space, most European countries have remained in their quadrant (Chart 9). Nonetheless, Spanish CD stocks have moved out of the Lagging zone into the Leading quadrant, while their Italian counterparts have recently entered the Weakening quadrant where they have joined French CDs. While both these countries’ consumer discretionary firms are witnessing weakening momentum, they remain in an upward trend against their European competitors. It is therefore too early to sell these countries within this industry. German Consumer discretionary equities are still in the Lagging quadrant, but they are trying to move into the Improving one, where UK CD names have remained for the past three months. European Tech The European tech sector is very much a story about The Netherlands versus the rest, due to the large size of the Dutch tech sector (Chart 10). For now, rotational patterns remain in favour of Dutch names; they have exited the Leading quadrant, but, while their momentum is weakening somewhat, they remain in a pronounced relative uptrend. A few small markets offer some promise. Over the past three months, both Spanish and German tech names have shifted from the Lagging quadrant into the Improving one. Their elevated momentum measures suggest that a shift into the Leading quadrant is imminent. As such, investors should consider switching some of their tech holdings into these two countries to diversify away from the Dutch behemoth.   Mathieu Savary, Chief European Strategist Mathieu@bcaresearch.com

Bank Credit Analyst

In this report we examine the risk of stagflation by comparing the current environment to that of the late-1960s and 1970s. Today, investors cannot rule out the possibility of a stagflationary outcome, for four reasons: long-term household inflation expectations have risen significantly over the past year; fiscal policy has been expansionary; monetary policy will remain expansionary at the Fed’s projected terminal Fed funds rate; and component shortages and price increases linked to energy market and supply chain disruptions may persist or worsen over the coming year. However, the strong demand-pull inflationary dynamics that existed in the late-1960s were mostly absent in the lead-up to the pandemic, supply-chain issues are in part due to strong goods demand and supply disruptions that will eventually dissipate, and economic agents do not expect severe price pressures to persist beyond the pandemic. On balance, this points to a stagflationary outcome over the coming 6-24 months as a risk, but not a likely event. Investors should use the Misery Index, which is the sum of the unemployment rate and headline PCE inflation, as a real-time stagflation indicator. The Misery Index underscores that the US economy is unlikely to experience true stagflation unless the unemployment rate rises. A portfolio of the US dollar, the Swiss Franc, and industrial commodities may serve as a useful hedge for investors who are concerned about absolute return prospects in a world in which long-maturity bond yields are rising and risks of stagflationary dynamics are present. Chart II-1The Misery Index Reflects The Risk Of Stagflation Over the past several weeks, concerns about a possible return to 1970s-style stagflation have re-emerged significantly in the minds of many investors. These investors have pointed toward similarities between the current environment and that of the 1970s, including shortages limiting output, a snarled global trade and logistical system, and rising energy prices. Chart II-1 highlights that the US “Misery Index” – the sum of the unemployment rate and headline PCE inflation – rose again over the past several months to high single-digit territory, after having fallen dramatically from April 2020 to February of this year. Panel 2 of Chart II-1 highlights that last year's rise in the Misery Index was driven almost entirely by the unemployment rate, whereas the current level is due to a combination of a modestly elevated unemployment rate and a pronounced acceleration in inflation. The headline PCE deflator has risen above 4%, a level that has not been reached since 1991 during the First Gulf War. In this report, we examine the risk of stagflation by comparing the current environment to that of the late 1960s and 1970s. We conclude that while investors cannot rule out the possibility of a stagflationary outcome, there are important differences that point toward a stagflation outcome over the coming 6-24 months as a risk, not a likely event. We conclude by highlighting assets that may produce absolute returns in a world in which long-maturity bond yields are rising and risks of stagflationary dynamics are present. Revisiting The 1960s And 70s Chart II-2The 1960s Laid The Groundwork For Elevated Inflation The first step in judging the risk of a return to 1970s-style stagflation is to review, in a detailed way, what caused those conditions. Investors are well aware of the role that two separate energy price shocks played in raising prices and damaging output during this period, but they are less cognizant of the impact that a persistent period of above-trend output and significant labor market tightness had in setting up the conditions for sharply higher inflation. This focus of investors on energy prices partially reflects the fact that the Misery Index increased most visibly in the 1970s and that policymakers in the 1960s may not have realized how extensively economic output was running above its potential. With the benefit of hindsight, Chart II-2 illustrates the extent to which inflationary pressures built up in the 1960s, well before the first oil price shock in 1973. The chart shows that the unemployment rate was below NAIRU – the non-accelerating inflation rate of unemployment – for 70% of the time during the 1960s, and that inflation had already responded to this in the latter half of the decade. Annual headline PCE inflation was running just shy of 5% at the onset of the 1970 recession; it fell to 3% in the aftermath of the recession, but had already begun to reaccelerate in the first half of 1973. Following the 1973/1974 recession, inflation did decelerate significantly, falling from 11-12% to 5% in headline terms, and from 10% to 6% in core terms. But the pace of price appreciation did not fall below 5-6% in the second half of the 1970s, despite a significant and sustained rise in the unemployment rate above its natural rate. The 1975 to 1978 period is especially important for investors to understand, because it is arguably the clearest period of true stagflation in the 1970s. The fact that the Misery Index rose sharply during two major oil price shocks is not particularly surprising in and of itself, given the direct impact of energy prices on headline consumer prices; it is the fact that the index remained so elevated between these shocks, the result of persistently high inflation in the face of significant labor market slack, that is most relevant to investors. There are two reasons that both inflation and unemployment remained high during this period. First, labor market slack was sizeable during these years because the US economy was more energy-intensive in the 1970s than it is today. Chart II-3 highlights that goods-producing employment lagged overall employment growth from late 1973 to late 1977, underscoring that the rise in oil prices significantly impacted jobs growth in energy-intensive industries. Second, it is clear that the combination of demand-pull inflation in the late 1960s and the predominantly cost-push inflation of the 1970s led to expectations of persistent inflation among households and firms. The original Phillips Curve, as formulated by New Zealand economist William Phillips in the late 1950s, described a negative relationship between the unemployment rate and the pace of wage growth. Given the close correlation between wage and overall price growth at the time, the Phillips Curve was soon extended and generalized to describe an inverse relationship between labor market slack and overall price inflation. But the experience of the 1970s highlighted that inflation expectations are also an important determinant of inflation, a realization that gave birth to the expectations-augmented (i.e. “modern-day”) Phillips Curve (more on this below). The Stagflation Era Versus Today Table II-1 presents a stagflation “threat matrix,” representing the Bank Credit Analyst service’s assessment of the various factors that could potentially contribute to a stagflationary environment today, relative to what occurred in the 1960s and 1970s. While we acknowledge that there are some similarities today to what occurred five decades ago, the most threatening factors have been present for a shorter period of time and appear to have a smaller magnitude than what occurred during the stagflationary era. In addition, key factors, such as the visibility available to policymakers and investors about household inflation expectations and the potential output of the economy, would appear to reduce significantly the risk of a stagflationary outcome today. We discuss each of the factors presented in Table II-1 below: Fiscal & Monetary Policy Chart II-4Government Spending Last Cycle Looked Nothing Like The 1960s The persistently tight labor market that contributed to the inflationary buildup in the 1960s occurred as a result of easy fiscal and monetary policy. Chart II-4 highlights that the contribution to real GDP growth from government expenditure and investment was very elevated in the 1960s. Chart II-5 shows that a positive output gap in the late 1960s and the first half of the 1970s is well explained by the fact that 10-year US government bond yields were persistently below nominal GDP growth. The relationship between the stance of monetary policy and the output gap only meaningfully diverged in the latter half of the 1970s, during the true stagflationary era that we noted above. Chart II-5Easy Monetary Policy Juiced Aggregate Demand In The 60s And Early 70s Chart II-6Monetary Policy Today Is Extremely Easy Today, it is clear that the stance of fiscal policy has recently been extraordinarily easy, and 10-year US government bond yields have remained well below nominal GDP growth for the better part of the last decade. Relative to estimates of potential nominal GDP growth, 10-year Treasury yields are the lowest they have been since the 1970s (Chart II-6). Ostensibly, this supports concerns that policy might contribute to a stagflationary outcome. These concerns were raised by Larry Summers in March, when he described the Biden administration’s fiscal policy as the “least responsible” that the US has experienced in four decades and warned of the potential inflationary consequences of overheating the economy.1 But there are two important counterpoints to these concerns. First, easy fiscal policy this cycle has followed a period during the last economic cycle in which government spending contributed to the most sustained drag on economic activity since the 1950s. Unlike the 1960s, the unemployment rate has been below NAIRU for only a third of the time over the past decade. In addition, Chart II-7 highlights that fiscal thrust will turn to fiscal drag next year, underscoring the temporary nature of the massive burst in fiscal spending that has occurred in response to the pandemic. Under normal circumstances, the fiscal drag implied by Chart II-7 would substantially raise the risks of a recession next year, but we have noted in previous reports that a significant amount of excess savings remain to support spending and employment. The net impact of these two factors results in a reasonable expectation that the US economy will return to maximum employment next year, but this is a far cry from the 1960s when the unemployment rate was below its natural rate for 70% of the decade. Based on conventional measures, US monetary policy has been easy for a decade, but easy monetary policy did not begin to contribute positively to a rise in household sector credit growth last cycle until 2014/2015. This underscores that the natural rate of interest (“R-star”) did fall during the early phase of the last economic expansion. However, we argued in an April report that R-star was likely rising in the latter half of the last expansion,2 and we believe that the terminal Fed funds rate is likely higher than what the Fed is currently projecting, barring any additional negative policy shocks. Thus, while we do not believe that the duration of easy monetary policy over the past decade has laid the groundwork for a major rise in prices, it is now clearly positively contributing to aggregate demand and does risk a future overshoot in prices if long maturity bond yields remain well below the pace of economic growth for a sustained period of time. The Impact Of Shortages Chart II-8Gasoline Shortages Plagued The US Economy In The 1970s Gasoline shortages occurred during the oil shocks of the 1970s and are a key similarity that some investors point toward when comparing the situation today with the stagflationary era. Chart II-8 highlights that the annual growth in real personal consumption expenditures on energy goods and services fell into negative territory on six occasions in the 1970s, although it was most pronounced during the two oil price shocks and their resulting recessions. Today, the impact of shortages appears to be broader