Developed Countries

Special Report

Generally-speaking, an increase in bank deposits occurs due to either Fed asset purchases, bank asset purchases, or bank loan creation. Deposits have grown massively over the past year because the Treasury has issued an enormous amount of bonds, and these bonds have been purchased both by the Fed and US banks. Relative to the 2008-2009 period, the comparatively better health of US bank balance sheets last year has been an even more important factor than Fed asset purchases in accounting for the difference in money growth between the two periods. Money growth used to be a good predictor of economic activity, but today it makes more sense to focus on interest rates rather than monetary aggregates as a leading economic indicator. Over the past 20 years, only the collapse in velocity that occurred after 2008 is meaningful for investors, and it appears to reflect already “known” information: the persistent household deleveraging that occurred following the global financial crisis, and the effect of Fed asset purchases on the stock of money at several points over the past decade. Our base case view is that a portion of the significant amount of household savings that have accumulated will not be spent, and that the US output gap will close but not move deeply into positive territory this year. But the enormous growth in money over the past year reflects unprecedented fiscal and monetary support, which could eventually change investor expectations about long-term interest rates (even absent rapid overheating). Rising long-term rate expectations could threaten the equity bull market, given the impact the secular stagnation narrative has had in keeping long-term rate expectations low and the extent to which easy money has boosted equity valuation multiples over the past year. Broad money growth has exploded higher over the past year, to a pace that has not been seen since WWII (Chart II-1). This growth in the money supply has vastly exceeded what investors witnessed during and immediately following the global financial crisis of 2008-2009, raising concerns among many investors of the potential cyclical and structural consequences. Chart II-1A Nearly Unprecedented Surge In Money Growth In this report we revisit the deposit creation process, and explain the specific factors that have led to surging money growth over the past year. We also review the usefulness of money growth as an economic indicator, and provide some perspective on the 20-year decline in money velocity. We conclude by noting that the surge in money growth is potentially concerning for investors for two reasons. First, if US households ignore likely future tax increases and decide to fully spend the vast amount of savings that have accumulated over the past year, then the US economy is likely to overheat rather quickly. The second, more likely, threat to investors is if the sharp increase in the money supply ends up changing market expectations about the neutral rate of interest. It remains too early to conclude whether investors will significantly revise up their long-term rate expectations, in large part because the scale of permanent damage in the wake of the pandemic is still unknown. But investors should remain vigilant, given the impact the secular stagnation narrative has had on keeping long-term rate expectations low and the extent to which easy money has boosted equity valuation multiples over the past year. Reviewing The Money And Bank Deposit Creation Process In order to fully understand the spectacular growth of the money supply over the past year and its potential implications for the economy and financial markets, it is important to revisit how money is created in a modern economy. My colleague Ryan Swift, BCA’s US Bond Strategist, reviewed this question in detail in a June 2020 Strategy Report and we summarize the report’s key points below.1 In the US, most of the stock of broad money aggregates is composed of bank deposits. Following the global financial crisis, the textbook view of how banks act purely as intermediaries, taking in deposits from the public and lending them out, was revealed to be a mostly inaccurate description of the financial system in the aggregate. Rather, while individual banks often compete for deposits as a source of funding, bank deposits in the aggregate are typically created by making loans. Central banks can also create money, by purchasing financial assets and crediting the banking system with reserve assets (central bank money). Table II-1 highlights the link between the Fed’s balance sheet and that of US banks in the aggregate, and highlights how changes in deposits – a liability of the banking system – must be offset by increases in bank assets or decreases in other bank liabilities. Table II-1The Link Between The Fed’s Balance Sheet And The Aggregate US Banking System The typical mechanics of three money-creating operations are described below, alongside the corresponding change in balance sheet items: Fed Asset Purchases: When the Federal Reserve purchases financial assets in the secondary market, it increases securities held (Fed asset) and typically increases reserves (Fed liability). In the increase in reserves (banking system asset) matches the increase in deposits (banking system liability), as the previous holders of the assets purchased by the Fed deposit the proceeds of the sale. Bank Asset Purchases: When banks purchase government securities from non-bank holders they credit the sellers with bank deposits.2 This increases bank holdings of securities or other assets (banking system asset) and increases deposits (banking system liability). Bank Loan Creation: When banks create a loan, they increase their holdings of loans & leases (banking system asset) and deposit the loan amount into the borrowers’ account (banking system liability). At the individual bank level, if Bank A creates a loan and the borrower withdraws the funds to pay someone with an account at Bank B, there will be an asset-liability mismatch relating to that loan transaction between those two banks if no other actions are taken. The result will be that Bank A experiences an increase in equity capital and Bank B experiences a decline. But for the banking system as a whole, the increase in bank assets exactly matched the increase in bank liabilities, and Bank A created the deposits that ended up as a liability of Bank B. The issuance or retirement of long-term bank debt and equity instruments can also create or destroy deposits but, for the purpose of understanding the difference in money growth during the pandemic compared with the 2008-2009 experience, it is sufficient to focus on the three money-creating operations described above. Explaining The Recent Surge In Money Growth The prevalent view among many financial market participants is that the money supply has surged in the US due to the fiscal stimulus provided by the CARES act. But an increase in the government’s budget deficit does not in and of itself create money, because the Treasury issues bonds to finance the difference between revenue and expenditures. If those bonds are purchased entirely by the nonfinancial sector, then an increase in deposits of stimulus recipients is offset by a decrease in deposits of those who purchased the bonds. A more precise answer is that deposits have grown massively over the past year because the Treasury has issued an enormous amount of bonds and these bonds have been purchased both by the Fed and US banks. Charts II-2A and II-2B highlight this by showing the change in the main items on the aggregate banking system balance sheet since the end of 2019. The charts show that the increase in deposits on the liability side of bank balance sheets have been matched by large increases in reserves and other cash (caused by the Fed’s asset purchases) and banks’ securities holdings (caused by bank asset purchases). Chart II-2AOver The Past Year, Fed And Bank Asset Purchases… Chart II-2B…Account For Most Of The Surge In Deposits   But this does not explain why money growth has been so much larger over the past year than it was in 2008-2009, when total Federal Reserve assets increased from $920 billion to $2.2 trillion. Chart II-1 on page 15 highlighted that growth in M2 has risen to a whopping 25% year-over-year growth rate, a full 15 percentage points above the strongest rate that prevailed following the global financial crisis. Charts II-3A and II-3B explain the discrepancy, by showing the change in the main items on the aggregate banking system balance sheet as a percent of the money supply during each of the two periods, as well as the difference. The charts show that while changes in bank reserves and cash assets – caused by Fed asset purchases – were significantly larger in 2020 than they were on average from 2008 to 2009, changes in loans & leases and securities in bank credit, as well as other assets were also quite significant and account for two-thirds of the difference when added together. Chart II-3ARelative To 2008/2009, The Health Of The Banking System… Chart II-3B…Helped Facilitate More Money Creation Last Year   Thus, while it is true that the Fed’s accommodation of extraordinary fiscal easing has helped create a sizeable amount of money over the past year, relative to the 2008-2009 period the comparatively better