Developed Countries

Executive Summary High profile economists Larry Summers and Olivier Blanchard have recently cast doubt on the Federal Reserve’s claim that a soft landing is possible for the US economy. We explore the arguments from both sides of the debate and conclude that the economic data will likely support the Fed’s soft landing thesis during the next six months. However, the unemployment rate will rise more significantly as we move deeper into 2023 and the Fed continues to run a restrictive monetary policy. This report also provides an update on our recommended portfolio duration and high-yield positioning, and suggests a tweak to our recommended positioning across the Treasury curve. Specifically, we advise clients to enter a duration-matched position long the 5/30 barbell and short the 10-year bullet. The Beveridge Curve Bottom Line: Investors should keep portfolio duration close to benchmark and maintain a neutral (3 out of 5) allocation to high-yield bonds. Investors should also exit positions long the 2-year bullet versus a duration-matched cash/5 barbell and enter a position long a 5/30 barbell versus the 10-year bullet. Feature This week’s report digs into a recent macro debate between two high profile economists – Larry Summers and Olivier Blanchard – and the Federal Reserve about whether a “soft landing” is possible for the US economy. We summarize the debate below and offer our own thoughts on its implications for investment strategy. But first, we provide a quick update on our recent thinking about US bond portfolio construction, including a change to our recommended yield curve positioning. Positioning Update Portfolio Duration In recent reports we have written that we would reduce our recommended portfolio duration stance from “at benchmark” to “below benchmark” if the 10-year Treasury yield falls to 2.5% or if core inflation converges to our 4%-5% estimate of its underlying trend (Chart 1).1 The 10-year yield came close to hitting our 2.5% trigger last week but then quickly reversed course. It moved even higher after Friday’s extremely strong employment report, and it now sits at 2.78%. We are sticking with our plan. Despite July’s blockbuster job gains, trends in both initial and continuing jobless claims suggest that the unemployment rate is more likely to rise than fall during the next few months (Chart 2). Supply chain indicators also point toward falling inflation (Chart 2, bottom panel). Against this backdrop, it wouldn’t be too surprising to see bond yields experience another downleg. Chart 1Stay Neutral For Now Chart 2Unemployment Has Bottomed High-Yield Turning to credit, we continue to recommend an underweight allocation to spread product (including investment grade corporate bonds) versus Treasuries, but with a slightly higher allocation (neutral) to high-yield. We think that high-yield spreads can tighten in the near-term as recession fears are allayed and inflation rolls over. However, the medium-to-long run macro environment is negative for spread product and we will be quick to reduce junk exposure if spreads reach their 2017-19 average (Chart 3) or if core inflation converges with our 4%-5% estimate of trend. Chart 3Tracking The Junk Rally Treasury Curve Chart 4Buy A 5/30 Flattener Finally, this week we tweak our recommended yield curve positioning by closing our prior recommendation: long 2-year bullet versus duration-matched cash/5 barbell, and by initiating a new trade: long 5/30 barbell versus a duration-matched 10-year bullet. We only initiated that 2 over cash/5 trade a couple weeks ago on the view that 2/5 Treasury curve inversions don’t tend to last very long.2 However, it has since become clear that our timing was premature. In fact, we probably shouldn’t anticipate a significant 2/5 steepening until the Fed’s tightening cycle is near its end, which we do not believe to be the case. Instead, we recommend that investors shift into a duration-matched position that is overweight a 5/30 barbell versus the 10-year bullet. This trade offers a positive yield differential of 16 bps (Chart 4) and will profit from a flattening of the 5-year/30-year Treasury slope. The 5/30 slope has steepened in recent weeks, but further steepening is only likely to occur near the end of a Fed tightening cycle. Given that we see significant further tightening ahead, it’s much more likely that the 5/30 slope will fall to zero or even turn negative (Chart 4, top panel). The Battle Of The Beveridge Curves Our battle begins with a speech from Fed Governor Christopher Waller that was given back in May.3 In that speech, Waller made the case for why the large number of job vacancies gave him “reason to hope that policy tightening in current circumstances can tame inflation without causing a sharp increase in unemployment.” Waller’s argument was based on the historical relationship between the job vacancy rate and the unemployment rate, a relationship known as the Beveridge Curve (Chart 5). In essence, Waller’s argument for a “soft landing” boils down to the observation that the Beveridge Curve shown in Chart 5 has shifted up since the pandemic. That is, since March 2020 we have consistently seen more job vacancies for any given unemployment rate. His contention is that, as economic activity slows, rather than moving to the right along the Beveridge Curve, the curve will shift down toward its pre-pandemic level. In other words, the job vacancy rate will decline significantly without a large uptick in the unemployment rate. Chart 5The Beveridge Curve Objection! In a paper published this month, Olivier Blanchard, Alex Domash and Larry Summers (BDS) take issue with Waller’s claims from two different angles, a theoretical one and an empirical one.4 First, from a theoretical perspective, BDS describe three factors that lead to either movements along the Beveridge Curve or shifts in the curve itself. 