than what occurred in the 1970s, but less impactful and not likely to be as long-lasting. Chart II-9 highlights that the OPEC oil embargo of 1973 raised the global oil bill by 2.4% of global GDP and permanently raised the price of oil. The global oil bill will only be fractionally above its pre-pandemic level in 2022, with oil prices at $80/bbl, and, while it is true that US gasoline prices have risen significantly, they are not higher than they were from 2011-2014 (Chart II-10). Chart II-9$80/bbl Oil Is Not Onerous Chart II-10US Gasoline Prices Are High, But They Have Been Higher It is certainly true that global shipping costs have skyrocketed and that this is contributing to the increase in US consumer prices. We estimate, however, that this increase in shipping costs as a share of GDP is no more than a quarter of the impact of the 1973 increase in oil prices, without the attendant negative effects on US goods-producing employment that occurred in the 1970s. If anything, surging shipping costs create an incentive to re-shore manufacturing production, which would contribute positively to US goods-producing employment. We also do not expect the rise in shipping costs to be meaningfully permanent, i.e., shipping costs may ultimately settle at a higher level than they were in late-2019, but at a much lower level than what prevails today. Chart II-11A Tight Labor Market Is Causing Wage Growth To Pick Up Semiconductor and labor shortages would appear to represent a more salient threat of stagflation in the US, as the domestic production of motor vehicles cannot occur without key inputs and a tight labor market is already contributing to an acceleration in wage growth (Chart II-11). As we noted in Section 1 of our report, auto production significantly impacted growth in the third quarter. However, Chart II-12 highlights that, for now, the breadth of impact of these shortages appears to be limited: the production component of the ISM manufacturing index remains in expansionary territory, industrial production of durable manufacturing excluding motor vehicles and parts has not broken down, and both housing starts and building permits remain above pre-pandemic levels despite this year’s downtrend in permits. Chart II-12Shortages Do Not Yet Seem To Be Broad-Based A physical shortage of components is a less relevant factor for the services side of the economy, which appears to have re-accelerated meaningfully in October. The services sector is more considerably impacted by shortages in the labor market, which seem to be linked to a still-low labor force participation rate. We noted in our September report that the decline in the participation rate has significantly overshot what would be implied by the ongoing pace of retirements. Chart II-13 highlights that this has occurred not just because of a significant retirement effect, but also because of the shadow labor force (people who want a job but are not currently looking for work) and family responsibilities. We expect that the recent expiry of expanded unemployment insurance benefits, a steady rise in the immunity of the US population, an abating Delta wave of COVID-19, and a likely upcoming reduction in school/classroom closures once the Pfizer/BioNTech vaccine is approved for school-age children will likely ease the labor shortage issue over the coming several months. Output Gap Uncertainty It remains a debate among economists why policymakers maintained such easy monetary policy in the 1960s and 1970s, but Chart II-14 highlights that uncertainty about the size of the output gap may have contributed to too-low interest rates. The chart shows the unemployment rate compared with today's estimate of NAIRU, alongside a simple proxy for policymakers’ real time estimate of the natural rate of employment: the cumulative average unemployment rate in the post-war environment. To the extent that policymakers used past averages of the unemployment rate as their guide for NAIRU, Chart II-14 highlights how they may have underestimated the degree to which output was running above its potential level in the 1960s, and would not have even concluded that output was above potential in the early 1970s. Chart II-14Policymakers Overestimated Labor Market Slack In The 60s And 70s Chart II-15Policymakers Know That NAIRU Is Likely At Or Below 4% Today, the environment is quite different, because the acceleration in wage growth at the tail end of the last expansion gives policymakers and investors a good estimate of where NAIRU is. Chart II-15 highlights that wage growth accelerated in 2018/2019 in response to a sub-4% unemployment rate, which is consistent with both the Fed’s NAIRU estimate of 3.5-4.5% and Fed Vice Chair Richard Clarida’s expressed view that a 3.8% unemployment rate likely constitutes maximum employment (barring any issues with the breadth and inclusivity of the labor market recovery). It is possible that the pandemic has structurally lowered potential output, which could mean that policymakers may no longer rely on the wage growth / unemployment relationship that existed in the latter phase of the last expansion. However, we do not find any credible arguments that would support the notion of a structurally lower level of potential output: the pandemic is likely to end at some point in the not-too-distant future, the negative impact of working-from-home policies on office properties and employment in central business districts is not sizeable,3 and productivity may have permanently increased in some industries because of the likely stickiness of a hybrid work culture. The Behavior Of Inflation Expectations Chart II-16Rising Long-Term Expectations Have Merely Normalized (For Now) One parallel to the argument that policymakers may have underestimated the degree of labor market tightness in the 1960s and early 1970s is the fact that they did not yet understand that inflation expectations are an important determinant of actual inflation, nor were they able to monitor them even if they did. Most credible surveys of inflation expectations began in the 1980s, and policymakers in the 1960s and 1970s were guided by the original Phillips Curve that solely related inflation to unemployment. Today, policymakers have the experience of the stagflationary episode to serve as a warning not to allow inflation expectations to get out of control, and both policymakers and investors have reliable measures of inflation expectations for households and market-participants. Chart II-16 highlights that households expect significant inflation over the coming year, but also expect prices over the longer term to rise at a pace that is almost exactly in line with their average from 2000-2014. The Rudd Controversy: (Adaptive) Inflation Expectations Do Matter One potential criticism of the idea that inflation expectations are signaling a low risk of higher future inflation has emerged through arguments made by Jeremy Rudd, a Federal Reserve economist. In a recent paper, Rudd questioned the view that households’ and firms’ expectations of future inflation are a key determinant of actual inflation; he suggested instead that relatively stable inflation since the mid-1990s might reflect a situation in which inflation simply does not enter workers’ employment decisions and expectations are irrelevant. Rudd’s paper was primarily addressed to policymakers who view inflation dynamics in a highly quantitative light. A full response to the paper would be mostly academic and thus not especially relevant to investors; however, we would like to highlight three points related to the Rudd piece that we feel are important.4 First, we disagree with Rudd’s argument that the trend in inflation has not responded to changes in economic conditions since the mid-1990s. Chart II-17 highlights that while the magnitude of the relationship has shifted, the trend in inflation relative to a measure of long-term expectations based on prior actual inflation has mimicked that of the output gap. The fact that inflation was (ironically) too high during the early phase of the last economic cycle provides some support for Rudd’s inflation responsiveness view, although we would still point toward the Fed’s strong record of maintaining low and stable inflation, its active communication with the public in the years following the global financial crisis, and the fact that a recovery began and the output gap began to (slowly) close as the best explanation for the avoidance of deflation during that period. Second, we agree with Rudd’s point that regime shifts in inflation’s responsiveness to economic conditions can occur, and that adaptive measures of inflation expectations, and even surveys of inflation, may not capture such a shift in real time. Chart II-18 shows that the 2014-2016 period was a good example of this, when adaptive expectations as well as household survey measures of long-term inflation expectations both lagged the actual decline in inflation that was caused by a collapse in the price of oil. Chart II-17The Trend In Inflation Continues To Respond To Economic Conditions Chart II-18Surveyed Inflation Expectations Can Lag, But This Time They Led But Chart II-18 also shows that long-term household survey measures of inflation led the rise in actual inflation (and thus our adaptive expectations measure) last year, underscoring that these measures are likely more reliable indicators today of whether a major regime shift is occurring. As noted above, long-term expectations have risen significantly relative to what prevailed prior to the pandemic, but this has merely raised expectations from extraordinarily depressed levels back to the average that prevailed prior to (and immediately after) the global financial crisis. Therefore, household expectations are not yet at dangerous levels. Chart II-19Unit Labor Costs Modestly Lead Inflation, But Are Far From Extreme Third, one of the core observations in Rudd’s paper is that unit labor cost (ULC) growth leads the trend in inflation, which he argued was evidence against the idea that expectations of future inflation are a key determinant of actual inflation. Chart II-19 highlights that Rudd is correct that ULC growth modestly leads inflation (especially core inflation), but we disagree with his conclusion that it argues against the importance of expectations. As we noted in Section 2 of our January 2021 Bank Credit Analyst,5 one crucial aspect of the expectations-augmented, or “modern-day” Phillips Curve is that, if inflation expectations are largely formed based on the experience of past inflation, then inflation is ultimately determined by three dimensions of the output gap: whether it is rising or falling, whether it is above or below zero, and how long it has been above or below zero. Our view is that ULC growth is fundamentally linked to slack in the labor market, which is directly incorporated in output gap measures. As we noted above, investors currently have a good estimate of the magnitude of the output/employment gap, meaning that it is possible to track the inflationary consequences of prevailing aggregate demand. As a final point about ULC growth, Chart II-19 highlights that while the five-year CAGR of unit labor costs is currently running at its strongest pace since the global financial crisis, investors should note that it remains well below the levels that prevailed in the late-1960s when persistently above-potential output laid the groundwork for a massive inflationary overshoot. Conclusions And Investment Strategy Our review of the 1960s and 1970s highlights that stagflation is a phenomenon in which supply-side shocks raise prices of key inputs to production, which lowers output and raises unemployment. Energy price shocks in the 1970s occurred after a long period of policy-driven above-trend growth in the 1960s, meaning that both demand-pull and cost-push inflation contributed to stagflation in the 1970s. Today, investors cannot rule out the possibility of a stagflationary outcome, for four reasons: long-term household inflation expectations have risen significantly over the past year; fiscal policy has been very expansionary; monetary policy will remain expansionary at the Fed’s projected terminal Fed funds rate; and component shortages and price increases linked to energy market and supply chain disruptions may persist or worsen over the coming year. Chart II-20It Is Not Stagflation If The Unemployment Rate Continues To Fall However, the strong demand-pull inflationary dynamics that existed in the late-1960s were mostly absent in the lead-up to the pandemic, supply-chain issues are in part the result of strong goods demand and disruptions that are clearly linked to the pandemic (and thus will eventually dissipate), and long-term inflation expectations are behaving differently than short-term expectations, signaling that economic agents do not expect severe price pressures to persist beyond the pandemic. Policymakers also have more visibility about the magnitude of economic / labor market slack than they did during the stagflationary era and better tools to track inflation expectations. On balance, this points to a stagflationary outcome over the coming 6-24 months as a risk, but not as a likely event. Using the Misery Index as real-time stagflation indicator, investors should note that the US economy is not likely experiencing true stagflation unless the unemployment