health of US bank balance sheets has been an even more important factor – in the sense that balance sheet restrictions did not prevent US banks from facilitating the creation of money as appears to have been the case in the aftermath of the global financial crisis. Money And Growth We noted above that fiscal easing does not create money in and of itself unless the bonds issued to finance an increase in the deficit are purchased either by banks or the Fed. Yet most investors would not disagree that significant increases in budget deficits boost short-term economic growth, particularly during recessions. This implies that the link between money and economic growth may not be particularly strong over a cyclical time horizon, which is in fact what the data shows – at least over the past 30 years. Charts II-4A and II-4B illustrate the historical relationship between real GDP and real M2 growth, pre- and post-1990. The chart makes it clear that the relationship between real money and GDP growth used to be strong, with real money growth somewhat leading economic activity. This relationship completely broke down after the 1980s, and is now mostly coincident and negative. There are three reasons behind the breakdown: 1. The money supply used to be the Federal Reserve’s monetary policy target, meaning that money growth directly reflected monetary policy shifts. Today, the Fed targets interest rates, and the portion of money created through loans simply mirrors the change in interest rates as loan demand rises (falls) and interest rates fall (rise). Specifically, Chart II-4A shows that the ability of money growth to lead economic activity seems to have ended in the late 1980s, when the Fed stopped providing targets for monetary aggregates. Chart II-4AMoney Growth Used To Predict Economic Activity… Chart II-4B…But Ceased To Do So Once The Fed Stopped Targeting The Money Supply Chart II-5US Banks Provide Meaningfully Less Private Sector Credit Than In The Past 2. The share of total US credit provided by US banks has fallen significantly over time – especially during the early 1990s – as corporate bond issuance, securitized loans, and mortgages backed by agency bonds issued to the private sector rose as a proportion of total credit (Chart II-5). 3. Since 2000, a Chart II-4B shows that a clearly negative correlation has emerged between money growth and economic activity during recessions. In 2008-2009 and again last year, money growth reflected emergency Fed asset purchases in the face of a sharp decline in economic activity. In 2000, the Fed did not expand its balance sheet, but the economy diverged from money growth due to the lingering impact of management excesses, governance failures, and elevated debt in the corporate sector in the 1990s. The conclusion for investors is straightforward: while money growth used to be a good predictor of economic activity, today it makes more sense to use interest rates than monetary aggregates as a leading indicator for growth. Money Velocity And Its Implications When discussing the impact of money on the economy, one point often raised by investors is the fact that money velocity has declined significantly over the past two decades. This observation is frequently followed by the question of whether the absence of this decline would have caused real growth, inflation, or both to have been higher over the past 20 years than they otherwise were. It is difficult to prove or refute the point, as monetary velocity is not a well-understood concept – investors do not have a good, reliable theory upon which to predict changes in velocity or understand their economic significance. Velocity is calculated from the equation of exchange as a ratio of nominal GDP to some measure of the money supply (typically a broad measure such as M2) and theoretically represents the turnover rate of money. But long-term changes in velocity do not seem to correlate well with measures of growth or inflation. Short-term changes in velocity correlate extremely well with inflation, but this simply reflects the fact that velocity tends to be driven by the numerator (nominal GDP) over short periods of time (see Box II-1). BOX II-1 Money Velocity Over The Short-Term Some investors have pointed to the relationship shown in Chart II-B1 to argue that M2 money velocity is a significant cyclical predictor of inflation. But Chart II-B2 illustrates that nearly two-thirds of annual changes in velocity since 1990 have been accounted for by changes in the numerator – nominal GDP – rather than the denominator. This underscores that the apparent explanatory power of short-term changes in money velocity at predicting inflation is simply capturing the normal relationship between real growth and inflation, as well as the naturally positive correlation between the implicit GDP price deflator and core consumer prices. Chart Box II-1Velocity Seemingly Predicts Inflation Over The Short-Term… Chart Box II-2…Because Short-Term Changes In Velocity Are Driven By Nominal Output   The bigger question is why velocity has declined so significantly over the past 20 years, and what this means for investors. Chart II-6Large Declines In Velocity Are Linked To Prolonged Periods Of Deleveraging Panel 1 of Chart II-6 shows a long-dated history of M2 velocity, and highlights that the average or “normal” level of M2 velocity has historically been just under 1.8. Over the past century, there have been just four major deviations from this level: A major decline that began at the start of the Great Depression and prevailed until the Second World War (WWII) Significant volatility during and in the years immediately following WWII A sharp rise in velocity during the 1990s to a record level A downtrend beginning in the late 1990s that remains intact today Abstracting from the war period in which the economy was heavily distorted by government intervention, Chart II-6 also highlights that persistent declines in velocity appear to be explainable by major deleveraging events. The second panel of the chart shows a measure of the duration of private sector deleveraging, and highlights that the two periods of low velocity have been strongly (negatively) correlated with the prevalence of deleveraging. This explanation is simple but intuitive: excessive leveraging eventually causes households and firms to redirect a larger portion of their income to servicing or paying down debt, which weighs on real growth and, by extension, prices. While it is true that the recent 20-year downtrend in velocity began in the late 1990s and thus well before household deleveraging began in 2008, this seems to mostly reflect the reversal of an anomalous rise in velocity in the late 1990s. We largely view the decline in velocity from the late 1990s to 2008 as a “reversion to the mean.” It remains an option question why velocity rose so sharply in the 1990s. Some evidence seems to point to financial innovation and technological change: Chart II-7 highlights that the number of automated bank teller and point-of-sale payment terminals rose massively in the 1990s, alongside a significant acceleration in real cash in circulation. This is theoretically consistent with an increased “turnover” rate of money. But Chart II-8 highlights that a substantial portion of the rise in velocity during this period was attributable to denominator effects (persistently weak money growth), rather than numerator effects. Chart II-7Some Evidence Of Increased Money Turnover In The 1990s Chart II-8The Rise In Velocity In The 1990s Was Driven By Slow Money Growth   Regardless of the cause, velocity was clearly anomalous on the upside in the 1990s, suggesting that it is not the downtrend in velocity over the past 20 years that is significant to investors. Rather, it is the collapse in velocity that has occurred since 2008 that is meaningful, and from the perspective of investors it appears to reflect already “known” information: the persistent household deleveraging that occurred following the global financial crisis, and the effect of Fed asset purchases on the stock of money at several points over the past decade. In the future, any meaningful increases in velocity are only likely to occur due to a significant reduction in the size of the Fed’s balance sheet, which is two to three years away at the earliest. The Fed could decide to taper its asset purchases sometime later this year or in early 2022, but tapering would merely slow the pace at which the Fed’s assets are increasing (and would thus not cause velocity to rise via a meaningful slowdown in money growth). Money And Future Inflation The final question to address is the issue of whether the enormous rise in money growth over the past year is likely to lead to higher, potentially much higher, inflation over the coming 6-12 months. This has been the main question from investors who have been unnerved by the surge in money growth and the collapse in the US government budget balance. Any link between money and inflation has to come through spending, so the question of whether a surge in money will lead to higher inflation is akin to asking whether the massive amount of savings that have been accumulated over the past year are likely to be spent, and over what period. We discussed this question in Section 1 of this month’s report, and noted that expectations of future tax increases