1) Economic Activity. Stronger economic activity leads to more job vacancies and a lower unemployment rate. In other words, a shift to the left along the Beveridge Curve, illustrated as the journey from point A to point B in Chart 6. Chart 6An Illustrated Beveridge Curve 2) Matching Efficiency. If available jobs are a worse match for the skills of the unemployed labor force, then it will lead to a higher job vacancy rate for any given unemployment rate. In other words, a shift up in the Beveridge Curve from point B to point C in Chart 6. 3) Reallocation Intensity. If people switch jobs more frequently, then there will also tend to be more vacancies for any given level of unemployment. Again, this would shift the Beveridge Curve up from point B to point C in Chart 6. Using a model and data from the JOLTS survey, BDS attempt to decompose how much of these three factors have contributed to the current positioning of the Beveridge Curve. The authors estimate that economic activity has increased significantly since the end of 2019, but also that the labor market’s matching efficiency has declined, and that reallocation intensity has increased (Chart 7). Chart 7An Illustrated Beveridge Curve   While monetary tightening can weaken economic activity, it cannot change the labor market’s matching efficiency or its reallocation intensity. Therefore, the authors argue, unless matching efficiency and reallocation intensity naturally revert to their pre-COVID levels, weaker economic activity will manifest as a movement to the right along the post-2020 Beveridge Curve, leading to a higher unemployment rate. This, in our view, is the crux of the “soft landing” debate. Are the recent changes in labor market matching efficiency and reallocation intensity temporary or permanent? Next, we move to BDS’ empirical arguments. The authors construct a time series of the job vacancy rate going back to the 1950s and then examine changes in both the job vacancy rate and the unemployment rate following cyclical peaks in the vacancy rate. Their results show that a falling job vacancy rate almost always coincides with a rising unemployment rate (Table 1). In other words, if history is any guide, it is very unlikely that the Fed will be able to push the job vacancy rate down without seeing an increase in unemployment. Table 1Average Change In The Unemployment Rate And The Vacancy Rate After A Peak In The Vacancy Rate That said, the authors’ results also reveal a dynamic known as the Beveridge Loop. Notice in Table 1 that a drop in the vacancy rate leads to a much smaller increase in the unemployment rate during the first six months following the vacancy rate peak than it does during the first 12 months or first 24 months. In other words, there is some empirical validity to Fed Governor Waller’s argument that the early impact of Fed tightening will be felt primarily through a falling job vacancy rate. The 2018/19 Example We can illustrate the Beveridge Loop with a recent example, one that interestingly was not included in BDS’ empirical analysis. The job vacancy rate peaked in November 2018 and then trended lower until the pandemic struck in early 2020. Interestingly, this 2018-19 drop in the job vacancy rate occurred alongside a modest decline in the unemployment rate. Chart 8 shows what the Beveridge Curve looked like during this period. Notice that, rather than moving back to its January 2018 point in a straight line, the Beveridge Curve formed a loop after peaking in November 2018. Chart 8The 2018/19 Beveridge Loop What allowed the labor market to achieve this “soft landing” in 2018/19? The most likely answer is that labor force participation rose significantly during this period (Chart 9). The influx of workers into the labor force allowed the unemployment rate to keep falling even as continuing unemployment claims bottomed out. Chart 9The 2018/19 Soft Landing The BCA Verdict Our view is that the incoming economic data will appear to validate the Fed’s “soft landing” view during the next six months, but that the unemployment rate will start to rise more significantly as we move deeper into 2023. As we have stated in prior reports, a significant increase in the unemployment rate will eventually be required to tame inflation, but that increase likely won’t occur as soon as many market participants expect.5 In essence, we anticipate a large Beveridge Loop. A loop that, in fact, appears to already be forming (Chart 5). We have shown that the empirical evidence supports the idea that a Beveridge Loop will occur during the early stages of a slowdown. Further, theory and empirical evidence demonstrate that the Beveridge Curve is convex. This suggests that the Beveridge Loop could be particularly large in this cycle given that the vacancy rate is starting from such a high level. Perhaps the bigger question, though, is whether the Beveridge Curve will re-converge with its pre-pandemic level during the next 6-12 months. On this question we side