rate rises. Chart II-20 highlights that there is no evidence yet of a contraction in goods-producing or service-producing jobs. Even if goods-producing employment slows meaningfully over the coming few months as a result of component shortages, the unemployment rate is still likely to fall if services spending normalizes, as it would imply that the gap in services-producing employment, which is currently 20% of the level of pre-pandemic goods-producing employment, will continue to close. Investors have been focused on the issue of stagflation because its occurrence would imply a sharply negative correlation between stock prices and bond yields. This is not our base case view, but we have highlighted that months with negative returns from both stocks and long-maturity bonds tend to be associated with periods of monetary policy tightening (or in anticipation of such periods). As we discussed in Section 1 of our report, we do expect the Fed to raise interest rates next year. We do not see a rise in bond yields to levels implied by the Fed’s interest rates projections as being seriously threatening to economic activity, corporate earnings growth, or equity multiples. But the adjustment to higher long-maturity bond yields may unnerve equity investors for a time, implying temporary periods of a negative stock price / bond yield correlation. Table II-2 highlights that, since 1980, commodities, the US dollar, and the Swiss franc have typically earned positive returns during non-recessionary months in which stock and long-maturity bond returns are negative. While the dollar is not likely to perform well in a stagflationary scenario, Chart II-21 highlights that CHF-USD and industrial commodities performed quite well in the late-1970s. As such, a portfolio of these three assets might serve as a useful hedge for investors who are concerned about absolute return prospects in a world in which long-maturity bond yields are rising and risks of stagflationary dynamics are present. Chart II-21The Swiss Franc and Raw Industrials Did Well During The Stagflationary Era Jonathan LaBerge, CFA Vice President The Bank Credit Analyst Footnotes 1  “Summers Sees ‘Least Responsible’ Fiscal Policy in 40 Years,” Bloomberg News, March 20, 2021. 2 Please see The Bank Credit Analyst “R-star, And The Structural Risk To Stocks,” dated March 31, 2021, available at bca.bcaresearch.com 3 Please see The Bank Credit Analyst “Work From Home “Stickiness” And The Outlook For Monetary Policy,” dated June 24, 2021, available at bca.bcaresearch.com 4 Rudd, Jeremy B. (2021). “Why Do We Think That Inflation Expectations Matter for Inflation? (And Should We?),” Finance and Economics Discussion Series 2021-062. Washington: Board of Governors of the Federal Reserve System. 5  Please see The Bank Credit Analyst “The Modern-Day Phillips Curve, Future Inflation, And What To Do About It,” dated December 18, 2021, available at bca.bcaresearch.com

Bank Credit Analyst

Highlights The circumstances of the pandemic improved in October, but data highlighting the economic consequences of the Delta wave grew more severe. US economic activity slowed meaningfully in the third quarter, driven by lower car sales and a slowdown in services spending. The imminent vaccination of school-aged children, and signs that services activity and spending are increasing, will likely raise labor force participation, boost education employment, and hasten the return of real services spending back to pre-pandemic levels. Investors have the right bond view, but the wrong reason. Investors believe that the Fed will be forced to raise interest rates earlier than it currently expects to prevent an out-of-control rise in prices, whereas it will likely do so because of a quicker return to maximum employment. Bond yields are likely to move higher over the coming year, but this will be driven by real yields, not inflation expectations. Once the Fed begins to raise interest rates, investors should be on the lookout for signs that market expectations for the real natural rate of interest, or “R-star,” are rising. The Fed’s terminal rate projection is well below nominal potential GDP growth, and a gap between these two measures no longer makes sense. Stocks are likely to generate mid-single digit returns next year, which will beat the returns offered by bonds and cash. But stocks will generate much lower returns compared with those enjoyed by investors over the past year. A benign rise in long-maturity bond yields argues for the outperformance of value versus growth stocks over the coming year. Cyclical stocks are now becoming stretched versus defensives on an equally-weighted basis; stay overweight for now, but a downgrade to neutral may be in the cards at some point next year. Feature Chart I-1The Waning Impact Of Delta Over the past month, the focus of investors has shifted from day-to-day developments to the consequences of the Delta wave of the pandemic. Chart I-1 highlights that, while an estimate of the COVID-19 reproduction rates in advanced economies has recently inched higher, it remains below one and hospitalizations continue to trend lower in most major economies. UK hospitalizations have increased over the course of the month, but remain at a level that is a quarter of their January peak – despite an elevated pace of confirmed cases. In the US, both these cases and hospitalizations continue to fall, trends that are likely to be reinforced by the vaccination of children over the coming weeks. A 50-60% vaccination rate for school-aged children would increase the US vaccination rate by 4-5 percentage points. Vaccinating all children at this rate would increase the total vaccination rate by 7-8 percentage points. In combination with a meaningful level of natural immunity, the vaccination of children is likely to bring the US very close to, if not above, the non-accelerating hospitalization rate of immunity (or “NAHRI”).1 The Delta Hangover While the circumstances of the pandemic improved in October, the economic consequences grew more severe. US economic activity slowed meaningfully in the third quarter, as highlighted by yesterday’s advance release. Chart I-2 highlights that durable goods spending subtracted almost three percentage points from Q3 growth, and that most other components of GDP contributed less to growth in Q3 than in Q2. The significant slowdown in Q3 growth is disappointing, but several factors point toward the conclusion that it is not likely to be sustained: Chart I-3Services PMIs Are Pointing To A Stronger Q4 The Delta wave very likely impacted services spending, which we have highlighted is likely to drive overall consumption over the coming year. Given the ongoing impact of semiconductor shortages on the availability of new cars, it is not surprising that a slowdown in services spending resulted in a significant slowdown in overall growth. After having declined significantly in Q3, Chart I-3 highlights that the US, UK, French, and Japanese October flash services PMI rose anew, underscoring that recent services weakness have been closely linked to the Delta variant of COVID-19 (whose impact is now waning). Chart I-3 also highlights that the US services PMI is currently at a level that has been historically consistent with solid real PCE growth. Finally, while it is true that manufacturing PMIs are being supported by supplier deliveries components, the October output component of the US Markit manufacturing index remained in expansionary territory, as was the case in Germany, Japan, and the UK (despite month-over-month declines in these components). Chart I-4 highlights that Q3’s real GDP reading was highly anomalous relative to the pace of jobs growth in the quarter, based on the relationship between the two since the global financial crisis. In quarters in which real GDP growth was 1% or less than implied by the trendline shown in Chart I-4, real GDP accelerated in the subsequent quarter 80% of the time. In conjunction with a pickup in services activity in October, this suggests that growth will be meaningfully stronger in Q4. Chart I-5Global Growth Is Peaking, But A Major Downturn Is Unlikely Chart I-5 shows our global Nowcast indicator, alongside our global LEI. Our Nowcast indicator is a high-frequency measure of economic activity that is designed to predict global industrial production. The chart shows that both the Nowcast and global LEI are declining, but that this decline is occurring from an extremely elevated level. The global economy is at an inflection point in terms of the pace of growth, but Chart I-5 still points to above-trend growth – and certainly not a major cyclical downturn. The expectation of a slowdown in growth in Q3 has significantly raised concerns about a possible return to 1970s-style stagflation in the minds of many investors. We address this topic in depth in this month’s Special Report, and conclude that, while investors cannot rule out the possibility of stagflation, there are important differences that point toward a stagflationary outcome over the coming 6-24 months as a risk, not a likely event. We note in our report that the risk of stagflation can be monitored in real time by tracking the Misery Index, which is the sum of headline PCE inflation and the unemployment rate. Currently, the Misery Index is elevated relative to the average of the past 30 years, but it is meaningfully lower than it was during the latter half of the 1970s. This also underscores that true stagflation is only likely to occur if the unemployment rate rises, which means that the economic and financial market outlook over the coming year is strongly tied to the pace of jobs growth (even more so than usual). Table I-1 presents an industry breakdown of the jobs gap relative to pre-pandemic levels, sorted by industries with the largest gap. The table highlights that leisure and hospitality, government, and education and health services jobs continue to account for two-thirds of the five million jobs gap, with the latter two largely reflecting the same effect: 60% of the government jobs gap is accounted for by state and local government education-related employment. Chart I-6Leisure And Hospitality Employment Tracks The Hotel Occupancy Rate US education employment has been impacted by school and classroom closures, which we noted above are likely to end once school-aged children are vaccinated against the disease. Chart I-6 highlights that leisure and hospitality employment is clearly predicted by the US hotel occupancy rate, which wobbled over the past few months as a result of the Delta wave of the pandemic. Correspondingly, monthly growth in leisure and hospitality employment slowed in August and September. Taken together, the imminent vaccination of school-aged children and signs that services activity and spending are increasing will likely raise labor force participation, boost education employment, and hasten the return of real services spending back to pre-pandemic levels. The Bond Market Outlook Chart I-7The Market Now Agrees With Us About The Timing Of Fed Rate Hikes... A continued normalization of the labor market over the coming 6-12 months argues in favor of Fed rate hikes next year, which is a view that we have maintained for several months. Recently, investors have come to agree with us, by moving forward their expectations for the Fed funds rate (Chart I-7). However, Chart I-8 highlights that investors have the right view for the wrong reason. The chart highlights that US government bond yields have risen entirely due to inflation expectations and that real yields have fallen. This means that investors believe that the Fed will be forced to raise interest rates earlier than it currently expects to prevent an out-of-control rise in prices, whereas we believe that they will do so because of a return to maximum employment. The implication for investors is that bond yields are still likely to rise over the coming year, but that higher yields are likely to occur alongside falling inflation expectations. This trend underscores that common hedges against inflation, such as precious metals and the relative performance of TIPS, are likely to underperform over the coming year. We have noted in previous reports that the fair value for long-maturity government bond yields implied by the Fed’s interest rate projections is not likely threatening for equity multiples, and certainly not for economic activity. A September 2022 rate hike, coupled with a pace of three hikes per year and a 2.5% terminal Fed funds rate, implies that 10-year Treasury yields will rise to 2.15% over the coming year, which would be only modestly higher than the level that prevailed prior to the pandemic (Chart I-9). Chart I-8...But For The Wrong Reason Chart I-9Higher Bond Yields Are Unlikely To Be Restrictive Next Year   However, once the Fed begins to raise interest rates, investors should be on the lookout for signs that market expectations for the real natural rate of interest, or “R-star,” are rising. The Fed’s terminal rate projection is well below nominal potential GDP growth, and, while a gap between these two measures made sense in the years following the global financial crisis, this no longer appears to be the case. Chart I-10 highlights that real household mortgage liabilities began to contract sharply in 2006, and did not turn positive on a year-over-year basis until the end