and a permanent decline in some services spending will likely prevent all of these savings from being deployed once the practice of social distancing durably ends later this year. This implies that a substantial closure of the output gap is likely to occur in the second half of the year, but that major economic overheating will be avoided. Moreover, even if the output gap does rise into positive territory over the coming 6-12 months, this does not necessarily suggest that inflation will rise quickly back above the Fed’s target. In last month’s Special Report, we highlighted two important points about inflation that are often overlooked by investors. First, inflation’s long-term trend is determined by inflation expectations. Second, 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. While market-based expectations of long-term inflation have risen well above the Fed’s target, both our adaptive expectations model for inflation as well as a simple five-year moving average are between 30-60 basis points below the 2% mark (Chart II-9). This may suggest that a persistent period of output above potential may be required in order to raise inflation relative to expectations and to raise expectations themselves above the Fed’s target unless the Fed’s efforts at “jawboning” them higher prove to be highly successful. Measured as a year-over-year growth rate of core prices, inflation is set to spike higher in April and May in the order of 50-60 basis points simply due to base effects (Chart II-10). However, inflation will only sustainably rise to an above-target rate over the coming 6-12 months if demand is even stronger than implied by consensus expectations, which is not our base case view. Chart II-9Adaptive Inflation Expectations Measures Are Still Well Below The Fed's Target Chart II-10The Fed Will Look Through Base Effects On Consumer Prices   Investment Conclusions Investors can draw two conclusions from our analysis above. First, there is reason to be concerned about the enormous rise in the money supply if we are wrong in our assessment that some portion of the savings accumulated over the past year will not ultimately be spent. If US households ignore likely future tax increases and decide to fully spend their savings windfall, then the US economy is likely to overheat rather quickly. The second, more likely, threat to investors is if the sharp increase in the money supply, reflecting monetized fiscal stimulus and a meaningfully healthier financial system compared with the global financial crisis, ends up changing market expectations about the neutral rate of interest. Chart II-11The Pandemic Response May Raise Long-Term Rate Expectations Chart II-11 that while 5-year/5-year forward Treasury yields are not much lower than they were pre-pandemic, that is an artificially low bar. Long-dated bond yields fell over 100 basis points in 2018 and 2019, in response to a global growth slowdown precipitated by the Trump administration’s trade war. While President Biden will pursue some protectionist policies, they are likely to be meaningfully less damaging to global growth than under President Trump and are extremely unlikely to act as the primary driver of macroeconomic activity over the course of Biden’s term (as they were during the period that long-dated bond yields fell). As such, if the pandemic ends this year with seemingly minimal lasting damage to the US economy, long-dated bond yields could re-approach their late 2018 levels or higher towards the end of the year. This would cause a meaningful rise in 10-year Treasury yields, even with the Fed on hold until the middle of 2022 or later. A significant rise in bond yields would be quite unwelcome to stock investors given how stretched equity multiples have become. Table II-2 presents a set of year-end scenarios to gauge the potential impact of an eventual rise in 10-year yields. We assume that forward earnings grow at 5% this year, and we allow the spread between the 12-month forward earnings yield and the 10-year yield (a proxy for the equity risk premium) to return to its 2003-2007 average as part of an assumed “normalization” trade. Table II-2Current Multiples Are Justified Only If The 10-Year Treasury Yield Does Not Rise Above 2.5% The table suggests that a 10-year Treasury yield of 2.5% will be the most that the interest rates could rise before the fair value of the S&P 500 falls below current levels. That roughly equates to a return to the late-2018 levels that prevailed for 5-year/5-year forward Treasury yields, given that the short-end of the curve will remain pinned close to the zero lower bound for some time. For now, it remains too early to conclude whether investors will significantly revise their long-term rate expectations, in large part because the scale of permanent damage in the wake of the pandemic is still unknown. But investors should remain vigilant and attentive to the fact that interest rates may pose a threat to financial markets later this year even in a scenario where the US economy is not immediately overheating, given the impact the secular stagnation narrative has had in keeping long-term rate expectations low and the extent to which easy money has boosted equity valuation multiples over the past year. Jonathan LaBerge, CFA Vice President The Bank Credit Analyst Footnotes 1 Please see USBS Strategy Report "The Case Against The Money Supply," dated June 30, 2020, available at usbs.bcaresearch.com 2 Please see “Money creation in the modern economy,” Bank of England, Q1 2014 Quarterly Bulletin.

Highlights The general public and the financial markets have unrealistic hopes for what a Covid vaccine can achieve. The Covid vaccine will not be like the ‘one and done’ measles vaccine, but more like the ‘frequent jab’ seasonal flu vaccine. Underweight the so-called ‘economy reopening’ plays, particularly basic resources. The so-called ‘pandemic plays’ still have a lot of structural upside. As such, long-term investors should avoid value, and stick with growth. Stay structurally overweight healthcare, as governments will realise that the smart strategy is to add capacity to their healthcare systems. Stay structurally overweight healthcare-heavy DM versus resource-heavy EM. Fractal trade: Short KRW/USD. Feature Chart of the WeekThe UK Healthcare System Reached Its Limit At 60 Thousand Daily Covid Infections The general public and financial markets have high hopes that a vaccine can banish Covid forever, just like the vaccine for measles gives us lifelong immunity against the disease. Unfortunately, these hopes are unrealistic (Chart I-2). Chart I-2Reopening' Plays Are Due A Reversal The latest evidence suggests that the Covid vaccine will not be like the ‘one and done’ measles vaccine. It will be more like the seasonal flu vaccine, requiring regular refreshing. As if to prove this point, Moderna announced this week that it will trial a new Covid vaccine to tackle variants of the virus. Vaccines Against RNA Viruses Are Not Highly Effective Measles, the flu, and Covid are all diseases caused by ‘RNA viruses’ in contrast to, say, smallpox which is caused by a ‘DNA virus.’ One of defining characteristics of RNA versus DNA is its inferior proofreading ability during replication. As a result, RNA viruses have very high mutation rates. This means there are very few effective vaccinations against them. As the Journal of Immunology Research puts it:1 “No vaccine or specific treatment is available for many of these RNA viruses and some of the available vaccines and treatments are not highly effective.” Measles is an exception. While the flu virus mutates constantly and requires a yearly jab, a two-dose vaccine against measles during childhood confers lifelong immunity. This is because the measles virus is ‘antigenically monotypic.’ In plain English, the proteins that the measles virus uses to infect a cell cannot mutate even slightly without breaking. Meaning that any mutation to the virus destroys its ability to infect. Hence, the measles virus cannot evade the ‘one and done’ childhood vaccine. Is the Covid virus antigenically monotypic? No, we now know that the Covid virus can mutate and still infect, because we have already seen three region-specific mutations – the Brazilian variant, the UK variant, and the South African variant. The next question is, can any of these mutations evade the current vaccinations? Unfortunately, the answer is yes. As researchers at the South Africa National Institute for Communicable Diseases point out:2 “SARS-CoV-2 501Y.V2 (the South African variant), a novel lineage of the coronavirus causing COVID-19, contains multiple mutations within two immunodominant domains of the spike protein… This lineage exhibits complete escape from three classes of therapeutically relevant monoclonal antibodies… and may foreshadow reduced efficacy of current spike-based vaccines.” (Figure I-1) Figure 1Vaccines Against RNA Viruses Are Not Highly EffectiveBut if the current spike-based vaccines have only limited efficacy, why aren’t the vaccine manufacturers and policymakers warning us? The answer is that they are, but in coded form. Moderna’s decision to trial a new vaccine is a coded warning that its current vaccine might have only limited efficacy against emerging variants. Meanwhile, the UK health minister, Matt Hancock recently warned: “If we vaccinated