more with Blanchard, Domas and Summers. While we think that matching efficiency can continue to improve along its current trend (Chart 7, panel 2), the widespread adoption of work-from-home suggests that the labor market has probably experienced a permanent increase in reallocation intensity. On matching efficiency, the best evidence for continued improvement comes from a breakdown of employment by industry (Table 2). Notice that the three sectors (other than government) that have experienced the greatest job losses since the pandemic – Health Care, Leisure & Hospitality and Other Services – also have three of the highest job openings rates. This suggests that there shouldn’t be a permanent friction between matching those missing workers to available jobs. Table 2Employment By Industry Finally, working from our 2018/19 example, we can assess the likelihood that an increase in labor force participation will cushion the upside in the unemployment rate. Here, we see some potential for the prime age participation rate to rise back to its pre-COVID level, but the re-entry of recently retired workers over the age of 55 is more in doubt. Overall, it’s highly unlikely that the overall participation rate will re-gain its pre-pandemic level (Chart 10). Chart 10Labor Force Participation The bottom line is that the next six months will likely look more like a soft landing than a hard one. The job vacancy rate will fall quickly and the unemployment rate will stay relatively low, causing the Beveridge Curve to form a large loop. However, the Beveridge Curve will not revert to its pre-COVID level any time soon. As we move deeper into 2023, the Beveridge Curve will stop looping and the unemployment rate will rise significantly.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Weekly Report, “Recession Now Or Recession Later?”, dated July 26, 2022. 2 Please see US Bond Strategy / Global Fixed Income Strategy Weekly Report, “A Low Conviction US Bond Market”, dated July 12, 2022. 3https://www.federalreserve.gov/newsevents/speech/files/waller20220530a.pdf 4https://www.piie.com/publications/policy-briefs/bad-news-fed-beveridge-space#:~:text=The%20Federal%20Reserve%20seeks%20to,together%20and%20remain%20unlikely%20now. 5 Please see US Bond Strategy Weekly Report, “Three Conjectures About The US Economy”, dated July 19, 2022. Recommended Portfolio Specification Other Recommendations Treasury Index Returns Spread Product Returns

S&P 500 Chart 1Macroeconomic Backdrop Chart 2Profitability Chart 3Valuations And Technicals Chart 4Uses Of Cash Cyclicals Vs Defensives Chart 5Macroeconomic Backdrop Chart 6Profitability Chart 7Valuation And Technicals Chart 8Uses Of Cash Growth Vs Value  Chart 9Macroeconomic Backdrop Chart 10Profitability Chart 11Valuations And Technicals Chart 12Uses Of Cash Small Vs Large Chart 13Macroeconomic Backdrop Chart 14Profitability Chart 15Valuations and Technicals Chart 16Uses Of Cash Table 1Performance Table 2Valuations And Forward Earnings Growth Recommended Allocation  

Executive Summary The constructive economic view that has us at odds with the consensus rests on three premises: excess pandemic savings will allow consumption to grow at trend, despite inflation; inflation will soon peak, moving to around 4% by year end; and inflation expectations will remain well anchored, keeping the Fed from moving immediately to stifle the economy. Our consumption thesis remains intact. Real consumption has kept pace despite falling real incomes, thanks to a steady, modest drawdown of excess savings. Though our calls for an inflation peak have been consistently premature, recent data suggest that inflation pressures are abating. Gasoline prices have been falling for seven weeks; the fever has broken in ISM survey price measures; and the labor market, notwithstanding July's potent employment report, is becoming less tight. Longer-run inflation expectations have resisted becoming unmoored despite soaring measured inflation and a breakout does not appear to be imminent. A Mighty Savings Cushion Bottom Line: We continue to expect the economy will be surprisingly resilient, allowing equities to rally further before the Fed squashes the expansion. We doubt the rally will persist very far into 2023, however, so we are reducing equities to equal weight over a twelve-month timeframe. Feature We will be holding our quarterly webcast next Monday, August 15th at 9:00 a.m. Eastern time in lieu of publishing a Weekly Report. Please join us with your questions to make it a fully interactive event. We will resume our regular publication schedule on the 22nd. Last week, an investor we were meeting for the first time asked us how anyone could have published on a weekly basis this year. “Things are so uncertain and they’re moving so fast, how do you keep up? What have you been writing about?” At long last, we felt seen. Feeding the weekly beast is not easy under the best of circumstances and investors know that this year has been far from ideal. Related Report  US Investment StrategyThe High Bar For Getting Worse Once the warm glow of unexpected empathy receded, we replied that we’ve been doing our best to anticipate how the key macro issues will impact financial markets over our cyclical 3-to-12-month timeframe, paying particular attention to consumers, inflation and the Fed. The outlook for consumption has been our primary focus from a growth perspective; we’ve been trying to assess