of 2016. It is likely that R-star was falling or weak during this period, but the correlation between the two series clearly shifted in the latter phase of the last economic cycle. Chart I-11 emphasizes this point by highlighting that the household debt service ratio is now the lowest it has been since the 1970s, underscoring the capacity that US consumers have to withstand higher interest rates. Chart I-10R-star Fell Post-GFC, For A Time Chart I-11Today, US Households Have A Lot Of Capacity To Tolerate Higher Rates     We doubt that investor expectations for the terminal rate will rise significantly before the Fed begins to normalize monetary policy, but it may happen. In addition, the Fed may begin raising interest rates next year as soon as late in the summer or early in the fall, which would locate the liftoff date within our 6-12 month investment time horizon. As such, our base case view is that a rise in interest rates over the coming year will not be threatening to the equity market, but this view may change at some point next year. Equities: Expect Modest Returns In 2022 A benign increase in long-maturity bond yields in 2022 suggests that equity multiples will neither contribute to, nor subtract from, equity returns. As such, return expectations for equities should be centered around expected earnings growth. Table I-2 presents consensus estimates for nominal GDP growth, S&P 500 revenue growth, and earnings growth for 2022. The table highlights that expectations for revenue growth estimates appear to be reasonable, given that bottom-up analysts continue to expect an expansion in profit margins next year. Chart I-12 highlights that margins have already risen back above their pre-pandemic high, and that this is true for both tech and ex-tech sectors. Chart I-12US Profit Margins Have Already Risen To Record Levels We doubt that further increases in profit margins will be sustained next year. It is possible that margins will actually decline – a view that was recently espoused by our US Equity Strategy service.2 Risks to profit margins underscore that stocks are likely to generate mid-single digit returns next year, which will beat the returns offered by bonds and cash. But stocks will generate much lower returns compared with those enjoyed by investors over the past year. Within the equity market, we remain of the view that even a benign rise in long-maturity bond yields argues for the outperformance of value versus growth stocks over the coming year. Chart I-13 highlights that the rolling one-year correlation between relative global growth versus value stock prices and the US 10-year Treasury yield has become increasingly negative over time, which bodes well for value. We also continue to recommend that investors favor small over large caps and cyclicals over defensives, although cyclical stocks are now becoming stretched versus defensives on an equally-weighted basis as they are closing in on their 2018 highs (Chart I-14). We think it is too early to position against cyclicals, but a downgrade to neutral may be in the cards at some point next year. Chart I-13Growth Will Underperform Value If Long-Maturity Bond Yields Rise Chart I-14Cyclicals Are Starting To Look Stretched Versus Defensives   Investment Conclusions Next month’s report will feature BCA’s 2022 outlook, as well as a transcript of our recently held annual discussion with Mr. X and his daughter Ms. X (who joined his family office a couple of years ago). Our annual outlook will provide a detailed walkthrough of our views for the upcoming year, as well as answers to sobering questions raised by Mr. X and Ms. X about the longer-term outlook. For now, we recommend that investors stick with a pro-cyclical view, favoring the following assets: Global stocks over bonds A short-duration position within a government bond portfolio Speculative-grade corporate bonds within a credit portfolio Global ex-US stocks vs US, focused on DM ex-US Global value versus growth stocks Cyclicals versus defensives, and small versus large caps Major currencies versus the US dollar Jonathan LaBerge, CFA Vice President The Bank Credit Analyst October 29, 2021 Next Report: November 30, 2021 II. Gauging The Risk Of Stagflation In this report we examine the risk of stagflation by comparing the current environment to that of the late-1960s and 1970s. Today, investors cannot rule out the possibility of a stagflationary outcome, for four reasons: long-term household inflation expectations have risen significantly over the past year; fiscal policy has been expansionary; monetary policy will remain expansionary at the Fed’s projected terminal Fed funds rate; and component shortages and price increases linked to energy market and supply chain disruptions may persist or worsen over the coming year. However, the strong demand-pull inflationary dynamics that existed in the late-1960s were mostly absent in the lead-up to the pandemic, supply-chain issues are in part due to strong goods demand and supply disruptions that will eventually dissipate, and economic agents do not expect severe price pressures to persist beyond the pandemic. On balance, this points to a stagflationary outcome over the coming 6-24 months as a risk, but not a likely event. Investors should use the Misery Index, which is the sum of the unemployment rate and headline PCE inflation, as a real-time stagflation indicator. The Misery Index underscores that the US economy is unlikely to experience true stagflation unless the unemployment rate rises. A portfolio of the US dollar, the Swiss Franc, and industrial commodities may serve as a useful hedge for investors who are concerned about absolute return prospects in a world in which long-maturity bond yields are rising and risks of stagflationary dynamics are present. Chart II-1The Misery Index Reflects The Risk Of Stagflation Over the past several weeks, concerns about a possible return to 1970s-style stagflation have re-emerged significantly in the minds of many investors. These investors have pointed toward similarities between the current environment and that of the 1970s, including shortages limiting output, a snarled global trade and logistical system, and rising energy prices. Chart II-1 highlights that the US “Misery Index” – the sum of the unemployment rate and headline PCE inflation – rose again over the past several months to high single-digit territory, after having fallen dramatically from April 2020 to February of this year. Panel 2 of Chart II-1 highlights that last year's rise in the Misery Index was driven almost entirely by the unemployment rate, whereas the current level is due to a combination of a modestly elevated unemployment rate and a pronounced acceleration in inflation. The headline PCE deflator has risen above 4%, a level that has not been reached since 1991 during the First Gulf War. In this report, we examine the risk of stagflation by comparing the current environment to that of the late 1960s and 1970s. We conclude that while investors cannot rule out the possibility of a stagflationary outcome, there are important differences that point toward a stagflation outcome over the coming 6-24 months as a risk, not a likely event. We conclude by highlighting assets that may produce absolute returns in a world in which long-maturity bond yields are rising and risks of stagflationary dynamics are present. Revisiting The 1960s And 70s Chart II-2The 1960s Laid The Groundwork For Elevated Inflation The first step in judging the risk of a return to 1970s-style stagflation is to review, in a detailed way, what caused those conditions. Investors are well aware of the role that two separate energy price shocks played in raising prices and damaging output during this period, but they are less cognizant of the impact that a persistent period of above-trend output and significant labor market tightness had in setting up the conditions for sharply higher inflation. This focus of investors on energy prices partially reflects the fact that the Misery Index increased most visibly in the 1970s and that policymakers in the 1960s may not have realized how extensively economic output was running above its potential. With the benefit of hindsight, Chart II-2 illustrates the extent to which inflationary pressures built up in the 1960s, well before the first oil price shock in 1973. The chart shows that the unemployment rate was below NAIRU – the non-accelerating inflation rate of unemployment – for 70% of the time during the 1960s, and that inflation had already responded to this in the latter half of the decade. Annual headline PCE inflation was running just shy of 5% at the onset of the 1970 recession; it fell to 3% in the aftermath of the recession, but had already begun to reaccelerate in the first half of 1973. Following the 1973/1974 recession, inflation did decelerate significantly, falling from 11-12% to 5% in headline terms, and from 10% to 6% in core terms. But the pace of price appreciation did not fall below 5-6% in the second half of the 1970s, despite a significant and sustained rise in the unemployment rate above its natural rate. The 1975 to 1978 period is especially important for investors to understand, because it is arguably the clearest period of true stagflation in the 1970s. The fact that the Misery Index rose sharply during two major oil price shocks is not particularly surprising in and of itself, given the direct impact of energy prices on headline consumer prices; it is the fact that the index remained so elevated between these shocks, the result of persistently high inflation in the face of significant labor market slack, that is most relevant to investors. There are two reasons that both inflation and unemployment remained high during this period. First, labor market slack was sizeable during these years because the US economy was more energy-intensive in the 1970s than it is today. Chart II-3 highlights that goods-producing employment lagged overall employment growth from late 1973 to late 1977, underscoring that the rise in oil prices significantly impacted jobs growth in energy-intensive industries. Second, it is clear that the combination of demand-pull inflation in the late 1960s and the predominantly cost-push inflation of the 1970s led to expectations of persistent inflation among households and firms. The original Phillips Curve, as formulated by New Zealand economist William Phillips in the late 1950s, described a negative relationship between the unemployment rate and the pace of wage growth. Given the close correlation between wage and overall price growth at the time, the Phillips Curve was soon extended and generalized to describe an inverse relationship between labor market slack and overall price inflation. But the experience of the 1970s highlighted that inflation expectations are also an important determinant of inflation, a realization that gave birth to the expectations-augmented (i.e. “modern-day”) Phillips Curve (more on this below). The Stagflation Era Versus Today Table II-1 presents a stagflation “threat matrix,” representing the Bank Credit Analyst service’s assessment of the various factors that could potentially contribute to a stagflationary environment today, relative to what occurred in the 1960s and 1970s. While we acknowledge that there are some similarities today to what occurred five decades ago, the most threatening factors have been present for a shorter period of time and appear to have a smaller magnitude than what occurred during the stagflationary era. In addition, key factors, such as the visibility available to policymakers and investors about household inflation expectations and the potential output of the economy, would appear to reduce significantly the risk of a stagflationary outcome today. We discuss each of the factors presented in Table II-1 below: Fiscal & Monetary Policy Chart II-4Government Spending Last Cycle Looked Nothing Like The 1960s The persistently tight labor market that contributed to the inflationary buildup in the 1960s occurred as a result of easy fiscal and monetary policy. Chart II-4 highlights that the contribution to real GDP growth from government expenditure and investment was very elevated in the 1960s. Chart II-5 shows that a positive output gap in the late 1960s and the first half of the 1970s is well explained by the fact that 10-year US government bond yields were persistently below nominal GDP growth. The relationship between the stance of monetary policy and the output gap only meaningfully diverged in the latter half of the 1970s, during the true stagflationary era that we noted above. Chart II-5Easy Monetary Policy Juiced Aggregate Demand In The 60s And Early 70s Chart II-6Monetary Policy Today Is Extremely Easy Today, it is clear that the stance of fiscal policy has recently been extraordinarily easy, and 10-year US government bond yields have remained well below nominal GDP growth for the better part of the last decade. Relative to estimates of potential nominal GDP growth, 10-year Treasury yields are the lowest they have been since the 1970s (Chart II-6). Ostensibly, this supports concerns that policy might contribute to a stagflationary outcome. These concerns were raised by Larry Summers in March, when he described the Biden administration’s fiscal policy as the “least responsible” that the US has experienced in four decades and warned of the potential