the population, and then you got in a new variant that evaded the vaccine, then we’d be back to square one.” These warnings are coded because anything more blatant would undermine the mass vaccination programs that governments have spent a lot of blood, sweat, and tears (and money) to initiate. Furthermore, when the healthcare system is at breaking point, a vaccine that provides limited efficacy is better than no vaccine at all. In fact, whether the current vaccines are effective or not against the South African variant is moot. The much bigger point is that, just like the flu virus, the Covid virus will continue to mutate. And the mutations of the virus that get around the population’s immunity are more likely to spread and become the new dominant strains. If the current mutations cannot evade the vaccinations, then a future mutation eventually will. It’s just a matter of time. The immunity from the Covid vaccine will be short-lived. Hence, the evidence strongly suggests that the immunity from the Covid vaccine will be short-lived. How To Contain Covid Unlike smallpox and the measles, a mass vaccination program for Covid is unlikely to banish the disease. Instead, like the flu, vaccination will only contain the disease before it re-emerges in a new guise within a year or, worse, six months. In this regard, the Australia/New Zealand strategy of sealing national borders is not a strategy. It just buys time to form a strategy. No advanced open economy can shut its borders for ever. Meanwhile, a strategy to implement an all-seeing government track and trace system seems to be successful in China. However, it would be impossible to implement in liberal democracies, where the public would not accept the government watching your every move. In which case, can we just adopt the same strategy for Covid as we use for the seasonal flu – offer vaccinations to the most medically vulnerable once or twice a year? In theory, yes. But in practice, this strategy would overwhelm our healthcare systems. This we know because a bad flu season, by itself, was already pushing some healthcare systems to the limit. So how would this strategy cope with Covid, which creates conservatively five times as many casualties? Understand that the key metric is not the mortality rate, but the morbidity (severe illness) rate. Or more specifically, the morbidity rate versus the healthcare system’s intensive care unit (ICU) capacity (Chart I-3). Death requires very little medical intervention and resource, whereas severe illness requires massive medical intervention and resource. Moreover, when the severe illness is a respiratory illness, it leaves the sufferer struggling to breathe and needing critical care. No civilized society can deny critical care to somebody who is struggling to breathe (Chart I-4). Chart I-3Critical Care Capacity Is Limited Chart I-4Healthcare Systems Are Running Out Of Capacity Let’s optimistically assume that the Covid morbidity rate is around 1 percent. So, if a hundred thousand people get infected, one thousand will need an ICU. If the average stay in an ICU is three days, this equals 3000 ICU days. But in the UK, there are only around 10 ICUs per a hundred thousand people. This means that the UK can tolerate only 333 new daily infections in a hundred thousand people before the ICU capacity is breached. Now assume that only 1 out of 4 Covid infections is officially recorded because of asymptomatic/mild infection or reluctance to get tested. This would equate to 83 recorded daily infections per hundred thousand people, or 56 thousand per the whole UK population. Interestingly, the empirical evidence supports our assumptions. When recorded daily infections in the UK recently breached 60 thousand, the healthcare system approached its capacity. At this point, the UK healthcare system was on the brink of turning way those that needed urgent care, such as cancer patients. Luckily, UK daily Covid infections have eased, but crucially, this is only because of a national lockdown (Chart of the Week). So yes, we can adopt the same strategy for Covid as for the seasonal flu – offer vaccinations to the most medically vulnerable once or twice a year. But it will only work with some form of additional restrictions, dialled up and down based on the capacity utilisation of the healthcare system. This leads to two important takeaways, one short-term and one long-term. The short-term takeaway is that varying degrees of social distancing and restrictions to movement will be around for much longer than the general public and financial markets expect. Varying degrees of social distancing and restrictions to movement will be around for a long time. The long-term takeaway is that the best strategy to liberate the people and the economy is to add lots of capacity to healthcare systems. The pandemic has taught us that investing in healthcare will increase our long-term freedoms, wellbeing, and wealth. The Investment Conclusions 1. The so-called ‘economy reopening’ plays are due a reversal. In particular, the strong recent outperformance of the basic resources sector will unwind. 2. The extended period of social distancing and restrictions to movements will solidify a more remote way of working, shopping, and interacting. Hence, the so-called ‘pandemic plays’ still have a lot of structural upside. As such, long-term investors should avoid value, and stick with growth (Chart I-5). Chart I-5Long-Term Investors Should Avoid Value, Stick With Growth 3. As governments realise that the smart strategy is to add capacity to their healthcare systems, it will give a strong tailwind to healthcare stocks. Stay structurally overweight healthcare. 4. It follows that healthcare-heavy developed markets (DM) will outperform resource-heavy emerging markets (EM). Stay structurally overweight DM versus EM (Chart I-6). Chart I-6DM Versus EM = Healthcare Versus Resources Fractal Trading System* Fractal analysis suggests that the Korean won has reached an intermediate top versus the US dollar. This is implied by both the 130-day and 65-day fractal structures. Accordingly, the recommended trade is short KRW/USD, setting a profit-target and symmetrical stop-loss at 2.5 percent. The rolling 12-month win ratio now stands at 56 percent. Chart I-7KRW/USD When the fractal dimension approaches the lower limit after an investment has been in an established trend it is a potential trigger for a liquidity-triggered trend reversal. Therefore, open a countertrend position. The profit target is a one-third reversal of the preceding 13-week move. Apply a symmetrical stop-loss. Close the position at the profit target or stop-loss. Otherwise close the position after 13 weeks. * For more details please see the European Investment Strategy Special Report “Fractals, Liquidity & A Trading Model,” dated  December 11, 2014, available at eis.bcaresearch.com.   Dhaval Joshi Chief European Investment Strategist dhaval@bcaresearch.com Footnotes 1 Journal of Immunology Research, Volume 2018: Immune Responses to RNA Viruses, by Elias A. Said 2 Source: National Institute for Communicable Diseases (NICD) of the National Health Laboratory Service (NHLS), Johannesburg, South Africa SARS-CoV-2 501Y.V2 escapes neutralization by South African COVID-19 donor plasma (biorxiv.org) Fractal Trading System   Cyclical Recommendations Structural Recommendations Closed Fractal Trades Trades Closed Trades Asset Performance Currency & Bond Equity Sector Country Equity Indicators Bond Yields Chart II-1Indicators To Watch - Bond Yields Chart II-2Indicators To Watch - Bond Yields Chart II-3Indicators To Watch - Bond Yields Chart II-4Indicators To Watch - Bond Yields   Interest Rate Chart II-5Indicators To Watch - Interest Rate Expectations Chart II-6Indicators To Watch - Interest Rate Expectations Chart II-7Indicators To Watch - Interest Rate Expectations Chart II-8Indicators To Watch - Interest Rate Expectations  

Highlights Biden’s initial political capital is moderate-to-strong according to our Political Capital Matrix. He will pass his American Rescue Plan and one or two budget reconciliation bills over the next 18 months. Investors will need to discount the impact of tax hikes eventually.  The Democrats’ second impeachment of President Trump is a distraction but the party will not let it derail their legislative agenda. The bipartisan power-sharing agreement in the Senate will keep the filibuster in place for now (though not permanently). This does not affect the most market-relevant aspects of Biden’s policies, at least not in 2021, but beyond that it is an open question.  