how representative the key drivers of inflation are and how persistent they’ll be; and we’ve continuously monitored longer-run inflation expectations to determine if inflation has gotten far enough into economic agents’ heads to become self-reinforcing and compel the Fed to dislodge it, no matter the near-term economic cost. We review what we see on all three fronts in today’s report, and how events are unfolding relative to our expectations. The direction remains especially uncertain, but our theses remain intact, and we are sticking with our constructive outlook on risk assets and the economy for the rest of the year. We are pulling in our horns on our twelve-month optimism, however, in line with the BCA house view and the dawning realization that twelve months of equity outperformance is overly ambitious. We continue to believe the recession will arrive too late for the gloomy consensus of investors judged by their quarterly performance, forcing them back into risk assets, but the rebound may not persist beyond the FOMC’s first 2023 meeting at the beginning of February. The Consumer’s Staying Power Since CARES Act transfer payments began driving a surge in personal savings, we have viewed them as dry powder to support consumption once households regained the freedom to spend as they see fit. When the payments stopped flowing and the pandemic continued to delay a return to normal, that view came under some fire. We are of the mind that households merely deferred much of the services demand they would otherwise have slaked in 2020 and 2021; others argue that consumption deferred is consumption destroyed, as households will be reluctant to spend windfall transfers that they’d mentally sorted as savings. While it will take a while for data to confirm either thesis, we are encouraged by what we’ve seen so far. The savings rate has declined considerably so far in 2022, supporting the view that households would be willing to reach into their savings to maintain trend consumption (Chart 1). It dipped to 5.2% in the second quarter from 5.6% in the first quarter, well below February 2020’s 8.3% pre-pandemic level and 2011 to 2019’s 7.4% quarterly mean (Chart 2). Based on the series’ stability over the previous nine years, 2020’s and 2021’s forced savings rates amounted to 11- and 6-sigma post-crisis events and this year’s approximately -2.5-sigma drawdown suggests the pendulum has further to swing in the direction of dissaving. We disagree with knee-jerk conclusions that spending in excess of income is unsustainable – it’s plenty sustainable for households who socked away a mountain of savings over the previous eight quarters while bars, restaurants, stadiums, concert venues and resorts were idled. Chart 1Right On Target Chart 22020 And 2021 Savings Were Enormous The estimates of excess savings that we’ve been calculating every month since the summer of 2020 peaked just above $2.3 trillion last August and remained around that level before embarking on a steady decline in the first half to reach our current estimate above $2 trillion (Chart 3, bottom panel). Quoting that figure has been nagging at us lately, however, as one of the two assumptions we used to calculate households’ no-pandemic savings baseline – annualized disposable income growth of 4% – took 2% annual inflation as given, a condition that no longer applies after a twelve-month stretch in which year-over-year CPI inflation has averaged 7.1%. Chart 3Nominal Excess Savings To determine how much households' purchasing power has eroded, we deflated our monthly excess savings estimates to a level equating to 2% annualized inflation (Chart 4, top panel). The adjustment knocked $450 billion off our current estimate, trimming it to $1.6 trillion (Chart 4, bottom panel). Perhaps more importantly for the outlook, our adjustment doubled the year-to-date burn rate to $500 billion. We have always worked with the (deliberately conservative) assumption that households would spend half of their excess savings; if inflation doesn’t decelerate soon, their cushion may not last very far beyond the end of the year. Chart 4Adjusted Excess Savings Bottom Line: Households have been willing to dip into savings to maintain trend consumption so far this year, in line with our hypothesis. We expect they will continue to do so, and the savings rate will remain around 5% or fall even lower, but inflation has eaten up some of their dry powder. Will Inflation Ever Peak? Shredding widely shared expectations that inflation would peak sometime in the first half, the year-over-year increase in headline CPI has kept climbing, all the way to 9% in June. July should finally provide some relief, as the average national retail gasoline price has fallen for seven consecutive weeks and ended July 13% below its June 30 level (Chart 5). Last week’s ISM manufacturing and services PMIs also suggested that inflation has begun to ease its grip somewhat, with the manufacturing input prices series plunging by nearly 20 points to its two-decade mean (Chart 6, top panel) and the services prices component cooling by 8 points, though it remains quite high (Chart 6, bottom panel). Chart 5Four Bucks A Gallon Is High, But Not Unfamiliar Chart 6The Fever May Have Broken ...​​​​​​ Chart 7... Though The Job Market Is Still Quite Hot​​​​​ The tight-as-a-drum labor market has been a fertile source of inflation worries, but there are signs that it is becoming less tight. Job openings remain 40% above their pre-COVID high but declined by 600,000 in June and are 10% off of March’s all-time peak (Chart 7). Elevated quits reveal that it's still easy to get a job, but the net share of small businesses in the NFIB survey planning to hire in the next three months is down 40% from its peak last summer (Chart 8). The July employment report challenged the under-the-radar indicators’ implication that the labor market is cooling, as net payroll expansion reaccelerated along with average hourly earnings growth (Chart 9). We are confident that net payroll growth will slow but compensation clearly has the cyclical wind at its back, and it is not certain that labor’s structural headwind will largely offset it, as per our thesis.   Chart 8Hiring Intentions Are Back To More Normal Levels ... Chart 9... But Wage Growth Remains Elevated Inflation Expectations Longer-run inflation expectations are a critical piece of the puzzle because they are the pathway for rising inflation to become self-reinforcing. If they expect persistently higher inflation, workers will negotiate more fiercely for larger compensation increases to stay ahead of it; businesses will push more vigorously to pass on their increased costs to preserve profit margins; lenders and bond investors will demand higher interest rates to protect their real returns; and consumers will seek to buy more now to get the most from their dwindling purchasing power, exacerbating supply-demand imbalances and keeping the heat on near-term inflation readings. We are therefore closely watching inflation expectations. Market-based measures like TIPS break-evens and CPI swaps shed some light on investor and business expectations, while the monthly University of Michigan consumer sentiment survey offers insight into households’ views. Market-based measures remain well-anchored: intermediate-term expectations as implied by TIPS break-evens are just nosing above the top of the Fed’s preferred 2.3-2.5% range (Chart 10, middle panel) while long-term expectations remain below it, as they have for most of the year (Chart 10, bottom panel). Intermediate- and long-term expectations derived from CPI swaps remain 20 to 30 basis points higher but are in the same position relative to their year-to-date path (Chart 11, bottom two panels). Chart 10Market-Based Inflation Expectations ...​​​​​​ Chart 11... Are Not Problematic​​​​​​ Chart 12Just Say No (To Bottleneck Prices) The Michigan survey doesn’t betray any pressing long-run concerns. The preliminary 3.3% June reading hinting at a breakout turned out to be a false alarm, as June’s final figure was 3.1% and July’s was 2.9%. Survey respondents continue to shun big-ticket purchases because they expect prices will fall from their current levels (Chart 12). 2-year TIPS and swaps price in an optimistic near-term outlook that is likely to be disappointed, as we think inflation will prove to be sticky around the 4% level, and that disappointment could bleed into higher longer-run expectations. While expectations are not problematic now, investors will need to watch them carefully going forward. Investment Implications It was policy, monetary and fiscal, that inspired our bullish turn in 2020 once we digested the COVID shock. We thought the macro backdrop would come down to policymakers versus the virus and our money was on the former. We remained bullish across 2021 on the idea that monetary and fiscal support would remain in place well after they ceased to be necessary. Mindful that there is no such thing as a free lunch, we expected that the emergency pandemic measures would ultimately have the effect of overstimulating demand, but we entered 2022 thinking that equities and credit would enjoy one more year of sizable excess returns over Treasuries and cash before the overstimulation manifested itself. Overweighting (underweighting) equities in a multi-asset portfolio is our default position when monetary policy is easy (tight), though we will override that default when appropriate. We have no appetite for overriding it once it becomes clear that market expectations for 2023 rate cuts are going to be disappointed and tight policy is just around the bend. Given our view that inflation will linger around 4% after easing smartly over the rest of this year, we expect that the Fed will impose restrictive monetary policy settings by the second half of 2023 in its quest to drive inflation back down to its 2% target. Markets’ overly rosy Fed expectations look sure to be disappointed and they could face a reckoning after the FOMC’s January 31-February 1 meeting. Chart 13Consolidation Now, 10%+ By The End Of The Year That meeting could herald an inflection for risk assets’ relative performance and we are therefore joining our colleagues in adopting a neutral 12-month view on equities. We continue to differ from the BCA consensus, however, in expecting a meaningful equity rally before year end. While we expect technical resistance at 4,200 will restrain the S&P 500 in the immediate term (Chart 13), we think it will find its way back into the mid-to-high 4,000s before the Fed signals that it will take the funds rate to 4% or above, dashing hopes for a February peak around 3.5%. We still want to overweight equities in multi-asset portfolios, but only until year-end or 4,500 to 4,600, whichever comes first.   Doug Peta, CFA Chief US Investment Strategist dougp@bcaresearch.com