inflationary consequences of overheating the economy.3 But there are two important counterpoints to these concerns. First, easy fiscal policy this cycle has followed a period during the last economic cycle in which government spending contributed to the most sustained drag on economic activity since the 1950s. Unlike the 1960s, the unemployment rate has been below NAIRU for only a third of the time over the past decade. In addition, Chart II-7 highlights that fiscal thrust will turn to fiscal drag next year, underscoring the temporary nature of the massive burst in fiscal spending that has occurred in response to the pandemic. Under normal circumstances, the fiscal drag implied by Chart II-7 would substantially raise the risks of a recession next year, but we have noted in previous reports that a significant amount of excess savings remain to support spending and employment. The net impact of these two factors results in a reasonable expectation that the US economy will return to maximum employment next year, but this is a far cry from the 1960s when the unemployment rate was below its natural rate for 70% of the decade. Based on conventional measures, US monetary policy has been easy for a decade, but easy monetary policy did not begin to contribute positively to a rise in household sector credit growth last cycle until 2014/2015. This underscores that the natural rate of interest (“R-star”) did fall during the early phase of the last economic expansion. However, we argued in an April report that R-star was likely rising in the latter half of the last expansion,4 and we believe that the terminal Fed funds rate is likely higher than what the Fed is currently projecting, barring any additional negative policy shocks. Thus, while we do not believe that the duration of easy monetary policy over the past decade has laid the groundwork for a major rise in prices, it is now clearly positively contributing to aggregate demand and does risk a future overshoot in prices if long maturity bond yields remain well below the pace of economic growth for a sustained period of time. The Impact Of Shortages Chart II-8Gasoline Shortages Plagued The US Economy In The 1970s Gasoline shortages occurred during the oil shocks of the 1970s and are a key similarity that some investors point toward when comparing the situation today with the stagflationary era. Chart II-8 highlights that the annual growth in real personal consumption expenditures on energy goods and services fell into negative territory on six occasions in the 1970s, although it was most pronounced during the two oil price shocks and their resulting recessions. Today, the impact of shortages appears to be broader than what occurred in the 1970s, but less impactful and not likely to be as long-lasting. Chart II-9 highlights that the OPEC oil embargo of 1973 raised the global oil bill by 2.4% of global GDP and permanently raised the price of oil. The global oil bill will only be fractionally above its pre-pandemic level in 2022, with oil prices at $80/bbl, and, while it is true that US gasoline prices have risen significantly, they are not higher than they were from 2011-2014 (Chart II-10). Chart II-9$80/bbl Oil Is Not Onerous Chart II-10US Gasoline Prices Are High, But They Have Been Higher It is certainly true that global shipping costs have skyrocketed and that this is contributing to the increase in US consumer prices. We estimate, however, that this increase in shipping costs as a share of GDP is no more than a quarter of the impact of the 1973 increase in oil prices, without the attendant negative effects on US goods-producing employment that occurred in the 1970s. If anything, surging shipping costs create an incentive to re-shore manufacturing production, which would contribute positively to US goods-producing employment. We also do not expect the rise in shipping costs to be meaningfully permanent, i.e., shipping costs may ultimately settle at a higher level than they were in late-2019, but at a much lower level than what prevails today. Chart II-11A Tight Labor Market Is Causing Wage Growth To Pick Up Semiconductor and labor shortages would appear to represent a more salient threat of stagflation in the US, as the domestic production of motor vehicles cannot occur without key inputs and a tight labor market is already contributing to an acceleration in wage growth (Chart II-11). As we noted in Section 1 of our report, auto production significantly impacted growth in the third quarter. However, Chart II-12 highlights that, for now, the breadth of impact of these shortages appears to be limited: the production component of the ISM manufacturing index remains in expansionary territory, industrial production of durable manufacturing excluding motor vehicles and parts has not broken down, and both housing starts and building permits remain above pre-pandemic levels despite this year’s downtrend in permits. Chart II-12Shortages Do Not Yet Seem To Be Broad-Based A physical shortage of components is a less relevant factor for the services side of the economy, which appears to have re-accelerated meaningfully in October. The services sector is more considerably impacted by shortages in the labor market, which seem to be linked to a still-low labor force participation rate. We noted in our September report that the decline in the participation rate has significantly overshot what would be implied by the ongoing pace of retirements. Chart II-13 highlights that this has occurred not just because of a significant retirement effect, but also because of the shadow labor force (people who want a job but are not currently looking for work) and family responsibilities. We expect that the recent expiry of expanded unemployment insurance benefits, a steady rise in the immunity of the US population, an abating Delta wave of COVID-19, and a likely upcoming reduction in school/classroom closures once the Pfizer/BioNTech vaccine is approved for school-age children will likely ease the labor shortage issue over the coming several months. Output Gap Uncertainty It remains a debate among economists why policymakers maintained such easy monetary policy in the 1960s and 1970s, but Chart II-14 highlights that uncertainty about the size of the output gap may have contributed to too-low interest rates. The chart shows the unemployment rate compared with today's estimate of NAIRU, alongside a simple proxy for policymakers’ real time estimate of the natural rate of employment: the cumulative average unemployment rate in the post-war environment. To the extent that policymakers used past averages of the unemployment rate as their guide for NAIRU, Chart II-14 highlights how they may have underestimated the degree to which output was running above its potential level in the 1960s, and would not have even concluded that output was above potential in the early 1970s. Chart II-14Policymakers Overestimated Labor Market Slack In The 60s And 70s Chart II-15Policymakers Know That NAIRU Is Likely At Or Below 4% Today, the environment is quite different, because the acceleration in wage growth at the tail end of the last expansion gives policymakers and investors a good estimate of where NAIRU is. Chart II-15 highlights that wage growth accelerated in 2018/2019 in response to a sub-4% unemployment rate, which is consistent with both the Fed’s NAIRU estimate of 3.5-4.5% and Fed Vice Chair Richard Clarida’s expressed view that a 3.8% unemployment rate likely constitutes maximum employment (barring any issues with the breadth and inclusivity of the labor market recovery). It is possible that the pandemic has structurally lowered potential output, which could mean that policymakers may no longer rely on the wage growth / unemployment relationship that existed in the latter phase of the last expansion. However, we do not find any credible arguments that would support the notion of a structurally lower level of potential output: the pandemic is likely to end at some point in the not-too-distant future, the negative impact of working-from-home policies on office properties and employment in central business districts is not sizeable,5 and productivity may have permanently increased in some industries because of the likely stickiness of a hybrid work culture. The Behavior Of Inflation Expectations Chart II-16Rising Long-Term Expectations Have Merely Normalized (For Now) One parallel to the argument that policymakers may have underestimated the degree of labor market tightness in the 1960s and early 1970s is the fact that they did not yet understand that inflation expectations are an important determinant of actual inflation, nor were they able to monitor them even if they did. Most credible surveys of inflation expectations began in the 1980s, and policymakers in the 1960s and 1970s were guided by the original Phillips Curve that solely related inflation to unemployment. Today, policymakers have the experience of the stagflationary episode to serve as a warning not to allow inflation expectations to get out of control, and both policymakers and investors have reliable measures of inflation expectations for households and market-participants. Chart II-16 highlights that households expect significant inflation over the coming year, but also expect prices over the longer term to rise at a pace that is almost exactly in line with their average from 2000-2014. The Rudd Controversy: (Adaptive) Inflation Expectations Do Matter One potential criticism of the idea that inflation expectations are signaling a low risk of higher future inflation has emerged through arguments made by Jeremy Rudd, a Federal Reserve economist. In a recent paper, Rudd questioned the view that households’ and firms’ expectations of future inflation are a key determinant of actual inflation; he suggested instead that relatively stable inflation since the mid-1990s might reflect a situation in which inflation simply does not enter workers’ employment decisions and expectations are irrelevant. Rudd’s paper was primarily addressed to policymakers who view inflation dynamics in a highly quantitative light. A full response to the paper would be mostly academic and thus not especially relevant to investors; however, we would like to highlight three points related to the Rudd piece that we feel are important.6 First, we disagree with Rudd’s argument that the trend in inflation has not responded to changes in economic conditions since the mid-1990s. Chart II-17 highlights that while the magnitude of the relationship has shifted, the trend in inflation relative to a measure of long-term expectations based on prior actual inflation has mimicked that of the output gap. The fact that inflation was (ironically) too high during the early phase of the last economic cycle provides some support for Rudd’s inflation responsiveness view, although we would still point toward the Fed’s strong record of maintaining low and stable inflation, its active communication with the public in the years following the global financial crisis, and the fact that a recovery began and the output gap began to (slowly) close as the best explanation for the avoidance of deflation during that period. Second, we agree with Rudd’s point that regime shifts in inflation’s responsiveness to economic conditions can occur, and that adaptive measures of inflation expectations, and even surveys of inflation, may not capture such a shift in real time. Chart II-18 shows that the 2014-2016 period was a good example of this, when adaptive expectations as well as household survey measures of long-term inflation expectations both lagged the actual decline in inflation that was caused by a collapse in the price of oil. Chart II-17The Trend In Inflation Continues To Respond To Economic Conditions Chart II-18Surveyed Inflation Expectations Can Lag, But This Time They Led But Chart II-18 also shows that long-term household survey measures of inflation led the rise in actual inflation (and thus our adaptive expectations measure) last year, underscoring that these measures are likely more reliable indicators today of whether a major regime shift is occurring. As noted above, long-term expectations have risen significantly relative to what prevailed prior to the pandemic, but this has merely raised expectations from extraordinarily depressed levels back to the average that prevailed prior to (and immediately after) the global financial crisis. Therefore, household expectations are not yet at dangerous levels. Chart II-19Unit Labor Costs Modestly Lead Inflation, But Are Far From Extreme Third, one of the core observations in Rudd’s paper is that unit labor cost (ULC) growth leads the trend in inflation, which he argued was evidence against the idea that expectations of future inflation are a key determinant of actual inflation. Chart II-19 highlights that Rudd is correct that ULC growth modestly leads inflation (especially core inflation), but we disagree with his conclusion that it argues against the importance of expectations. As we noted in Section 2 of our January 2021 Bank Credit Analyst,7 one crucial aspect of the expectations-augmented, or “modern-day” Phillips Curve is that, if inflation expectations are largely formed based on the experience of past inflation, then inflation is ultimately determined by three dimensions of the output gap: whether it is rising or falling, whether it is above or below zero, and how long it has been above or below zero. Our view is that ULC growth is fundamentally linked to slack in the labor market, which is directly incorporated in output gap measures. As we noted above, investors currently have a good estimate of the magnitude of the output/employment gap, meaning that it is possible to track the inflationary consequences of prevailing aggregate demand. As a final point about ULC growth, Chart II-19 highlights that while the five-year CAGR of unit labor costs is currently running at its strongest pace since the global financial crisis, investors should note that it remains well below the levels that prevailed in the late-1960s when persistently above-potential output laid the groundwork for a massive inflationary overshoot. Conclusions And Investment Strategy Our review of the 1960s and 1970s highlights that stagflation is a phenomenon in which supply-side shocks raise prices of key inputs to production, which lowers output and raises unemployment. Energy price shocks in the 1970s occurred after a long period of policy-driven above-trend growth in the 1960s, meaning that both demand-pull and cost-push inflation contributed to stagflation in the 1970s. Today, investors cannot rule out the possibility of a stagflationary outcome, for four reasons: long-term household inflation expectations have risen significantly over the past year; fiscal policy has been very expansionary; monetary policy will remain expansionary at the Fed’s projected terminal Fed funds rate; and component shortages and price increases linked to energy market and supply chain disruptions may persist or worsen over the coming year. Chart II-20It Is Not Stagflation If The Unemployment Rate Continues To Fall However, the strong demand-pull inflationary dynamics that existed in the late-1960s were mostly absent in the lead-up to the pandemic, supply-chain issues are in part the result of strong goods demand and disruptions that are clearly linked to the pandemic (and thus will eventually dissipate), and long-term inflation expectations are behaving differently than short-term expectations, signaling that economic agents do not expect severe price pressures to persist beyond the pandemic. Policymakers also have more visibility about the magnitude of economic / labor market slack than they did during the stagflationary era and better tools to track inflation expectations. On balance, this points to a stagflationary outcome over the coming 6-24 months as a risk, but not as a likely event. Using the Misery Index as real-time stagflation indicator, investors should note that the US economy is not likely experiencing true stagflation unless the unemployment rate rises. Chart II-20 highlights that there is no evidence yet of a contraction in goods-producing or service-producing jobs. Even if goods-producing employment slows meaningfully over the coming few months as a result of component shortages, the unemployment rate is still likely to fall if services spending normalizes, as it would imply that the gap in services-producing employment, which is currently 20% of the level of pre-pandemic goods-producing employment, will continue to close. Investors have been focused on the issue of stagflation because its occurrence would imply a sharply negative correlation between stock prices and bond yields. This is not our base case view, but we have highlighted that months with negative returns from both stocks and long-maturity bonds tend to be associated with periods of monetary policy tightening (or in anticipation of such periods). As we discussed in Section 1 of our report, we do expect the Fed to raise interest rates next year. We do not see a rise in bond yields to levels implied by the Fed’s interest rates projections as being seriously threatening to economic activity, corporate earnings growth, or equity multiples. But the adjustment to higher long-maturity bond yields may unnerve equity investors for a time, implying temporary periods of a negative stock price / bond yield correlation. Table II-2 highlights that, since 1980, commodities, the US dollar, and the Swiss franc have typically earned positive returns during non-recessionary months in which stock and long-maturity bond returns are negative. While the dollar is not likely to perform well in a stagflationary scenario, Chart II-21 highlights that CHF-USD and industrial commodities performed quite well in the late-1970s. As such, a portfolio of these three assets might serve as a useful hedge for investors who are concerned about absolute return prospects in a world in which long-maturity bond yields are rising and risks of stagflationary dynamics are present. Chart II-21The Swiss Franc and Raw Industrials Did Well During The Stagflationary Era Jonathan LaBerge, CFA Vice President The Bank Credit Analyst III. Indicators And Reference Charts BCA’s equity indicators highlight that the “easy” money from expectations of an eventual end to the pandemic have already been made. Our technical, valuation, and sentiment indicators remain very extended, highlighting that investors should expect positive but modest returns from stocks over the coming 6-12 months. Our monetary indicator has retreated below the boom/bust line, although this mostly reflects the use of producer prices to deflate money growth. In nominal terms, the supply of money continues to grow. Still, the retreat in the indicator over the past year highlights that the monetary policy stance is likely to shift in a tighter direction over the coming year. Forward equity earnings are pricing in a substantial further rise in earnings per share. Net earnings revisions and net positive earnings surprises appear to have peaked, but there is not yet any meaningful sign of waning forward earnings. Bottom-up analyst earning expectations remain too high, but stocks are likely to be supported by robust revenue growth over the coming year. Within a global equity portfolio, we would continue to recommend that investors position for the underperformance of financial assets that are negatively correlated with long-maturity government bond yield. The US 10-Year Treasury yield remains above its 200-day moving average after failing to break meaningfully below it. 10-Year Treasury Yields remain below the fair value implied by a late-2022 rate hike scenario, underscoring that a move higher over the coming year is likely. However, most of the recent move higher in government bond yields has occurred due to rising inflation expectations, whereas the increase in yields over the coming year will likely occur in real terms. Commodity prices remain elevated, and our composite technical indicator highlights that they are still overbought. An eventual slowdown in US goods spending, coupled with eventual supply-chain normalization and the absence of a significant reflationary impulse from Chinese policy, may weigh on commodity prices at some point over the coming 6-12 months. US and global LEIs remain very elevated but have started to roll over. Our global LEI diffusion index has declined very significantly, but this likely reflects the outsized impact of a few emerging market countries (whose vaccination progress is still lagging). Still-strong leading and coincident indicators underscore that the global demand for goods is robust, and that output is below pre-pandemic levels in most economies because of very weak services spending. The latter will recover significantly at some point over the coming year, as social distancing and other pandemic control measures disappear. EQUITIES: Chart III-1US Equity Indicators Chart III-2Willingness To Pay For Risk Chart III-3US Equity Sentiment Indicators Chart III-4US Stock Market Breadth Chart III-5US Stock Market Valuation Chart III-6US Earnings Chart III-7Global Stock Market And Earnings: Relative Performance Chart III-8Global Stock Market And Earnings: Relative Performance FIXED INCOME: Chart III-9US Treasurys And Valuations Chart III-10Yield Curve Slopes Chart III-11Selected US Bond Yields Chart III-1210-Year Treasury Yield ComponentsChart III-13US Corporate Bonds And Health Monitor Chart III-14Global Bonds: Developed Markets Chart III-15Global Bonds: Emerging Markets CURRENCIES: Chart III-16US Dollar And PPP Chart III-17US Dollar And Indicator Chart III-18US Dollar Fundamentals Chart III-19Japanese Yen Technicals Chart III-20Euro Technicals Chart III-21Euro/Yen Technicals Chart III-22Euro/Pound Technicals COMMODITIES: Chart III-23Broad Commodity Indicators Chart III-24Commodity Prices Chart III-25Commodity Prices Chart III-26Commodity Sentiment Chart III-27Speculative Positioning ECONOMY: Chart III-28US And Global Macro Backdrop Chart III-29US Macro Snapshot Chart III-30US Growth Outlook Chart III-31US Cyclical Spending Chart III-32US Labor Market Chart III-33US Consumption Chart III-34US Housing Chart III-35US Debt And Deleveraging Chart III-36US Financial Conditions Chart III-37Global Economic Snapshot: Europe Chart III-38Global Economic Snapshot: China Jonathan LaBerge, CFA Vice President The Bank Credit Analyst   Footnotes 1  Please see Section 1 of the September 2021 Bank Credit Analyst for a detailed discussion of the US immunity level. 2  Please see US Equity Strategy "Marginally Worse," dated October 11, 2021, available at uses.bcaresearch.com 3  “Summers Sees ‘Least Responsible’ Fiscal Policy in 40 Years,” Bloomberg News, March 20, 2021. 4 Please see The Bank Credit Analyst “R-star, And The Structural Risk To Stocks,” dated March 31, 2021, available at bca.bcaresearch.com 5 Please see The Bank Credit Analyst “Work From Home “Stickiness” And The Outlook For Monetary Policy,” dated June 24, 2021, available at bca.bcaresearch.com 6 Rudd, Jeremy B. (2021). “Why Do We Think That Inflation Expectations Matter for Inflation? (And Should We?),” Finance and Economics Discussion Series 2021-062. Washington: Board of Governors of the Federal Reserve System. 7  Please see The Bank Credit Analyst “The Modern-Day Phillips Curve, Future Inflation, And What To Do About It,” dated December 18, 2021, available at bca.bcaresearch.com EQUITIES:FIXED INCOME:CURRENCIES:COMMODITIES:ECONOMY:

Highlights Increasing consumption should be a lot easier than increasing savings. After all, most people like to spend! It is getting them to work that should be challenging. Yet, the conventional wisdom is that deflation is a much tougher problem to overcome than inflation. It is true that the zero-bound constraint on interest rates makes it more difficult for central banks to react to deflationary forces. However, monetary policy is not the only game in town; fiscal policy becomes more effective as interest rates fall because governments can stimulate the economy without incurring onerous financing costs. When the borrowing rate is below the growth rate of the economy, the more profligate a government has been in the past, the more profligate it can be in the future, while still maintaining a stable debt-to-GDP ratio. The pandemic banished the bond vigilantes. Governments ran massive budget deficits, but bond yields still dropped. While budget deficits will decline from their highs, fiscal policy will remain structurally more accommodative in the post-pandemic period. The combination of easier fiscal policy, increased household net worth, and other factors has raised the neutral rate of interest in the US and most other economies. This means that monetary policy is currently much more stimulative than widely believed. This is good news for equities and other risk assets in the near term, even if it does produce a major hangover down the road. New trade: Short US consumer discretionary stocks relative to other cyclicals. Consumer durable goods spending will slow as services spending and capex continue to recover. A Paradoxical Problem Economic pundits like to say that deflation is a tougher problem to overcome than inflation. We hear this statement so often that we do not think twice about it. In many respects, it is a rather strange perspective. Inflation results from too much spending relative to output, whereas deflation results from too little spending. Yet, people like to spend! One would think it would be much easier to get people to consume than to get them to work. The claim that deflation is a bigger problem than inflation is really just a statement about the limits of monetary policy. If the economy is overheating, central banks can theoretically raise rates as high as they want. In contrast, if the economy is in a deflationary funk, the zero-bound constraint limits how far interest rates can fall. Fortunately, there are other ways of stimulating the economy when interest rates cannot be cut any further. Most notably, governments can utilize fiscal policy by cutting taxes, spending more on goods and services, or increasing transfer payments. Getting Paid To Eat Lunch When interest rates are very low, not only is fiscal stimulus a free lunch, but you actually get paid for eating more. If the borrowing rate is below the growth rate of the economy, the more profligate a government has been in the past, the more profligate it can be in the future, while still maintaining a stable debt-to-GDP ratio. This sounds so counterintuitive that it is worth thinking through a simple example. Suppose you currently earn $100,000 per year and expect your income to rise by 8% per year. You have $100,000 in debt, which incurs an interest rate of 3%, and want to keep your debt-to-income ratio constant at 100% over time. Next year, your income will be $108,000, so you should target a debt level of $108,000. Thus, this year, you can spend $105,000 on goods and services, make $3,000 in interest payments, and take on $8,000 in additional debt. Now, suppose you have been spendthrift in the past and have accumulated $200,000 in debt. You still want to keep your debt-to-income ratio constant, but this time at 200%. How much can you spend this year? The answer is $110,000. If you spend $110,000 and pay an additional $6,000 in interest, your cash outflows will exceed your income by $16,000, taking your debt to $216,000 — exactly twice next year’s income. Notice that by maintaining a higher debt balance, you can actually spend $5,000 more while still keeping your debt-to-income ratio constant. Appendix A proves this point mathematically. One might protest that the interest rate you face would be higher if you had more debt. Fair enough, although in our example, the interest rate would need to rise above 5.5% for spending to decline. The more important point is that unlike people, governments which issue debt in their own currencies get to choose whatever interest rate they want. Granted, if central banks set interest rates too low, the economy will overheat, leading to higher inflation. But this just reinforces the point we made at the outset, which is that inflation and not deflation is the real constraint to macroeconomic policy. A Blissful Outcome For Stocks We would not have waded through this theoretical discussion if it did not serve a practical purpose. In April of last year, we wrote a controversial report asking if, paradoxically, the pandemic could turn out to be good for stocks. We noted that by combining monetary easing with fiscal stimulus, policymakers could steer equity markets towards a “blissful outcome” where the economy was operating at full capacity, yet interest rates were lower than they were before (Chart 1). If such a blissful state were reached, earnings would return to their pre-pandemic level, but the discount rate would remain below its pre-pandemic level, thus allowing stock prices to rise above their pre-pandemic peak. In the months following our report, the stock market played out this narrative.   From Blissful To Blissless? Chart 2Both The Fed And Investors Have Lowered Their Estimate Of The Neutral Rate More recently, bond yields have risen, stoking fears that we are moving towards less auspicious conditions for equities. There is no doubt that many central banks are looking to normalize monetary policy. That said, what central banks regard as normal today is very different from what they thought was normal in the past. Back in 2012, when the Fed began publishing its “dot plot,” the FOMC thought the neutral rate of interest was around 4.25%. Today, it thinks the neutral rate is only 2.5%. And based on the New York Fed’s survey of market participants and primary dealers, investors believe the neutral rate is even lower than the Fed’s estimate (Chart 2). Even if the Fed did not face political pressure to keep interest rates low, it probably would not want to raise them all that much anyway. The same applies to most other central banks. Why The Neutral Rate Is Higher Than The Fed Believes There are at least four reasons to think that the neutral rate of interest is higher than what the Fed believes:   Reason #1: The drag on growth from the household deleveraging cycle is ending As a share of disposable income, US household debt has declined by nearly 40 percentage points since 2008. Debt-servicing costs are now at record low levels (Chart 3). The Fed’s Senior Loan Officer Survey points to an increasing willingness to lend (Chart 4). The Conference Board’s Leading Credit Index also remains in easing territory (Chart 5). Chart 3The Deleveraging Cycle Has Run Its Course Real personal consumption increased by only 1.6% in Q3. However, this was largely driven by a 54% drop in auto spending on the back of the semiconductor shortage. While vehicle purchases normally account for only 4% of consumer spending, the sector still managed to shave 2.4 percentage points off GDP growth in Q3. Chart 4Banks Are Easing Credit Standards Chart 5A Positive Signal For Credit Growth Spending on services rose by 7.9%, an impressive feat considering the quarter saw the peak in the Delta variant wave.   Reason #2: Fiscal policy is likely to remain accommodative in the post-pandemic period The combination of lower real rates and higher debt levels has increased the budget deficit consistent with a stable debt-to-GDP ratio in the US and most developed markets (Chart 6). This point has not been lost on governments. While the flow of red ink will abate, the IMF estimates that the US cyclically-adjusted primary budget deficit will be 3% of GDP larger in 2022-26 than it was in 2014-19. The IMF also expects most other advanced economies to run larger budget deficits (Chart 7). Chart 8A Record Rise In Household Net Worth Reason #3: Higher asset prices will bolster spending According to the Federal Reserve, US household net worth rose by over 113% of GDP between 2019Q4 and 2021Q2, the largest six-quarter increase on record (Chart 8). Empirical estimates of the wealth effect suggest that households spend about 5-to-8 cents on goods and services for every additional dollar of housing wealth, and 2-to-4 cents for every additional dollar of equity wealth. Based on the latest available data, we estimate that US homeowner equity has increased by $5 trillion since the start of 2020, while household equity holdings have increased by $15.8 trillion. Together, this would translate into 2.5%-to-4% of GDP in additional annual consumption. And this does not even include any spending arising from the $2.4 trillion in incremental bank deposits that households have amassed since the start of the pandemic.    Chart 9Most Of The Deceleration In US Potential Real GDP Growth Has Already Occurred Reason #4: Population aging will drain savings Aging populations can affect the neutral rate either by dragging down investment demand or reducing savings. The former would lead to a lower neutral rate, while the latter would lead to a higher rate. As Chart 9 shows, most of the decline in US potential GDP growth has already occurred. According to the Congressional Budget Office, real potential GDP growth fell from over 3% in the early 1980s to about 1.8% today, mainly due to slower labor force growth. The CBO expects potential growth to edge down to 1.5% over the next few decades. The average age of the US capital stock is now the highest on record (Chart 10). Whereas real business fixed investment is 6% below its pre-pandemic trend, core capital goods orders – a leading indicator for capex – are 17% above trend. Capex intentions remain near multi-year highs (Chart 11). All this suggests that investment spending is unlikely to fall much in the future. Chart 10The Average Age Of The US Capital Stock Is Now The Highest On Record Chart 11Capex Intentions Remain At Lofty Levels In contrast, the depletion of national savings from an aging population is just beginning. Baby boomers are leaving the labor force en masse. They hold over half of US household wealth, considerably more than younger generations (Chart 12). As baby boomers transition from net savers to net dissavers, national savings will fall. UnTaylored  Monetary Policy The Taylor Rule prescribes the Fed to hike rates by between 50-to-100 bps for each percentage point that output rises relative to its potential. Over the past decade, the Fed has favored the higher output gap coefficient, meaning that a permanent one percentage-point increase in aggregate demand should translate, all things equal, into a one percentage-point increase in the neutral rate of interest. Taken at face value, the combination of increased household wealth and looser fiscal policy may have raised the neutral rate in the US by more than five percentage points since the pandemic. This estimate, however, does not consider feedback loops: A higher term structure for interest rates would depress asset prices, thus obviating some of the wealth effect. Higher rates would also reduce the incentive for governments to run large budget deficits. Taking these feedback loops into account, a reasonable estimate is that the neutral rate in the US is about 2% in real terms, or slightly over 4% in nominal terms based on current long-term inflation expectations. This is close to the historic average for real rates, although well above current market pricing. The implication for investors is that US monetary policy is currently more stimulative than widely believed. This is the good news. The bad news is that in the absence of fiscal tightening, the Fed will eventually be forced to raise rates by more than investors are discounting. Higher Inflation Won’t Force The Fed’s Hand… Just Yet When will the Fed be forced to move away from its baby-step approach to monetary policy normalization and adopt a more aggressive stance? Our guess is not for another two years. Last week, we argued that inflation in the US and many other countries is likely to follow a “two steps up, one step down” trajectory of higher highs and higher lows over the remainder of the decade. We are currently near the top of those two steps: Most of the recent increase in inflation has been driven by surging durable goods prices (Chart 13). Considering that durable goods prices usually fall over time, this is not a sustainable source of inflation. Chart 13ADurable Goods Spending Has Further To Fall (I) Chart 13BDurable Goods Spending Has Further To Fall (II) In modern service-based economies, structurally high inflation requires rapid wage growth. While US wage growth has picked up recently, most of the increase in wages has occurred at the bottom end of the income distribution (Chart 14). The Fed welcomes this development, given its expanded mandate to pursue “inclusive growth.” At some point in the future, long-term inflation expectations could become unmoored. However, that has not happened yet, whether one looks at market-based or survey-based expectations (Chart 15). Thus, for now, investors should remain constructive on stocks. Chart 14Wages At The Bottom End Of The Income Distribution Are Rising Briskly   New Trade: Short Consumer Discretionary Stocks Relative To Other Cyclicals We continue to favor cyclical stocks over defensives. Within the cyclical category, however, we are cautious on consumer discretionary names. Spending on consumer durable goods still has further to fall in order to return to trend. Durable goods prices will also come down, potentially squeezing profit margins. Go short the Consumer Discretionary Select Sector SPDR Fund (XLY) versus an S&P 500 sector-weighted basket of the Industrial Select Sector SPDR Fund (XLI), the Energy Select Sector SPDR Fund (XLE), and the Materials Select Sector SPDR Fund (XLB). Appendix A Peter Berezin Chief Global Strategist pberezin@bcaresearch.com View Matrix Special Trade Recommendations This table provides trade recommendations that may not be adequately represented in the matrix on the preceding page. Current MacroQuant Model Scores

Highlights Inflation in the US and many other countries is likely to follow a “two steps up, one step down” trajectory of higher highs and higher lows over the remainder of the decade. Goods inflation will ease in 2022, while energy price pressures will abate. This suggests that we are currently near the top of those two steps. Any decline in inflation will be short-lived, however. Tight labor markets will bolster wages. Rent inflation is also poised to pick up, especially in the US. The Fed and other central banks will face political pressure to keep interest rates low in order to suppress debt-servicing costs. This could lead to overheating. While we are not as bullish on stocks as we were at the start of the year, the combination of low interest rates and above-trend growth over the next 12 months will support equities. Investors should favor cyclicals, value stocks, small caps, and non-US markets. The Stairway To Higher Inflation In past reports, we argued that global inflation had reached a secular bottom and would begin to reaccelerate (see here, here, and more recently here). While it is still too early to be certain, recent developments appear to have vindicated that view. The path to structurally higher inflation is likely to be a bumpy one. We have generally contended that the shift to a more inflationary regime would follow a “two steps up, one step down” pattern, characterized by a series of higher highs and higher lows for inflation. In thinking about the inflation process, it is useful to distinguish between transitory shocks and structural forces. Unfortunately, much of the recent discussion about inflation has been politically charged, with one camp arguing that high inflation is entirely transitory (mainly due to pandemic disruptions) and another camp arguing that it is entirely structural in nature (big budget deficits, QE, and “dollar debasement” are often cited). The idea that both transitory shocks and structural forces may be driving inflation seems to generate a lot of cognitive dissonance in peoples’ minds. Our view is that transitory shocks have pushed up inflation, but that structural forces (both policy and non-policy related) are playing an important role too. In other words, we think that we are near the top of those metaphorical two steps. The next step for inflation is likely down, even though the longer-term trend is to the upside. Team Transitory Is Right About One Thing During most recessions, cyclically-sensitive durable goods spending falls, while the service sector serves as a ballast for the economy. The pandemic flipped this pattern on its head (Chart 1). While durable goods spending did dip briefly, it came roaring back due to generous stimulus payments and stay-at-home restrictions which cut many households off from the services they normally purchase. In March of this year, US real consumer durable spending was 27% above its pre-pandemic trend (Chart 2A and 2B). Chart 1Unlike During Most Recessions, Durable Goods Spending Spiked Due To Stimulus Checks And Stay-At-Home Restrictions Chart 2ADurable Goods Spending Has Begun To Normalize, But Durable Goods Prices Keep Rising Due To Supply Bottlenecks (I) Chart 2BDurable Goods Spending Has Begun To Normalize, But Durable Goods Prices Keep Rising Due To Supply Bottlenecks (II)     Durable goods spending has retreated since then, however. As of August, it was only 8% above its trendline. Supply-chain bottlenecks have curbed durable goods spending over the past eight months. A tell-tale sign of a supply shock is when spending declines and prices nonetheless rise. Between January 2020 and March 2021, durable goods spending increased at an annualized rate of 29% while prices rose at an annualized pace of 2%. Since March 2021, durable goods spending has fallen at an annualized pace of 28%, but price inflation has accelerated to 15% (Chart 3). Even more than other categories of durable goods, vehicle production has been stymied by supply-chain disruptions. Motor vehicles and auto parts represent about 40% of the durable goods sold in the US and accounted for nearly two-thirds of the decline in real durable goods spending between March and August. The downward trend in vehicle sales continued in September, with unit sales declining by 7.2% on the month. In the US, vehicle sales are now back to where they were in 2011 when the unemployment rate was 9%. In the euro area, they are below their sovereign debt crisis lows (Chart 4). The chip shortage hampering vehicle production will abate in 2022. However, vehicle prices are likely to come down only slowly. Auto inventories in the US are only a third of what they were prior to the pandemic (Chart 5). Until dealers are able to rebuild inventories, they will have little incentive to cut prices. Chart 4The Chip Shortage Has Caused Auto Sales To Tumble Chart 5Dealer Inventories Have Collapsed   Energy Price Pressures Should Abate, But Probably Not As Fast As Investors Expect Investors believe the recent surge in energy prices will reverse. The futures curves for oil, natural gas, and coal are all in steep backwardation (Chart 6). We agree that energy price pressures are likely to abate in 2022. However, as we discussed last week, the odds are that prices do not fall as quickly as anticipated. This concern is especially acute in Europe, where La Niña could lead to another cold winter and uncertainty abounds over the status of the Nord Stream 2 pipeline. Looking beyond the next 12 months, the risk is that years of declining investment in the oil and gas sector lead to continued energy shortages during the remainder of the decade. In 2020, 12% of global energy production came from renewable sources such as solar, wind, and hydro. The IEA estimates that this share will rise to 20% in 2030. However, the IEA also reckons that the global economy will still need about 5% more oil and natural gas than it consumes now (Table 1). Given the reluctance of many countries to invest in nuclear power generation, the phase-out of carbon-based fuels may take longer than expected. Table 1Oil And Gas Consumption Will Not Peak Until The Next Decade   Near-Term Upside For Rents Despite increasing home prices in most economies, rent inflation decelerated in the first year of the pandemic (Chart 7). More recently, however, the rental market has begun to heat up. US rents rose by 0.5% in September, the fastest monthly growth since the 2006 housing boom (Chart 8). The Zillow rent index, which looks only at units turning over, has spiked (Chart 9). Chart 7Rent Inflation Is Bouncing Back After Falling During The Pandemic Chart 8More Upside To Rent Inflation   Strong job growth, the end of the nationwide eviction moratorium, and the loosening of regulations freezing rents in a number of US cities and states are all contributing to higher rent inflation. A shortage of homes is also putting upward pressure on home prices and rents. After having surged during the Great Recession, the homeowner vacancy rate has fallen to record low levels (Chart 10). Chart 9Newly Listed Apartments Are Being Marked Up Sharply Chart 10The Home Vacancy Rate Is Very Low In addition to encouraging more construction, higher home prices could indirectly boost inflation through the wealth effect. According to the Federal Reserve, homeowner equity increased by $4.1 trillion, or 21%, between 2019Q4 and 2021Q2. Empirical estimates of the wealth effect suggest that consumption rises between 5 and 8 cents for every additional dollar in housing wealth. For the US, this would translate into 0.9%-to-1.4% of GDP in incremental annual consumption since the start of the pandemic. Higher Nominal Income Growth Would Make Housing More Affordable Chart 11Many Developed Economies Feature Overheated Housing Markets The housing wealth effect would turn negative if home prices were to fall. While this is less of a risk in the US where housing is still reasonably affordable in many states, it is more of a risk in countries such as Canada, Australia, New Zealand, and Sweden where home prices have reached stratospheric levels in relation to incomes and rents (Chart 11). Not only would a decline in nominal home prices curb construction and consumer spending, but it would also potentially undermine the financial system by reducing the value of the collateral backing mortgage loans. To support spending and preclude an outright fall in home prices, central banks would likely keep interest rates at fairly low levels. Low rates, in turn, would incentivize governments to maintain accommodative fiscal policies. The IMF expects the cyclically-adjusted primary budget deficit to be 2% of GDP larger in advanced economies in 2022-26 compared to 2014-19 (Chart 12). The combination of low interest rates and loose fiscal policies will help drive nominal income growth, thus allowing for improved home affordability without the need for a disruptive decline in home prices. As Japan’s experience demonstrates, a deflationary environment is toxic for the property market and the financial system. Labor Markets Getting Tighter There is little doubt that the US labor market is heating up. Even though there are 5 million fewer people employed now than at the start of the pandemic, the job vacancy rate is near record high levels and workers are displaying few misgivings about quitting their jobs (Chart 13). Part of the apparent tightness in the US labor market stems from pandemic-related factors. Although enhanced federal unemployment benefits have expired, households are still sitting on $2.4 trillion in excess savings (Chart 14). This cash cushion has allowed workers to be choosy in entertaining job offers. In addition, decreased immigration flows and a spate of early retirements have decreased labor supply. More recently, the introduction of vaccine mandates has caused some disruptions to the labor market. About 100 million US workers are currently subject to the mandates. According to the Census Household Pulse Survey, about 8 million of them are unvaccinated and attest that “they will definitely not get the vaccine.” Although many of them will reconsider, the anecdotal evidence suggests that some will not. In one glaring example, 4.6% of workers resigned from a rural hospital in upstate New York, causing the maternity ward to temporarily suspend operations. Prospects For A Wage-Price Spiral Chart 15Wages At The Bottom End Of The Income Distribution Are Rising Briskly So far, much of the pick-up in wage growth has been confined to the bottom end of the income distribution (Chart 15). Wage pressures are likely to become more broad-based over time as the unemployment rate continues to decline. A full-blown wage-price spiral would worry the Fed. However, such a spiral does not appear imminent. While respondents to the University of Michigan survey in October expected inflation to reach 4.8% over the next 12 months, they anticipated inflation of only 2.8% over a 5-to-10-year horizon (Chart 16). This is not much higher than their pre-pandemic expectations and is lower than the 3.0% figure reported for September. Chart 16Long-Term Inflation Expectations Have Risen But Remain At Historically Low Levels It is easy to dismiss households’ beliefs about future inflation as being largely irrelevant. However, these beliefs do influence spending decisions. For example, a record share of households say that this is a bad time to buy a car (Chart 17). The top reason given is that prices are too high. In other words, many households are deferring the purchase of a vehicle in the hopes of getting a better deal. Automobile demand would be a lot higher now if households thought that prices would keep rising, as this would incentivize them to buy a car before prices rose even more. Chart 17Households Think That This Is The Worst Time Ever To Buy A Car Chart 18Inflation Started Accelerating Quickly Only When Unemployment Reached Very Low Levels In The 1960s What should be acknowledged is that inflation expectations tend to be governed by complex social feedback loops, which makes the relationship between slack and inflation highly non-linear. The experience of the 1960s provides a pertinent example. The US unemployment rate reached NAIRU in 1962. However, it was not until 1966, when the unemployment rate was two percentage points below NAIRU, that inflation expectations became unhinged. Within the span of ten months, both wage growth and CPI inflation doubled, with the latter reaching 6% by the end of the decade (Chart 18). The lesson is clear: While long-term inflation expectations are well anchored today, there is no guarantee they will stay that way indefinitely. Is this a lesson that the Fed will heed? Like Larry Summers, we have our doubts, suggesting that the long-term risks to inflation are to the upside. Fighting The Last War Just as military generals are prone to fighting the last war, the same is true of economic policymakers. Central bankers have been staring down the barrel of the deflationary gun for over two decades. In the 1960s, policymakers prioritized high employment over low inflation. With memories of the Great Depression still fresh in their minds, they kept policy rates too low for too long. This time around, policymakers have an additional reason to drag their heels in raising rates: government debt is very high. Higher borrowing costs would force governments to shift spending from social programs to pay off bondholders. Needless to say, that would not be very popular with most voters. Reducing debt-to-GDP ratios via higher nominal income growth will prove to be more politically palatable than fiscal austerity. Investment Conclusions The path to high interest rates is lined with low interest rates. Structurally higher inflation will eventually lead to higher nominal interest rates, but not before an extended period of negative real rates. Chart 19Neither The Fed Nor The Markets Think The Neutral Rate Of Interest Is All That High Neither the Fed nor the markets think the neutral rate of interest is all that high (Chart 19). We think the neutral rate is higher than widely believed. However, this will not become apparent until the unemployment rate falls well below its full employment level. For now, the Fed’s leadership will want to avoid rocking the boat by turning more hawkish. While the US 10-year Treasury yield will trend higher over time, it will pause at around 1.8% in the first half of next year as the unwinding of pandemic-related bottlenecks leads to a “one step down” for inflation. The ECB and the Bank of Japan are even more reluctant to tighten monetary policy than the Fed. Some developed economy central banks like those of the UK, Norway, Sweden, Canada, and New Zealand are more inclined to normalize monetary conditions. That said, they too will be constrained by the fear that going it alone in raising rates will put undue upward pressure on their currencies. While we are not as bullish on stocks as we were at the start of the year, the combination of low interest rates and above-trend growth over the next 12 months will support equities. As we discussed in our recent strategy outlook, investors should favor cyclicals, value stocks, small caps, and non-US markets. Bitcoin Trade Update After being up as much as 50%, our short Bitcoin trade got stopped out for a loss. We remain bearish on Bitcoin and have decided to reinstate the trade.   Peter Berezin Chief Global Strategist pberezin@bcaresearch.com View Matrix Special Trade Recommendations Current MacroQuant Model Scores