The stock rally is stretched, so prepare for volatility in the near term. But over the long run continue to prefer stocks over bonds, cyclicals over defensives, and value over growth stocks. Feature The US equity rally is getting frothy even as President Joe Biden kicks off his administration with a flurry of executive orders. Financial exuberance stems from combined monetary and fiscal stimulus that will provide a positive backdrop for risk assets for most of this year. Still, most of the good news is priced so we expect volatility to revive in the short run. The BCA Equity Capitulation Indicator is nearing the highest points of its historic range, which is typically a signal for a 10% equity correction or more (Chart 1). Not all indicators point decisively to a bubble that will pop imminently but several suggest that a bubble is being formed.1 The policy backdrop of fiscal largesse combined with an ultra-dovish Fed makes it easy to see why some parts of the market are getting manic. In this context, the Biden administration’s regulatory and tax agenda will become a negative catalyst in the short run even though its big spending will secure the economic recovery, which is positive in the long run. Chart 1Mania Unfolding Biden’s First Executive Orders Biden’s initial decrees brought zero surprises so far. He rejoined the Paris climate agreement, canceled the Keystone XL pipeline, suspended new oil and gas leasing on federal land, reversed President Trump’s border emergency and immigration curbs, ordered federal workers to wear masks, and directed the federal government to “Buy American.” The energy sector suffered the brunt of Biden’s initial regulatory salvo but the relative performance of energy stocks did not drop as much as financials, where Biden’s regulatory risks are less immediate. Biden’s policies are negative for health care stocks but they suffered least from what was a general setback for value plays in the context of a small bounce in the dollar and fears about global growth weakness stemming from the pandemic which has not yet been quelled. Large caps in all three of these sectors are underperforming small caps, suggesting that Biden’s new regulations and looming tax hikes are not driving the markets – at least not yet (Chart 2). Rather these cyclical small caps stand to benefit from the administration’s large spending plans, which include the $1.9 trillion American Rescue Plan currently being negotiated (Table 1). These plans are highly likely to pass as explained below.    Chart 2Biden's Executive Orders: No Surprises So Far Table 1Biden’s American Rescue Plan (With Previous COVID Relief) Going forward, Biden’s regulatory onslaught will bring negative surprises eventually as it expands and deepens but these will not counteract the stronger tailwinds of the vaccine and fiscal spending. Democrats have yet to invoke the Congressional Review Act, which enables them rapidly to reverse the regulations that the Trump administration ordered just before leaving office.2 The regulatory risk is greater for health care and energy than it is for financials and tech, though the latter two are not void of risk. Health care is the Democrats’ top priority outside of pandemic relief and economic recovery. (See Appendix for our updated political risk matrix by sector.) While the market can look through Biden’s regulatory threat, at least for now, it cannot look through the impact of higher taxes on corporate earnings forever. Over the next two months House Democrats will start revealing details of their budget proposals, which could serve as a negative catalyst for the overstretched equity rally. Other negative catalysts from an ambitious new administration are also possible with a market at such dizzy heights. Secretary of Treasury Janet Yellen has discouraged raising taxes initially but investors know that taxes will go up sooner or later. Moreover the specific legislative vehicle for Biden to push his agenda – “budget reconciliation” – requires tax hikes to offset spending increases. Thus if Democrats initiate a reconciliation bill in February or March then it will imply at least some revenue offsets, even if the biggest tax increases are saved for the second reconciliation bill for FY2022. Bottom Line: Value stocks have taken a breather but will continue to outperform over the cyclical 12-month time horizon. Looming Democratic tax proposals are more likely to serve as a near-term negative catalyst for the overstretched equity rally than Biden’s regulatory onslaught, which will take time to be felt. We are sticking with value over growth stocks due to the extremely accommodative fiscal and monetary policy setting. The Filibuster Preserved (For Now) A critical check on lawmaking in the Senate, the filibuster, has been preserved – at least for the moment. This is positive news for markets as it lowers the odds of major legislative surprises this year. The filibuster enables senators to block normal legislation through endless debate. Sixty senators are needed to invoke “cloture” and bring debate to a close. Otherwise the bill goes nowhere. With the Senate divided evenly at 50-50 seats between the two parties, Biden’s agenda will now depend on any bills that can garner 10 Republican senators, plus two “budget reconciliation” bills for fiscal 2021-22. Reconciliation bills only require a simple 51-seat majority in the Senate. Eliminating the filibuster will remain a risk over the long run. It was only preserved because two centrist Democratic senators, Joe Manchin of West Virginia and Kyrsten Sinema of Arizona, declared that they would not vote to abolish it. This prompted Republican Senate Minority Leader Mitch McConnell to drop his chief demand, that the filibuster be kept, in his negotiations with Democratic Majority Leader Chuck Schumer toward an agreement for the two evenly divided parties in the Senate to share power. Now a power-sharing agreement is in place so the legislative process can begin, albeit within the filibuster’s guardrails. Notice that Schumer never conceded to McConnell that the filibuster would be preserved. And two Democrats is not very many. Later these centrists may succumb to party pressure, say amid Republican obstructionism of a voting rights bill, to eliminate the filibuster. The last time the Senate was evenly divided, after the 2000 election, the power-sharing agreement only lasted six months, from January to June 2001. A single retirement or death could turn the balance. Moreover since Democrats have the option of two reconciliation bills first, the filibuster is not a substantial check on them until 2022 or beyond, at which point the centrists could fall under sustained pressure.3 Bottom Line: Preserving the filibuster provides a source of stability – it reduces policy uncertainty and polarization. It restricts Biden’s agenda largely to his major initiatives: entrenching the Affordable Care Act, expanding infrastructure spending, partially repealing Trump tax cuts, and various other tax-and-spend measures known to investors. It lowers the chance that financial markets will be blindsided in 2021 by a sweeping new legislative initiative – for example, the Green New Deal – or radical redistributive schemes. While markets will need to discount the tax hikes they will be able to recover more quickly than if they also expected a stream of unpredictable legislation from a Senate unshackled from the filibuster. Stimulus And The Tax Hike Timeline The American Rescue Plan could pass in February at the earliest or April at the latest. If at least 10 Republican senators cooperate then it will fly through Congress. The advantage of this bipartisan route is that it would achieve an early Biden objective while still leaving Democrats with two full chances to pass reconciliation bills covering fiscal 2021-22. The economic recovery would be on sure footing thereafter, giving Biden more room to maneuver (Charts 3 and 4). Chart 3Is More Stimulus Necessary? Bipartisan talks are under way. Senator Joe Manchin of West Virginia set up talks with about 15 other senators and three White House aides, including National Economic Council director Brian Deese, toward revising and passing the rescue plan.4 Winning over ten Republicans is a tall order but GOP senators are aware that the pandemic is still going and even Republican voter opinion favors more relief. So far Democrats have not allowed any compromise in the size of the deal but that could change to get 60 votes, since they can always make up the difference through reconciliation later. The rescue plan is unlikely to be passed before Trump’s second impeachment trial begins on February 8, however. If 10 Republicans cannot be found, the Senate will be slowed down by juggling reconciliation and impeachment. Trump’s first impeachment took 49 days, leaving the average at 65 days (Table 2). It will keep the Senate busy at least through mid-March. Chart 4More Checks Coming For Households? Table 2Impeachment Takes At Least A Month Since Democrats are highly unlikely to win over 17 Republicans to convict Trump of inciting insurrection, the impeachment could be a policy mistake. Democrats are determined not to let slide the opportunity to position themselves as the arch defenders of democracy. Acquitting Trump would put several prominent Republicans on record endorsing him even after his alleged interference with the peaceful transition of power. However, impeachment will not be allowed to derail Biden’s agenda. The Democratic Party controls both processes. The Senate can wrap up the trial if it becomes an obstacle. Diagram 1 presents the timeline for these events to occur. The implication is that March 14, when the latest expansion of unemployment benefits starts to expire, will serve as a deadline for Biden’s rescue plan. Diagram 1Timeline Of Impeachment, Budget Reconciliation, And Regular Legislation Budget reconciliation takes seven months on average but it only took three months in 2017, which is the proper analogy for today. Even if tax hikes are passed in Q2 there is an open question as to when they would take effect (Diagram 2). Prudent investors should be prepared for a retroactive January 1, 2021 effective date, even if it is more likely that they will kick in on January 1, 2022 to give the economy more time to recover. Again, taxes pose a risk to the rally. Diagram 2How Long Does It Take To Pass A Budget Reconciliation Bill? If Republicans do not cooperate on Biden’s rescue plan then Democrats will cite it as obstructionism from the beginning, despite Biden’s call to unity, and it will play into any future efforts to eliminate the filibuster. But those will likely center on the period after the two reconciliation bills. Bottom Line: As the House Democrats begin to draft their first budget resolution, to initiate the reconciliation process, tax hikes will come more into focus. The near-term upside risk is that Democrats skip taxes in the first bill and save it for later. But there will have to be at least some revenue raisers in any reconciliation bill. So a near-term pullback is entirely reasonable to expect. We would be buyers on the dip given the extremely accommodative fiscal and monetary backdrop. Introducing Our Political Capital Index To assess any government’s capability – namely its ability to alter the policy setting that affects the economy and financial markets – we need to measure its political capital or grounds of support. To this end we have constructed a Political Capital Index to measure the strength and capability of US ruling parties and presidencies (Table 3). Table 3Political Capital Index The Political Capital Index shows a series of political and economic indicators, as of the latest available data (December or January), as well as the change since Biden’s election in November.5 Below we describe the political and economic categories of political capital that we chose and the data we use to represent them: Political Strength: The most basic measure of political capital is President Biden’s margin of victory in the popular vote (4.4%) and Electoral College vote (306/538), the number of days he has been in power, his party’s Congressional majorities, and the Supreme Court’s ideological leaning. These components will last for two-to-four years and can only be changed by new elections or deaths (Table 4). Even a president elected in a landslide would see his political capital decay over time. The sooner the next election, the less political capital the ruling party has. The president and Congress will have more trouble passing legislation just before the election and will be more careful about what they do pass to avoid punishment at the ballot box. Any difficult economic policies or reforms will tend to be done at the beginning of the term, as political capital is still abundant and the next election is not a clear and present danger. President Biden has moderate political capital. His popular victory was solid, his electoral victory was the same as President Trump’s, but his congressional majorities are weak. His initial legislative efforts should be assumed to pass but aside from his rescue plan and one or two reconciliation bills he will not be able to get much else done. Table 4Political Capital: White House And Congress Household Sentiment: Household sentiment is the origin of political capital since households are voters. We measure it through presidential net approval ratings, both in general and in handling the economy, as well as through consumer confidence (Chart 5). Household sentiment changes easily – it can drive policies and react to them. Even if the economy is objectively improving, sentiment can remain downbeat if politicians fail to communicate their policies, which could cost them the election. Measures that improve household pocketbooks or welfare are more popular than those that impose structural changes like taxes and regulation. But reforms are possible when a politician has sufficient political capital, or when a worse outcome would follow from doing nothing. Biden will start with a higher approval rating than President Trump but his average approval is not much higher at present and consumer confidence has ticked down as a result of the pandemic. His economic stimulus should create an improvement in household sentiment in the coming year. Chart 5US Households: Still Downbeat Business Sentiment: Business sentiment is another important element of political capital. Businesses that are confident about the economy’s prospects will spend on capex, new orders, and new hires, and they will also deplete their inventories (Table 5). Animal spirits respond to spending, taxation, regulation, and trade – all areas where politicians have some control. Table 5Political Capital: Household And Business Sentiment Policymakers can run down business sentiment by enacting painful policies for business, in favor of government or households or personal whim – or they can pass business-friendly policies to boost animal spirits. Businesses cannot vote like households but they have a powerful influence over politicians through lobbyists and political donations and a powerful influence on voters through employment. Higher animal spirits encourage new employment, which improves household welfare, thus boosting political capital. Biden is starting out fairly strong with respect to business sentiment, with the exception of the service sector, which is still beaten down by the pandemic. This is an area where his political capital could decay over time. Big business was happy to get rid of Trump’s trade war but now it faces larger government encroachment. This risk is flagged by small businesses, which are already highly distrustful of new taxes and regulation (Chart 6). Chart 6US Business Sentiment Chart 7Measures Of Polarization Political Polarization: Starkly divided populations and governments are often gridlocked or obstructionist, preventing policies from getting approved or implemented (Chart 7). Our polarization proxy measures the difference in approval of the sitting president according to party, while our economic polarization measure does the same for economic sentiment. Structural polarization is a low-frequency data series from political science literature that measures whether House members and senators tend to vote with the “party line” or “reach across the aisle.”6 The Philly Fed Partisan Index also measures the degree of political disagreement among politicians at the federal level. A highly polarized environment ensures that there will be strong opposition to any policy put forward by lawmakers and a higher likelihood of reversal by the next governing party. This leads to erratic policymaking and policy uncertainty among households and businesses. Lower polarization increases the durability of policies. Fiscal Policy: The government sector contributes to political capital through fiscal policy, especially fiscal thrust (the change in the cyclically adjusted primary budget deficit) (Table 6). An expansionary fiscal policy affords policymakers greater latitude – especially in times and places where inflation is not a public concern. It can also be an effort by the ruling party to boost its political capital when it is low, or when an election looms. The Biden administration is lucky to start off with a new business cycle, as Obama did in 2009, but the large dose of fiscal support today will become a fiscal drag by 2024 so the long-term effectiveness of today’s “pump priming” will be essential. Table 6Political Capital: The Economy And Markets Economic Conditions: Economic conditions are arguably the most important component of political capital. We included several objective measures of household wellbeing such as unemployment, inflation, gasoline prices at the pump, and wage growth. If voters have seen their quality of life improve under the current set of leaders then they are more likely to vote to continue their windfall. To judge whether a party will be re-elected, it is critical to know whether household wellbeing has changed since the last election. High unemployment, high inflation, high economic uncertainty, and high bankruptcy levels point to struggling voters who are more likely to take their grievances to the ballot box. By the same token, leaders will struggle to get anything done if voters are beset with these ills. Asset Markets: Asset markets play at least some role in determining political capital. Most voters are not highly exposed to the stock market, though they care about their pension fund. Most voters are highly exposed to the property market. A euphoric stock market will not necessarily buoy the political capital of a president or ruling party, as demonstrated by the recent election: President Trump’s approval was closely linked to the stock market, which also restrained his actions, yet a rallying market did not get him re-elected. A market crash will always hurt policymakers, especially if it happens just before an election. We watch the stock market primarily as a downside risk to the ruling party’s political capital rather than upside. Bottom Line: Our Political Capital Index is how we will monitor President Biden’s and the Democratic Party’s capability in the coming months and years. The administration begins with moderate political capital but it is likely to improve on economic recovery, which will be secured through control of Congress and the purse strings. Our confidence that Biden’s American Rescue Plan and one or two reconciliation bills will pass stems from this assessment. This means a large spending program and tax hikes are highly probable and investors should prepare for them. Investment Takeaways Signs of mania – from Bitcoin to TESLA to GameStop – have gripped the market as the combined effect of ultra-dovish monetary and fiscal policy is priced. This process can continue beyond reasonable expectations. Nevertheless we are prepared for near-term volatility and a correction at any time. The rollout of the COVID-19 vaccine faces inevitable bumps and the pandemic is still triggering government lockdown measures and consumer caution – though these will improve over time. Biden’s regulatory agenda and especially looming tax hikes will also spur some risk aversion in the near term as the House Democrats begin preparing a reconciliation bill. Overcoming the hurdle of Trump’s impeachment will free up the Senate to move forward on reconciliation as well, which means tax hikes will fall under the market’s radar sooner or later. A regular bill could be passed in February without new taxes but otherwise a reconciliation bill will pass as early as April and include at least some new taxes, even if they take effect next year. We would still use the opportunity to buy into any further weakness in value plays relative to growth plays (Chart 8). Fundamentally the economy is set to improve this year, the pandemic is set to subside, and the policy support will be reinforced and expanded as necessary. Chart 8A Setback For Growth Versus Value Chart 9Equity Correction Looms The reflation trade is technically over-extended, investors are complacent, and some profit-taking is due. The extremely depressed put-to-call ratio tracks well with the US dollar index, both of which are showing signs of life (Chart 9). We would fade a rebound in the dollar, however, as the Democratic Party’s policies will ensure widening twin deficits (budget and trade deficits) even as the Fed demonstrates its commitment to its new goal of allowing an inflation overshoot to make up for past undershoots.   Matt Gertken Vice President Geopolitical Strategy mattg@bcaresearch.com Jesse Anak Kuri Associate Editor jesse.Kuri@bcaresearch.com   Appendix Table A1Political Risk Matrix Table A2Biden’s Cabinet Position Appointments   Footnotes 1     See BCA Research US Equity Strategy, “Overdose?” January 25, 2021, bcaresearch.com. 2     The Congressional Review Act of 1996 enables Congress to speed up the removal of regulations that were adopted recently, in this case since August 21, 2020. The process requires both houses of Congress to repeal a regulation but the Senate cannot prevent repeal via filibuster. The Trump administration used the law aggressively to remove several of President Barack Obama’s outgoing regulations. See Jonathan H. Adler, “Will Democrats Learn To Love The Congressional Review Act?” Reason, January 23, 2021, reason.com.  3    Democrats are explicitly interested in repealing the filibuster, as Biden and Senate Majority Leader Chuck Schumer have indicated (not to mention former President Obama who characterized it as a relic of the racist Jim Crowe era). 4    See Ed O’Keefe et al, “16 senators from both parties meet with White House on COVID-19 relief plan,” CBS News, January 25, 2021, cbsnews.com; Aamer Madhani and Lisa Mascaro, “White House Begins Talks With Lawmakers On COVID-19 Relief,” Associated Press, January 25, 2021, apnews.com.  5    Biden’s term technically began on January 20 but voters in 2024 will judge the president and ruling party based on whether they are better off than they were four years ago, i.e. when they last made a major judgment. 6    See Jeffrey Lewis, Keith Poole, Howard Rosenthal, et al, at voteview.org.  

As the economy is transitioning from liquidity to growth, the oil-to-gold price ratio has caught our attention again this year. As a reminder, last year we successfully traded this high-octane pair using the S&P oil & gas exploration & production (O&G E&P) index on the long side and the global gold miners index on the short side. We pocketed gains of 10% in early May of 2020, only to reinstate the trade again and to scoop a further 32% in gains. This year, the latest ISM manufacturing survey release painted a bright picture for this intra-commodity price ratio once again (see chart), and while we are not reinstituting the pair trade just yet, it is now flashing on our radar screen; we are patient and await a better entry point. Reopening of the economy and related energy demand recovery will underpin oil prices and producers going forward, at the same time as rising real yields will weigh on the shiny metal and gold mining stocks. Bottom Line: Put a stop buy on long S&P O&G E&P/short global gold miners via the XOP/GDX exchange traded funds at a ratio of 1.2.

Highlights Fed: We will use the monthly US employment data to track progress toward the first Fed rate hike. At present, our base case outlook calls liftoff in late-2022 or the first half of 2023. Investors should maintain below-benchmark portfolio duration. Corporate Bonds: The macro environment is supportive for spread product returns, but there are better opportunities than in investment grade corporate bonds. We prefer high-yield over investment grade within the US corporate space, particularly the Ba credit tier. Munis: Muni value has deteriorated markedly, but the sector still looks attractive compared to investment grade corporate bonds. EM Sovereigns: We recommend owning investment grade USD-denominated EM Sovereign bonds instead of investment grade US corporates. Within high-yield, US corporates still offer a better opportunity than EM Sovereigns. Using Employment Data To Time Fed Liftoff The current debate raging in fixed income circles revolves around whether large-scale fiscal stimulus will cause inflation to flare this year, possibly leading to a much earlier fed funds liftoff date than is currently priced into the yield curve (Chart 1). Chart 1Fed Liftoff Priced For July 2023 Last week’s report discussed our outlook for inflation in 2021.1 In short, our base case calls for 12-month PCE inflation to peak above the Fed’s 2% target in April but to then fall back below 2% by the end of the year. However, there is a compelling case to be made that inflation could rise more quickly. Table 1A Checklist For Liftoff Last week, our Global Investment Strategy service pointed out that the combined effect of December’s fiscal stimulus deal and President Biden’s newly proposed American Rescue Plan would inject an average of $300 billion per month into the economy through the end of September.2 The Congressional Budget Office estimates that the monthly output gap – the difference between what the economy is capable of producing and what it is actually producing – is currently $80 billion. In that environment, it’s not hard to see how excess demand could lead to price increases in certain sectors. Chart 2How Far From "Maximum Employment"? Of course, for bond investors what matters is not just the path of inflation but how the Fed responds. If rising inflation prompts the Fed to lift rates before July 2023 – the liftoff date currently priced into the market – then bonds will sell off. If liftoff occurs later, then yields will fall. This makes timing the liftoff date critical, and fortunately, the Fed has given us three explicit criteria that must be met before liftoff will occur (Table 1). This week’s report focuses, not on inflation, but on the condition related to “maximum employment.” Our sense is that if the Fed does not think the economy is at “maximum employment” it will ignore modest overshoots of its 2% inflation target on the view that the large amount of labor market slack will eventually cause inflationary pressures to wane. We define “maximum employment” as an unemployment rate of 4.5%, consistent with the upper-bound of the Fed’s most recent range of NAIRU estimates (Chart 2). Using that assumption, and an assumption for the path of the labor force participation rate (Chart 2, bottom panel), we can calculate the average monthly payroll gains that must occur for the unemployment rate to hit the 4.5% target by specific future dates. Our results are shown in Table 2. We use four different scenarios for the labor force participation rate. The lowest estimate assumes that the participation rate remains at its current level. The highest estimate assumes that it re-converges to its pre-COVID level at the same time as the unemployment rate hits 4.5%. The two middle estimates assume smaller increases of 1% and 0.5%, respectively. Table 2Average Monthly Nonfarm Payroll Growth Required For The Unemployment Rate To Reach 4.5% Over The Given Horizon We expect the participation rate to rise as the economy recovers and people are drawn back into the labor force, but some workers have likely been permanently displaced by the pandemic and a full convergence back to pre-COVID levels may not occur until well after the unemployment rate reaches 4.5%, if at all. With that in mind, the “Convergence To Pre-COVID” scenario probably overstates the monthly payroll gains necessary to hit full employment and the “Stays At 61.5%” scenario almost certainly understates them. If we focus on the two middle scenarios, we see that average monthly payroll gains of between 472k and 572k are required for the unemployment rate to hit 4.5% by the end of this year. This range falls to 346k - 413k if we push the liftoff date out until mid-2022 and to 283k – 334k if we move out until the end of 2022. At first blush, these numbers look unattainable. Between 2010 and 2019, average monthly payroll growth averaged a mere +97k. But, given the downturn that just occurred, employment growth will likely be much stronger going forward. Our research into past economic cycles has found that the two main determinants of average monthly employment growth during the first year following a recession are: The drawdown in employment that occurred during the recession (a larger drawdown correlates with greater payroll growth in the first 12 months of recovery) Real GDP growth during the first 12 months of recovery Chart 3 shows the correlation between the peak-to-trough decline in nonfarm payrolls during the past eight US recessions and the average monthly payroll gains seen during the first 12 months of economic recovery. The correlation is quite linear except for the 2008 recession where the peak-to-trough decline in payrolls was 8.7 million but the bounce-back was incredibly weak. Chart 4 explains why the 2008 recession looks like such an outlier in Chart 3. Real GDP growth during the first 12 months of recovery coming out of the 2008 recession was very low, only 2.6%. Chart 3Large Payroll Drawdowns Tend To Be Followed By Strong Gains… Chart 4…And Occur Alongside Strong Economic Recovery Thinking about the current recovery from the COVID recession. Nonfarm payrolls fell by about 22 million from peak to trough in 2020. This is literally off the charts (looking at Chart 3), about 2.5 times the job loss seen in 2008. Then, the Fed’s most recent median estimate for real GDP growth in 2021 is a robust 4.2%, and this estimate was made before Democrats took control of the Senate and proposed a massive new stimulus bill. Considering both the large drawdown in employment and the outlook for rapid GDP growth in 2021, average monthly payroll gains should be quite strong this year. A return to a 4.5% unemployment rate by the end of 2021 is probably a long shot, but we can easily envision average monthly payroll gains on the order of 300k to 400k per month, enough to prompt Fed tightening by late-2022 or the first half of 2023. Whatever transpires, we will monitor monthly payroll growth in the coming months and use this analysis to continuously reassess our liftoff expectations. For the time being, investors should keep portfolio duration low. Alternatives To Investment Grade Corporates Another conclusion that falls out of the above analysis is that the runway for spread product outperformance remains long. With Fed tightening unlikely until late-2022 or the first half of 2023, monetary conditions will remain accommodative for some time. This will drive a continued search for yield, supporting the outperformance of spread product relative to Treasuries. But despite the supportive macro environment, bond investors face a problem that the most popular US spread sector – investment grade corporate bonds – looks very expensive. The average option-adjusted spread for the Bloomberg Barclays investment grade corporate index is only 2 bps above its pre-COVID low, and the spread on Baa-rated bonds is exactly equal to its pre-COVID low. Aa- and A-rated bonds appear somewhat cheaper (Chart 5). The valuation picture is even bleaker after adjusting the index to ensure a constant average credit rating and average duration over time. The 12-month breakeven spread for the credit rating-adjusted corporate index has only been tighter 3% of the time since 1995 (Chart 6). Chart 5IG Spreads Are Tight... Chart 6...Especially After Adjusting For Risk The remainder of this report discusses potential alternatives to investment grade corporate bonds. Specifically, we’re looking for spread products that will benefit from the same macro environment as investment grade corporates, but where investors can pick up some additional risk-adjusted value. Candidate #1: Junk Bonds Chart 7Ba-Rated Corporates Are Cheap One obvious thing investors might consider is a move down the quality spectrum into high-yield bonds. This move comes with greater credit risk, but we believe the incremental spread pick-up provides more than fair additional compensation. The Bloomberg Barclays High-Yield index’s average option-adjusted spread is still 33 bps above its pre-COVID low, and the spread pick-up in the Ba credit tier relative to the Baa credit tier looks particularly compelling (Chart 7). The supportive macro environment makes us less worried about taking additional credit risk in a portfolio, and we recommend that investors pick up the additional spread offered in the high-yield space. The elevated incremental spread pick-up in Ba bonds makes that credit tier look like the best risk-adjusted opportunity. Candidate #2: Tax-Exempt Municipal Bonds Municipal bond spreads have tightened dramatically during the past couple of months and Aaa-rated Munis no longer look cheap compared to Treasuries (Chart 8). That said, if we match the duration and credit rating between the Bloomberg Barclays Municipal bond indexes and the US Credit index, we find that both General Obligation (GO) and Revenue Munis appear attractive compared to US investment grade Credit. Both GO and Revenue Munis offer a before-tax spread pick-up relative to US Credit for maturities above 12 years (Chart 9), the same goes for Revenue bonds with 8-12 year maturities. Revenue bonds in the 6-8 year maturity bucket offer an after-tax yield pick-up versus Credit for investors with an effective tax rate of 10% or higher. GO bonds in the 8-12 year and 6-8 year maturity buckets offer breakeven effective tax rates of 14% and 26%, respectively. Chart 8Muni / Treasury Yield Ratios Chart 9Munis Still Attractive Versus Corporates All in all, municipal bond value has deteriorated markedly in recent months and we therefore downgrade our recommended allocation slightly from “maximum overweight” (5 out of 5) to “overweight” (4 out of 5). Investors should still prefer tax-exempt municipal bonds relative to investment grade corporate bonds with the same credit rating and duration. Candidate #3: USD-Denominated Emerging Market Sovereigns For all of last year we advised investors to favor investment grade corporate bonds over USD-denominated EM Sovereigns of equivalent credit rating and duration. This positioning worked out well. Since the March 23rd peak in credit spreads, the A3/Baa1-rated EM Sovereign index has only outperformed the duration-matched A-rated US Credit index by 159 bps while it has underperformed the Baa-rated US Credit index by 571 bps (Chart 10). In the high-yield space, the B1/B2-rated EM Sovereign index has significantly underperformed both the Ba and B-rated US junk bond indexes. Chart 10EM Sovereigns Underperformed US Corporates In 2020 But now, after nine months of poor relative performance, value is starting to look more compelling in the EM Sovereign space. Chart 11 shows that EM Sovereigns offer a yield pick-up versus duration-matched US corporate bonds for all credit tiers except Ba. At the country level, the yield advantage in the A and Aa credit tiers is attributable to opportunities in Qatar, UAE and Saudi Arabia (Chart 12). In the Baa credit tier, investors should look for opportunities in Mexico, Russia and Colombia, while avoiding the Philippines. Chart 11USD-Denominated EM Sovereign Spreads Versus Credit Rating And Duration-Matched US Credit: By Credit Rating Chart 12USD-Denominated EM Sovereign Spreads Versus Credit Rating And Duration-Matched US Credit: By Country All in all, investors should shift some allocation away from investment grade corporates and into USD-denominated EM Sovereigns with equivalent duration and credit rating, focusing on the countries that offer a yield pick-up. Turning to high-yield, we would rather own junk-rated US corporate bonds than junk-rated EM Sovereigns. US corporates offer a yield pick-up over EM Sovereigns in the Ba credit tier, and the sky-high spreads offered by B and Caa-rated EMs are due to overly risky opportunities in Turkey and Argentina. We don’t see these countries benefiting from the supportive US macro environment in the same way as US corporate credit, and therefore recommend overweighting US corporate junk bonds over EM Sovereign junk bonds. Bottom Line: Investors should continue to overweight spread product versus Treasuries in US fixed income portfolios but should look for opportunities outside of investment grade corporate bonds. We recommend owning municipal bonds and USD-denominated EM Sovereign bonds in place of investment grade US corporate bonds with the same credit rating and duration. We also recommend taking additional credit risk in US junk bonds, particularly in the Ba credit tier. Investors should prefer US junk bonds over junk-rated EM Sovereigns.   Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1  Please see US Bond Strategy Weekly Report, “Trust The Fed’s Forward Guidance”, dated January 19, 2021, available at usbs.bcaresearch.com 2  Please see Global Investment Strategy Weekly Report, “Stagflation In A Few Months?”, dated January 22, 2021, available at gis.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification