Policy
Your feedback is important to us. Please take our client survey today. Highlights Duration: The Fed’s adoption of an Average Inflation Target and the emerging correlation between bond yields and a “blue sweep” election outcome were the two main catalysts that caused us to reduce our recommended portfolio duration stance last week. Monetary Policy: The Fed is unlikely to increase its pace of Treasury purchases, unless a selloff in risky assets (equities and credit) threatens the economic recovery. As long as the Fed sticks with its ultra-dovish interest rate guidance, a moderate bear-steepening of the Treasury curve will not cause such a selloff. Economy: Weekly claims data are consistent with a continued decline in the unemployment rate, due mostly to fewer temporarily unemployed workers. All in all, the US economy is recovering, but it remains very far from full employment. Feature Chart 1Real Yields Have Troughed After having advocated “at benchmark” portfolio duration since March, we officially lowered our recommended duration stance to “below benchmark” in last week’s Special Report.1 Two main catalysts led us to this decision. First, there was the Fed’s late-August adoption of an Average Inflation Target. This was an important bond-bearish catalyst because it signaled that the Fed’s reaction function has reached its maximum dovishness. Treasury yields stayed low throughout the summer even as the economy recovered because the Fed was simultaneously guiding the market toward a more dovish reaction function. As evidence for this dynamic, notice that between March and August the uptrend in the cost of inflation compensation was completely offset by falling real yields (Chart 1). But now, the Fed has officially adopted its new Average Inflation Targeting framework. In addition, it has promised not to lift rates at all until inflation is above its 2% target and is expected to overshoot that target for some time. To get more dovish from here, the Fed would probably need to actually increase its long-run inflation target from 2% to 3%, a step it is not prepared to take. For this reason, we expect that the Fed has exhausted its ability to push real yields lower as the economy recovers. With the Fed’s interest rate reaction function at maximum dovishness, only a negative economic growth shock can push yields lower. With the Fed’s interest rate reaction function at maximum dovishness, only a negative economic growth shock can push yields lower. Such a shock would cause investors to anticipate a slower return of inflation and thus push bond yields down, even if the market’s assessment of the Fed’s willingness to respond to inflationary pressures (aka its reaction function) remains constant. The failure of Congress to deliver additional fiscal stimulus after the expiry of the CARES act’s main income-supporting provisions is the sort of thing that could cause such a shock. However, September’s market action made it clear that investors are willing to look past the failure to deliver a bill if they can look forward to a larger fiscal stimulus in January. As we wrote last week, the most likely election outcome of the Democrats winning the House, Senate and White House would certainly deliver on that promise. Bottom Line: The Fed’s adoption of an Average Inflation Target and the emerging correlation between bond yields and a “blue sweep” election outcome were the two main catalysts that caused us to reduce our recommended portfolio duration stance last week. Will The Fed Use Its Balance Sheet To Keep Bond Yields Low? Chart 2Fed Treasury Holdings Over Time One possible counterargument to our bond-bearish view is that, even if the Fed’s interest rate guidance is as dovish as it will get, the central bank will simply ramp up asset purchases to prevent any significant rise in long-maturity bond yields. Certainly, the Fed has not shied away from transacting in the Treasury market this year (Chart 2). In fact, the Fed took down more than 100% of gross note and bond issuance in the second quarter (Chart 3) and its current stated policy is to purchase at least $80 billion of Treasury securities per month. At present, the Fed holds more than 35% of all outstanding Treasuries with more than 10 years to maturity and about one quarter of the outstanding supply for all other maturities (Chart 4). This is a significant presence in the Treasury market, but not so large that the Fed would think twice about increasing its pace of Treasury purchases if such a policy was deemed necessary. But what would actually make the Fed increase its pace of asset purchases? Would a modest bear-steepening of the Treasury curve (our base case outlook) be enough? We doubt it. Chart 3Fed Purchases Peaked In Q2 Chart 4Fed Owns A Good Chunk ##br##Of The Market Chart 5Financial Conditions Are Highly Accommodative The Fed would be concerned if broad measures of financial conditions started to tighten, as that would indicate a looming period of slower economic growth. But that’s very different from long-maturity Treasury yields moving somewhat higher in response to an improving economic outlook. In fact, to get a meaningful tightening in broad measures of financial conditions, we would need to see significant credit spread widening and weaker equity prices (Chart 5). A bear-steepening Treasury curve, even if long-dated yields move 40-50 bps higher, will not prompt a selloff in credit markets or equities as long as the market believes that the Fed is committed to maintaining an accommodative monetary policy stance. Bottom Line: The Fed is unlikely to increase its pace of Treasury purchases, unless a selloff in risky assets (equities and credit) threatens the economic recovery. As long as the Fed sticks with its ultra-dovish interest rate guidance, a moderate bear-steepening of the Treasury curve will not cause such a selloff. Can The Fed Use It’s Balance Sheet To Keep Bond Yields Low? Chart 6Fed Purchases Work Mainly Through Signaling Interest Rate Intentions Of course, in the extreme, the Fed could decide to set a cap on the 10-year Treasury yield and promise to purchase as many securities as necessary to maintain that cap. This sort of Yield Curve Control would effectively prevent long-maturity Treasury yields from rising, even as the economy recovered. As discussed above, we think the economic situation would have to turn quite dire for the Fed to pursue such a policy. A more relevant question is whether, in the absence of a stated yield cap, the Fed’s current pace of $80 billion of Treasury purchases per month (or even $100 billion per month) will prevent Treasury yields from rising. Our sense is that, without a stated yield cap, Fed Treasury purchases won’t stop bond yields from rising. In fact, we see very little evidence to support the notion that changes in Fed Treasury purchases influence the trend in bond yields, beyond what the purchase announcements signal to markets about the Fed’s intentions with regards to interest rate policy. Consider Chart 6, which shows the 10-year Treasury yield alongside the Global Manufacturing PMI and two different measures of Fed Treasury purchases. At first blush, between 2010 and 2012, there is a fairly strong relationship between a falling 10-year Treasury yield and an increase in Fed Treasury holdings greater than five years. However, the correlation between the 10-year Treasury yield and the Global Manufacturing PMI during this period is even stronger. This suggests an alternative explanation where the decline in bond yields is driven by the market shifting its rate hike expectations out into the future in response to slowing economic growth. Greater Fed purchases only served to reinforce the Fed’s increasingly dovish interest rate guidance during this time. Without a stated yield cap, Fed Treasury purchases won’t stop bond yields from rising. The key point from Chart 6 is that it is difficult to identify periods when the Fed’s balance sheet policy and interest rate guidance suggest opposite outcomes for bond yields. Typically, when the Fed is ramping up asset purchases it is also signaling to the market that it will shift toward a more dovish interest rate policy. Similarly, when the Fed is reducing its asset purchases, it tends to also be preparing the market for eventual rate hikes. This makes it impossible to say conclusively whether a given move in bond yields is driven by interest rate guidance or balance sheet actions. In our minds, this casts a lot of doubt on the notion that the Fed could maintain its current interest rate guidance during the next 6-12 months while also preventing a rise in Treasury yields by increasing asset purchases. The policy of increasing asset purchases would appear incoherent if it wasn’t also paired with increasingly dovish forward rate guidance, guidance that the Fed is likely unwilling to deliver. Bottom Line: Even if the Fed modestly increases its monthly pace of Treasury purchases, or shifts some purchases further out the curve as some FOMC participants have suggested, the impact on long-dated Treasury yields will be negligible without a concurrent shift in interest rate guidance. The Hedging Effectiveness of Treasuries Is Diminished, But Not Gone Completely One topic that has come up a lot recently in our client interactions is the idea that Treasury securities are no longer an effective hedge for equity portfolios. There are two possible reasons why this could be true. The first is that with Treasury yields so close to the zero-lower-bound there is very little scope for capital appreciation in bonds. The second is that Treasury yields may not respond to falling equity prices by declining, as they have in the past. We don’t currently see much evidence for the second reason. The only way that bond yields wouldn’t decline alongside a major equity sell-off is if that sell-off was driven by high and rising inflation and expectations that the Fed would aggressively hike rates to combat higher prices. Inflation is nowhere near high enough for this to be a concern. The first reason, however, could hold some water. To test it, we first looked at this year’s COVID-driven 34% drop in the S&P 500 that occurred between February 19th and March 23rd (Table 1). If an investor purchased a 5-year Treasury note on February 19th and sold it on March 23rd, they would have earned 5.21% on that trade, offsetting some portion of the equity decline. The same trade in a 30-year T-bond would have earned 16.65%. Long-maturity Treasuries still perform their role as hedging instruments for equity portfolios. But with starting bond yields much lower today, Treasuries may not offer the same protection. Table 2 shows how much an investor would earn if they bought a Treasury security today, held it for six months, and during that timeframe the entire spot yield curve fell to zero. We also show what returns would be earned if the yield curve shifted in the same way it did during the Feb 19th – Mar 23rd equity selloff, except we don’t let any yields fall into negative territory.2 Table 1Total Returns From Stock Market Peak (Feb. 19th) To Stock Market Through (Mar. 23rd) Table 2Treasury Total Returns Over A Hypothetical 6-Month Period Notice that projected returns for shorter maturities in Table 2 are significantly lower than the returns earned during the Feb 19th – Mar 23rd episode. A 5-year Treasury note will earn only 1.90% during the next six months if the entire yield curve falls to zero, this is well below the 5.21% earned in February and March. However, for maturities beyond 10 years, returns are similar between Table 1 and Table 2. This makes sense because, unlike the short-end, those long-dated yields are still fairly far from the zero bound. Bottom Line: Treasuries still perform their role as hedging instruments for equity portfolios, but investors now have to move further out the curve, and thus take more interest rate risk, to get the same protection they received from less-risky shorter-maturity notes in the past. US Economy: Digging Into Claims Chart 7Labor Market Update Initial unemployment claims spooked some investors when the weekly number jumped to +898k two weeks ago, even as the consensus was calling for a small decline.3 But that spike was completely reversed last week, and a good chunk of it was actually revised out of the data. A delay in California’s reporting was one reason for the big swing. The state had failed to report claims data for two consecutive weeks at the beginning of the month. This meant that the national claims number was based on an assumed figure for the state. California resumed reporting last week and the result was a big drop in overall claims, back down to +787k. Stepping back from the large swings of the past two weeks, a good portion of which are statistical artifacts stemming from California’s reporting delay, we see that overall employment trends haven’t changed much. Initial claims have flattened-off since late-August, but continuing claims are still falling rapidly (Chart 7). This suggests that the unemployment rate will drop again next week when October’s employment report is released. The big question in next week’s employment report will be whether the decline in the unemployment rate is once again driven by falling temporary unemployment, or whether the number of permanently unemployed workers will reverse its uptrend (Chart 7, bottom panel). All in all, the labor market data continue to paint a picture of an economy that is recovering but that still has a very large output gap. In other words, the economy is a long way from generating meaningful inflationary pressure. Appendix A: Buy What The Fed Is Buying The Fed rolled out a number of aggressive lending facilities on March 23. These facilities focused on different specific sectors of the US bond market. The fact that the Fed has decided to support some parts of the market and not others has caused some traditional bond market correlations to break down. It has also led us to adopt of a strategy of “Buy What The Fed Is Buying”. That is, we favor those sectors that offer attractive spreads and that benefit from Fed support. The below Table tracks the performance of different bond sectors since the March 23 announcement. We will use this to monitor bond market correlations and evaluate our strategy’s success. Table 3Performance Since March 23 Announcement Of Emergency Fed Facilities Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy/Global Fixed Income Strategy Special Report, “Beware The Bond-Bearish Blue Sweep”, dated October 20, 2020, available at usbs.bcaresearch.com 2 Blank cells in Table 2 mean that returns are the same as in the “Spot Curve Falls To 0%” scenario. 3 Please see BCA Daily Insights, “Slowing US Labor Market Momentum: Investors Should Not Have Been Surprised”, dated October 15, 2020, available at din.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification
Highlights Global Duration: US Treasury yields have started to creep higher and the move is likely to continue in the coming months regardless of who wins the White House. Reduce overall global duration exposure to below-benchmark, focused on the US. Country Allocation: Based on our view that US Treasury yields have more upside, we are making the following changes to our recommended country allocations in the government bond portion of our model bond portfolio: downgrading the US to underweight, downgrading higher-beta Canada and Australia to neutral, and raising lower-beta Germany, France, Japan and the UK to overweight. Treasury-Bund Spread: We introduce a new trade in our Tactical Overlay to capitalize on our expectation of higher US bond yields and a wider Treasury-Bund spread: selling 10-year Treasury futures versus buying 10-year German bund futures. Feature In a Special Report jointly published last week with our colleagues at BCA Research US Bond Strategy, we laid out the case for why US Treasury yields have bottomed and should now begin to drift higher.1 We reached that conclusion for two reasons: 1) there will be a major US fiscal stimulus after the upcoming US election, especially so if Joe Biden becomes president and the Democrats take the Senate; and 2) the Fed’s shift to Average Inflation Targeting in late August represented the point of maximum Fed dovishness. The investment conclusions were to reduce duration exposure, while also downgrading our recommended allocation to US government bonds to underweight. We also advised cutting exposure to non-US government bond markets with relatively higher sensitivity to changes in US bond yields, while increasing allocations to countries with a lower “yield beta” to US Treasuries (Table 1). Table 1Updated GFIS Model Bond Portfolio Recommended Positioning In this follow-up report, we will further discuss the implications of our changed view on US yields for non-US developed market government bonds. This includes specific adjustments to the recommended country allocations in our model bond portfolio, as well as a new tactical trade to profit from a move higher in US yields that will not to be matched in Europe. Our Recommended Overall Duration Stance: Now Below-Benchmark The case for a future cyclical bottoming of global yields has been building for the past few months, even as yields have remained range-bound at very low levels across the developed economies. Our Global Duration Indicator, comprised of economic sentiment measures and leading economic indicators, bottomed back in March and has soared sharply since then (Chart of the Week). Given the usual lead time between peaks and troughs of the Indicator and global bond yields - around nine months, on average – that suggests yields should bottom out sometime before year-end. Chart of the WeekA Cyclical, US-Led Bottoming Of Global Bond Yields Chart 2UST Yields About To Break Out? In the US, we now think we are past that point, as we discussed last week. The 10-year US Treasury yield has been drifting higher during the month of October and is now bumping up against its 200-day moving average of 0.83% (Chart 2). This is only the first such attempt at a trend breakout in yields, and such a move is unlikely prior to US Election Day - or, more accurately, “US Election Is Decided Day” which may not be November 3! The case for a future cyclical bottoming of global yields has been building for the past few months, even as yields have remained range-bound. Outside the US, however, momentum of bond yields and potential trend breakouts paint a more mixed picture. German and French bond yields remain stable and generally trendless, with Italian and Spanish yields continuing to grind lower. At the same time, yields in the UK, Canada and Australia have started to perk up but remain just below their 200-day moving averages. Bond yields have not responded to the sharp cyclical rebound across the developed world, with large gaps between elevated manufacturing PMIs and stagnant bond yields (Chart 3). Low inflation, ample spare economic capacity and dovish monetary policies are all playing a role, with bond markets not expecting an imminent inflation surge that could drive up yields and fuel expectations of tighter monetary policy. By way of contrast, China - where domestic services sectors have improved at a rapid pace from the COVID-19 recession and where the central bank is not running an overly accommodative monetary policy – has seen a more typical positive correlation between government bond yields and the rising manufacturing PMI over the past several months (Chart 4). This suggests that developed market bond yields can begin to normalize if the domestic services side of those economies emerges more forcefully from the lockdown-induced downturn. Chart 3A Wide Gap Between Growth & Yields Chart 4Are Chinese Yields Sending A Message? The news on that front is more optimistic in the US compared in Europe. The Markit services PMIs for the euro area and UK have all weakened over the past few months, with headline inflation rates flirting with deflation (Chart 5). Similar data in the US has trended in the opposite direction, with stronger US services activity with rising inflation. Chart 5Deflation Risks In Europe, Not The US The pickup in new COVID-19 cases, and the degree of the response by governments to contain it, has been far stronger in Europe and the UK than in the US on a population-adjusted basis (Chart 6). Lockdowns have become more widespread across Europe to contain the second larger wave of the virus. The recent softer services PMI data in the euro area and UK are a reflection of those greater economic restrictions and weaker confidence. This gap between the US economy and non-US economies is only magnified by the fiscal stimulus measures proposed by both US presidential candidates. In the US, governments have been far less willing to implement politically unpopular restrictions in an election year, while lockdown-weary consumers have been more willing to go about their lives rather than stay sheltered at home. The result is a healthier tone to the US data compared to other countries, even with the number of new US cases on the rise again. This gap between the US economy and non-US economies is only magnified by the fiscal stimulus measures proposed by both US presidential candidates. As we discussed in last week’s Special Report, both the Biden and Trump platforms are calling for major fiscal stimulus – between $5-6 trillion over the next decade, including tax changes – although the Biden plan has much more front-loaded direct government spending, only partially offset by tax increases, if fully implemented. This is the “Blue Sweep” scenario, with a Biden victory and Democratic Party control of the US Congress, that is most bearish for US Treasuries, as the outcome would eventually help reduce the expected 2021 US fiscal drag of -7.2% of GDP as estimated by the latest IMF Fiscal Monitor (Chart 7). Even a re-elected Trump, however, would also mean more US fiscal stimulus, although with a mix of tax cuts and spending increases. Chart 6The Latest COVID-19 Wave Is Hitting Europe Harder Combined with an improving services sector and rising inflation, this puts the US in a much different economic position than the major economies of Europe. Chart 7Post-Election US Stimulus Will Offset Fiscal Drag There, the IMF is also projecting some fiscal drag in 2021, but now with a much less healthy domestic economy due to the COVID-19 surge and where inflation is already near 0%. Our decision to reduce our recommended overall global duration stance to below-benchmark is largely driven by trends in the US that are more bond-bearish than in the rest of the developed world. There will likely be another round of fiscal measures to help combat virus-stricken economies in Europe and elsewhere, but the US election is bringing the issue to the forefront more quickly. In other words, the US will get a more bond-bearish fiscal stimulus before Europe does. Bottom Line: US Treasury yields have started to creep higher and the move is likely to continue in the coming months regardless of who wins the White House. Reduce overall global duration exposure to below-benchmark, focused on the US. Our Recommended Country Allocation: Downgrade US, Upgrade Lower-Beta Countries Net-net, our decision to reduce our recommended overall global duration stance to below-benchmark is largely driven by trends in the US that are more bond-bearish than in the rest of the developed world. This also has implications for our recommend country allocation in our model bond portfolio. First, are downgrading our recommended US Treasury allocation to underweight. We are also increasing our desired weighting in countries where government bond yields are less sensitive to changes in US Treasury yields – especially during periods when the latter are rising. We call this “upside yield beta”. The countries that have the highest such beta to US Treasuries are Canada, Australia and New Zealand, making them downgrade candidates (Chart 8). Similarly, lower upside beta countries like Germany, France, Japan and the UK are upgrade possibilities. Chart 8Favor Countries With Lower Yield Betas To USTs Already, we are seeing the widening of yield spreads between US Treasuries and non-US government markets – with more to come as US Treasuries grind higher over the next 6-12 months. We see the greatest upside for spreads between the US and the low upside yield beta countries – that means wider spreads for US-Germany, US-France, US-Japan and US-UK (Chart 9). Chart 9Expect More Underperformance From USTs Chart 10Fed QE Momentum Peaking, Unlike Other CBs Thus, this week are making significant changes to our strategic government bond country allocations (see page 15), as well as the country weightings in our model bond portfolio (see pages 13-14), based on our new view on US bond yields and non-US yield betas. Specifically, we are not only cutting our recommended US weighting to underweight, but we are also downgrading Canada and Australia from overweight to neutral. On the other side, we are upgrading UK Gilts to overweight from neutral, while also upgrading Germany, France and Japan to overweight. Importantly, we are maintaining our overweight stance on Italian and Spanish sovereign debt, as those markets are supported by greater European fiscal policy integration in the world of COVID-19 and, just as importantly, large-scale ECB asset purchases. More generally, the relative “aggressiveness” of central bank quantitative easing (QE) does play a role in our recommended country allocation. We expect the Fed to be more tolerant of higher Treasury yields if the move is driven by improving US growth and/or greater US fiscal stimulus – as long as the higher yields were not having a negative impact on equity or credit markets. We expect the Fed to be more tolerant of higher Treasury yields if the move is driven by improving US growth and/or greater US fiscal stimulus – as long as the higher yields were not having a negative impact on equity or credit markets. This means less expected QE buying of Treasuries by the Fed. Conversely, given how aggressive the Reserve Bank of Australia and Bank of Canada have been with expanding their balance sheet via QE (Chart 10), this makes us reluctant to shift to the underweight stance on those countries implied by their high beta to rising US Treasury yields. Therefore, we are only downgrading those two countries to neutral. Bottom Line: Based on our view that US Treasury yields have more upside, we are making the following changes to our recommended country allocations in the government bond portion of our model bond portfolio: downgrading the US to underweight, downgrading higher-beta Canada and Australia to neutral, and raising lower-beta Germany, France, Japan and the UK to overweight. A New Tactical Trade: A UST-Bund Spread Widener Using Futures This week, we are also introducing a new recommended trade in our Tactical Overlay portfolio on page 16 to take advantage of our view on US bond yields: a 10-year US-Germany spread widening trade using government bond futures. Chart 11A Tactical Opportunity For A Wider UST-Bund Spread This trade makes sense for several reasons: Germany has one of the lowest yield betas to US Treasuries during periods when the latter is rising, as shown earlier. Our US Treasury-German Bund fundamental fair value spread model – which uses relative policy interest rates, unemployment and inflation between the US and the euro area as inputs - suggests that the spread is now far too tight after the massive rally in US Treasuries in 2020 (Chart 11). The main reason why the spread looks so “expensive” is that the underlying fair value has risen with US inflation rising and euro area inflation falling (Chart 12, bottom panel). The UST-Bund yield differential is not stretched from a technical perspective, when looking at deviations of the spread from its 200-day moving average or the 26-week change in the spread; both measures suggest room for additional spread widening before reaching historical extremes (Chart 13). Also, duration positioning by US fixed income investors is only around neutral, according to the JP Morgan duration survey, suggesting scope to push yields higher if bond investors become more defensive. Chart 12Inflation Differentials Justify A Wider UST-Bund Spread Chart 13Technical Trends Favor A Wider UST-Bund Spread As a reference, we are initiating this trade with the cash bond 10-year US-Germany spread at +138bps, with a target range of +170-190bps over the 0-6 month horizon we maintain for our Tactical Overlay positions. Bottom Line: We introduce a new trade in our Tactical Overlay to capitalize on our expectation of higher US bond yields and a wider Treasury-Bund spread: selling 10-year Treasury futures versus buying 10-year German bund futures. Robert Robis, CFA Chief Fixed Income Strategist rrobis@bcaresearch.com Footnotes 1 Please see BCA Research US Bond Strategy Special Report, "Beware The Bond-Bearish Blue Sweep", dated October 20, 2020, available at usbs.bcaresearch.com and gfis.bcaresearch.com. Recommendations The GFIS Recommended Portfolio Vs. The Custom Benchmark Index Duration Regional Allocation Spread Product Tactical Trades Yields & Returns Global Bond Yields Historical Returns
Highlights We are upgrading Trump’s odds of winning to 45%. We have bet on a Democratic sweep all year. Incumbent parties rarely survive recessions, and President Trump has mishandled the pandemic. However, our updated quantitative election model – which relies heavily on short-term changes in the 50 states’ economies – points to a Trump victory with 279 Electoral College votes. The model puts Biden’s odds at 49%, i.e. “too close to call.” Opinion polls still favor Biden – and polls are generally accurate with sitting presidents. Yet Biden’s lead in swing states is comparable to Hillary Clinton’s in 2016. And we all know how that ended. Trump’s comeback, successful or not, will increase the chances of a contested election and will boost Republicans in Senate races. Our Senate model is also now flagging Republican control. The US fiscal policy outlook hinges on control of the Senate. Democrats would add 4%-7% of GDP to the fiscal thrust next year. We give 28% odds to a risk-off scenario, leaving a 72% chance that the policy setting will favor reflation. Feature We are upgrading President Trump’s odds of winning the US election from 35% to 45%. Looking at opinion polls, Biden is still favored as we go to press. But according to our quantitative election model, which relies heavily on the economy, Trump will eke out an Electoral College victory. What matters is that the media and financial markets are once again underrating Trump. The race is getting closer in the final days. Not only is our model flagging a Trump win, but the V-shaped economic recovery is boosting Trump’s popular support in the battleground states critical to winning an Electoral College majority. At very least investors should hedge their bets on former Vice President Joe Biden, who is not, after all, an extraordinarily charismatic challenger. Biden is not polling much better than Hillary Clinton polled against Trump four years ago (Chart 1). Chart 1ABiden Not Polling Much Better Than Clinton … Chart 1B… Against Trump The polling so far suggests that Trump suffered permanent damage from this year’s crisis and his support will hit a ceiling and relapse over the next week, confirming the month’s general tendency of a Biden win. But our confidence in the outcome is lower than before. The implication for investors is that the current volatility and risk-off sentiment could extend for one-to-three months, particularly given Congress’s failure to pass a new COVID relief package. However, beyond the near term, most scenarios are reflationary, positive for global equities and negative for low-yielding government bonds. There Are Still Undecided Voters Trump beat expectations in the final presidential debate on October 22, according to CNN polls. But debate performance does not accurately predict the winner of US elections. Moreover around 58 million voters have already voted based on prior information.1 Chart 2Still Enough Undecided Voters To Turn Election Still, Trump’s recovery in swing state polling is simultaneous with a lot of early voting in October, and there are enough undecided voters to change the outcome in critical swing states. About 6% of voters are undecided – virtually the same as in 2016. While the number of third-party supporters has fallen by 5.7 percentage points (ppt), this trend will not hurt Trump (Chart 2). In swing states in 2016, Libertarian Party voters outnumbered Green Party voters by a ratio of four to one, which does not suggest that these voters will all flock to Biden. They could even lean Trump. A large third party vote points to popular discontent, which hurts the incumbent party, as in 1980 and 1992. A lower third party vote is thus neutral for Trump. This is a major difference in 2020 from 1980 and 1992, which are the only two modern examples of a president losing after his first term. The best demographic projections have long shown that a rerun of the 2016 election, albeit with a normalization of the third-party vote share, would lead to an Electoral College tie. While it is virtually impossible for Trump to win the popular vote, he has a lifeline if state results are contested and/or the Electoral College is indecisive. Quant Model Gives Trump 279 Electoral College Votes Economic activity in the US continues to bounce back, according to flash PMIs in services and manufacturing as well as the latest data release from the Philadelphia Federal Reserve’s Coincident Economic Index. This index is the key input in our quantitative US election models for the White House and Senate, both of which now flag Republican victories. The latest reading pushes Trump’s odds of winning re-election up by 2ppt, to 51%, thus predicting that he will win with 279 Electoral College votes, an increase of 20 votes since our September update (Chart 3). Obviously this is not a high-confidence reading but rather an outcome that says the election is “too close to call.” Our model correctly predicts all election outcomes since 1984 during in-sample back testing, and all elections since 2000 on an out-of-sample basis. Chart 3Quant Model Points To Trump Victory … A Risk To Our View A Trump victory would be a massive upset – as in 2016. According to PredictIt.org, there is a 40% chance that Republicans will keep the White House. Other prominent forecasting groups, like FiveThirtyEight and The Economist, give Trump much lower odds, at 12% and 4%, respectively. In our model, Michigan has moved comfortably toward a Republican win (74% odds), opposite the conventional wisdom. Michigan is the crux of our subjective difference with our quantitative model – we don’t see a path for Trump to win as the polls currently stand. New Hampshire is the model’s only toss-up state, with a 53% chance of switching to Republicans, another surprising find, albeit one punctuated by President Trump’s decision to campaign in the state over the weekend. Pennsylvania and Wisconsin, states won by the Republicans in 2016, are still expected to flip to the Democrats. State-level coincident economic indices have the largest impact in determining the outcome in the model. Our other explanatory variables are state-by-state margins of victory in 2016, a “time for change” variable that favors incumbent parties, and the range of Trump’s approval rating. These variables have not changed recently and will not change in the final days of the election. Improving economic activity across the US is the basis for our model’s finding. Chart 4Improvements In Swing State Economies The Philly Fed data show that 48 out of 50 states’ coincident economic indices increased over the past three months, an increase by 10 states since the previous month’s release. All swing states rose, while the previous toss-up state, Michigan, turned positive, according to our weighting method, which takes the three-month changes in the economic indicators and weights the final months of useable data more heavily than previous months in an election year (Chart 4). Michigan and New Hampshire account for 20 electoral votes, raising Trump from 259 to 279. Clients have asked us why we use the range of President Trump’s approval rating rather than the level (Chart 5). We found this measure more statistically significant than other measures. If we manipulate the data we find that the model would still favor Trump if we looked at the two-year change of the approval rating or the October momentum of the approval rating. However, the model flags a Trump loss if we judge by the July or October level of his approval rating (which is historically low), or if we exclude the approval rating data altogether. The result of these alternate versions of our model is a Trump loss, with 246 Electoral College votes and Florida and Michigan remaining the critical toss-up states (Table 1). We are sticking with our original model, as the range of approval predicts electoral votes with a higher confidence level than other measures. Chart 5Trump’s Approval Range Is Narrow, Stable Table 1Variations In Quant Model Show Range Of Possibilities Bottom Line: Our quant model now favors Trump for re-election with 279 Electoral College votes. This economy-heavy model suggests that Trump is once again underrated, that the odds of a contested election are rising (in which Trump has some institutional advantages), and that Senate Republicans will benefit from the final sprint. Uncertainty and volatility will affect the market if the election result is indecisive, delayed, or if the GOP keeps the Senate (see below). Why We Do Not Favor Trump Outright Biden has been our pick since March based on the year’s huge external shock. The pandemic and recession have been harmful to the material wellbeing of the American public and therefore have sharply reduced the odds that the current president and ruling party will be re-elected. Looking at the level of Trump’s approval rating, he is comparable to George Bush Sr, who lost re-election in 1992 after a recession and race riots in Los Angeles. He is well beneath George Bush Jr and Barack Obama, who were re-elected handily in 2004 and 2012 (Chart 6). Chart 6Trump’s Approval Rating Level Is Relatively Low Joe Biden has a 7.9ppt lead in average national opinion polling. Looking at the breakdown across demographic groups reveals Trump’s serious liabilities. Biden has a 17ppt lead among women, compared to Clinton’s 15ppt lead in 2016 exit polls, and he is tied with Trump among men, compared to a 11ppt Trump lead in 2016 (Chart 7).2 Chart 7Trump Lagging In Key Demographic Groups Ethnic white voters still favor Trump by 5ppt but Trump has lost ground with this group since 2016, when he had a 15ppt advantage. Biden leads among voters who have some college education, while Trump’s lead among non-college graduates has fallen from 7ppt in 2016 to 4ppt this year. Chart 8Consumer Confidence Sounds Warning For President However, Black and Hispanic voters support Biden by a 74ppt and 31ppt margin, respectively, down from Clinton’s larger margins of 85ppt and 38ppt in 2016. While Trump is an exclusively commercial president, his approval rating never rose above 47% even when the economy was booming and consumer confidence soared. The collapse in consumer confidence has taken a toll on his approval, which struggles to break above 45% (Chart 8). Expectations have shot up, but voters are unhappy about current conditions. Consumer spending has not fully recovered and disposable income is in a freefall due to the failure of Congress to agree to a new fiscal relief deal since August, when benefits began to expire (Chart 9). Trump wanted a deal but so far Senate Republicans have proven unable to capitulate to House Democrats’ demands. Median family income has fallen over the course of Trump’s term. It spiked on the fiscal relief but then fell back when the latest phase of stimulus fell through (Chart 10). Chart 9Lack Of Fiscal Stimulus Weighs On Households Chart 10Median Income Down Over Four Year Term Under Trump’s watch the unemployment rate has risen from 4.7% to 7.9%. Obviously the surge was due to the pandemic and unemployment has fallen from its peak. But rising joblessness weighs on a president’s approval rating in the final reckoning – this is a good rule of thumb for identifying one-term presidents (Chart 11). Permanent unemployment is also rising, creating a group of unhappy voters that could make a difference in elections with thin margins. Chart 11AUnemployment Often Predicts … Chart 11B... The Election End-Game The pandemic is not over. COVID-19 hospitalizations and deaths are climbing in Arizona, Michigan, Pennsylvania, and Wisconsin (Chart 12). Trump’s net approval rating is deeply negative with regard to his handling of the crisis, as opposed to the economy where his approval is still net positive (Chart 13). Chart 12Pandemic Re-Emerging, Hurts Trump Chart 13Trump Ailing On Pandemic Handling Biden, a traditional Democrat, is an acceptable alternative to Trump. His lead over Trump is 7.9% in national polling and 4% in swing state polling. He polls considerably better than Hillary Clinton did. In Arizona his polling is rising; elsewhere it is flat (Chart 14A & B). Chart 14ABiden Polling Stable … Chart 14B… And Better Than Hillary Can the opinion polls be trusted? National polling is generally close to the mark – especially for incumbent presidents – and the winner of the national vote wins the Electoral College 91% of the time. Challengers who lose elections typically outperform their final polling by 1.4%. Those who win outperform by 3%. Whereas incumbent presidents who win outperform by 0.8% and those who lose outperform by 1% – i.e. they do better than expected but still lose (Chart 15A & B). Presidents are well known so they don’t tend to bring big surprises. However, there are major exceptions, namely Harry Truman. Chart 15AOpinion Polls Fairly Accurate On Sitting Presidents Chart 15BOpinion Polls More Often Underrate Challengers Chart 16Trump Is Rising In Battleground State Polls What about state level polls? The big errors in 2016 occurred on the state level. However, swing state pollsters have improved their methods. This can be confirmed by the fact that Trump’s performance in battleground opinion polls closely aligns with his job approval rating (Chart 16). The approval rating is the most reliable of all US political polls. The fact that these two are in lockstep, as against Trump’s national support rate (which is weighed down by dyed-in-the-wool Democrats in populous states), suggests that swing state polling is not wildly off the mark. On the other hand, Biden’s 4ppt lead is not very large. Voter turnout will be very high this year. Both Professor Michael McDonald from the US Elections Project and Nate Silver of FiveThirtyEight expect turnout to be around 65%. High political polarization, get-out-the-vote campaigns by both parties, and expanded access to mail-in voting due to the pandemic have created a high-turnout environment. High turnout does not necessarily disfavor Trump, given that his political base consists of many low turnout groups. But it should hurt him in the context of higher unemployment, as was the case for the incumbent party in 1992 and 2008 (Chart 17). Bottom Line: History suggests the incumbent party will lose the White House. So do opinion polls, which tend to be accurate when it comes to sitting presidents. Trump’s momentum has picked up in swing state opinion polls this month, though it is pausing as we go to press. If he gains momentum in the final week then he could still win the election. Chart 17AHigh Turnout Amid High Unemployment … Chart 17B…Hurts Incumbent Trump’s Path To Victory Biden’s 4ppt lead in swing states is within the range of polling error. A last-minute Trump comeback is a risk. While presidents usually lose re-election if they suffer a recession, especially during the year of the election, there have been exceptions – namely in 1900, 1904, and 1924. The basis for Trump to make a comeback is the economic snapback and the fact that voters consistently rate the economy as the most important issue in the election. The crisis struck early enough in the year that the massive fiscal relief package has propped up demand in what could be the nick of time for the president (Chart 18).3 Chart 18Trump’s Biggest Help Is V-Shaped Recovery Trump is generally polling better than he did in 2016 and his polling is ticking up in the final weeks of the race despite a disastrous year (Chart 19). His polling is improving in Florida and Arizona, meaning that a single victory in the upper Midwest would keep him in the White House. Chart 19ATrump Rallying In Some Swing States … Chart 19B… Critical Trend If It Continues Wage growth is also seeing a V-shaped recovery – particularly in the blue states, where services and knowledge-based sectors drive the economy, but also in “purple” swing states (Chart 20), though admittedly the purple states that voted for Trump are the laggards. The manufacturing sector is also bouncing back, which is critical for the Midwestern Rust Belt that got pummeled by Trump’s trade war prior to the pandemic. The surge in credit fueled by the Federal Reserve’s liquidity provisions is a positive for this region (Chart 21). Chart 20Swing State Wage Growth Bounces Back Chart 21Midwestern Economy Snaps Back The stock market rally is also positive for the incumbent. The S&P 500 predicts the election result 77% of the time going back to 1896. Specifically, its year-to-date performance as of October 31 of an election year is positively correlated with an incumbent party’s likelihood of winning the White House and is statistically significant at the 5% confidence level. Back in May, with the S&P down 13%, the stock market gave Trump a 16% chance of re-election. Today, up 6% YTD, it gives him a 66% chance (Chart 22). Chart 22Simple Stock Market Model Says Trump Favored For Re-Election We would not put too much emphasis on this measure, as the market also rallied prior to Carter’s and Bush’s losses in 1980 and 1992. But Trump is uniquely tied to the stock market and it is clearly good for him if the market does not collapse (though the failure to pass fiscal stimulus is a liability). Simply put, Trump is stronger than Mitt Romney 2012 and Biden is weaker than Barack Obama. The 3.9ppt margin of victory in the popular vote that year should be narrower this year. Run-of-the-mill Democrats have not received more than 49% of the popular vote in recent memory. And that was the popular Bill Clinton in 1996 (Table 2). If Trump clocks in at 46%, as in 2016, then he could squeak through the Electoral College once again. Bottom Line: We are upgrading Trump’s odds to 45%. Table 2US Presidential Election Popular Vote The Senate Is Too Close To Call Even if Trump’s comeback is “too little, too late,” it increases the chance of a contested election – in which he could get a lifeline through the Supreme Court or the House of Representatives – and also gives a boost to Republican Senators in tight races. Our Senate election model, like our presidential model, uses the Philly Fed coincident economic indicators. It has also flipped from favoring Democrats to narrowly predicting Republican control, with 51-49 seats. Specifically, Montana and North Carolina shifted into the Republican camp, though North Carolina remains a toss-up and would turn the overall balance of power (Chart 23).4 Chart 23Quant Model Says Senate Favors Republicans – A Risk To Reflation Trade Again the proper way of interpreting this reading is that the election is “too close to call,” with a 49% chance of Democratic control. Notably our Senate model relies more heavily on opinion polling than our presidential model – it incorporates the president’s approval rating level as well as the incumbent party’s net support rate in the generic congressional ballot (a poll that measures which party voters generally prefer for Congress). The economic recovery is the source of the boost for Republicans but marginal improvements in Republican polling do not hurt. The Senate race is critical to the overall policy significance of the US election. You cannot pass major legislation in the US without control of the Senate. And the Senate races are clearly tightening. This means uncertainty is rising, not falling, as the election approaches. Position For Reflation, The Likeliest Policy Outcome In particular the US fiscal outlook depends on the Senate. Chart 24 simulates the different courses of the deficit depending on election scenarios. If the Democrats win the White House, Senate, and House of Representatives, the budget deficit will rise from 16% of GDP in FY 2020 to 23% of GDP in FY 2021, as Biden will largely execute his policy agenda. Chart 24Democratic Sweep Offers Massive Fiscal Boost If Trump and the Republicans win the White House and retain the Senate, they will keep cutting deals with House Democrats as in recent years, and the deficit will at least remain flat. The only scenario in which the budget deficit contracts – i.e. a negative fiscal thrust threatens the US economic recovery – occurs if Biden wins the White House but Republicans obstruct his agenda. Realistically, this would result in something like the Republican status quo scenario in Chart 24 above, rather than the Congressional Budget Office’s baseline scenario. The baseline scenario would produce an intolerable 7.4% contraction in fiscal thrust under baseline scenario in 2021. GOP senators would not go so far. They are not the same as the House Freedom Caucus members who were so hawkish in 2010-16. Nevertheless investors cannot rule out the baseline scenario – which could cause a double-dip recession – until GOP senators allay their fears. The market will cheer if President Trump and the Republicans retain the White House and Senate, as the fiscal thrust will be neutral or slightly expansive. It will especially cheer if the Democrats win a clean sweep, adding anywhere from 4%-7% of GDP in fiscal thrust for FY 2021 – the most reflationary outcome. It will even cheer in the odd chance that Trump wins with a unified Democratic Congress, which would also be reflationary. But the market will not cheer if the election threatens premature fiscal tightening via Republican obstructionism under a Biden presidency. This is the only scenario that is deflationary. The market would have to riot to force Republican senators to cooperate with a Democratic president – and this would be the case in the lame duck session as well as for each new stimulus package and budget over 2021-22. Based on our updated quant models, this Biden+GOP scenario is about a 28% probability, a slight increase from our previous view. The flip side is that there is about a 72% probability of a reflationary outcome. Beyond the near term, a Biden presidency with a Republican senate is actually market positive. Republican senators would eventually have to agree to House-drawn budgets, but would prevent tax hikes and legislative overreach (the downside of a Democratic sweep). Meanwhile a President Biden would avoid sweeping unilateral tariffs against China and the EU (the downside of any Trump victory). Bottom Line: A Democratic sweep is the most fiscally proactive scenario but the odds have fallen from around 45% to 27% according to our quant models. The odds of Biden plus a GOP Senate have risen from 20% to 28%. The market would have to digest significant new fiscal risks in this case, so the dollar and US treasuries would initially rally. The other scenarios combine to a 72% probability and are initially reflationary, albeit less so than a Democratic sweep, with the likelihood of massive trade war risk in 2021. Trade Recommendations Courtesy Of The BCA Equity Analyzer As the US election approaches and the effects of the global pandemic linger, economic policy uncertainty remains elevated. Equity markets tend to behave very differently in times of acute uncertainty. In order to gauge the effects of uncertainty at the individual stock level, we turn to BCA’s stock-picking engine, the Equity Analyzer. We looked at factor performance when economic policy uncertainty (as defined by Baker, Bloom and Davis) exceeds the 150-line (Chart 25). This is quite high compared to history. Chart 25Policy Uncertainty: How High Will It Go? We look at the 30 factors included in the BCA Equity Analyzer and examine the Sharpe Ratio (Chart 26). The Sharpe ratio expresses the risk adjusted performance of long/short strategies based on each factor. Long/short strategies, in turn, are defined as going long the top 25% based on a factor and going short the bottom 25%. Chart 26Equity Analyzer Shows Key Traits For Navigating Uncertainty The results show that the best performing factors in times of high uncertainty are: Relative earnings yield Low accruals5 BCA Style, which is an in-house combined measure for (1) value versus growth and (2) small caps versus large caps. One-month momentum With these results, we go back to the BCA Equity Analyzer to extract the top 25 stocks filtered by our top 4 factors during times of uncertainty. The results are shown in Table 3.6 The BCA score in this table ranges from 0 to 100% (from a strong sell to a strong buy). It is based on 30 factors distributed among seven broad categories: Macro, Value, Safety, Sentiment, Technical, Quality, Payout. These picks will improve performance during the upcoming spike in uncertainty, which is now even more likely than it was given the rising odds of a contested election and/or deflationary partisan gridlock. Table 3BCA Equity Analyzer Stock Picks For Election Uncertainty Matt Gertken Vice President Geopolitical Strategy mattg@bcaresearch.com Guy Russell Research Analyst GuyR@bcaresearch.com David Boucher Chief Quantitative Strategist DavidB@bcaresearch.com Footnotes 1 See Lauren King and Lauren Lantry, “More than 58 million Americans have already voted,” ABC News, October 25, 2020, abcnews.go.com. 2 See “An examination of the 2016 electorate, based on validated voters,” Pew Research Center, August 9, 2018, pewresearch.org. 3 Back in April, former Obama economic adviser Jason Furman predicted that the likely timing of the economic snapback would be very favorable for President Trump: “We’re about to see the best economic data we’ve seen in the history of this country,” he said. See Ryan Lizza and Daniel Lippman, “The general election scenario that Democrats are dreading,” Politico, May 26, 2020, politico.com. 4 Given the Senate’s critical importance to US fiscal policy, we weren’t joking when we said that Democratic candidate Cal Cunningham’s sex scandal in North Carolina could end up making the difference for the election’s overall consequences as well as the global macro outlook! 5 Accounting accruals are the non-cash component of a firm’s earnings and can be used as a metric to gauge the overall quality of a firm. Firms with high accruals tend to underperform firms with low accruals because of the potential to mask poor performance through the practice of accrual accounting. 6 Screener is based on US exchanges, top 30% based on market cap, Relative Earning Yield Score above 50%, Accruals Score above 50%, 1-month Momentum above 50%, and BCA Style above 50%.
Your feedback is important to us. Please take our client survey today. Highlights For now, there is little evidence that the pandemic has adversely affected the global economy’s long-run growth potential. Even if one counts those who will be unable to work due to long-term health complications from the virus, the pandemic will probably reduce the global labor force by only 0.1%-to-0.15%. Labor markets have healed more quickly over the past few months than after the Great Recession. In the US, the ratio of unemployed workers-to-job openings has recovered most of its lost ground. Thanks in part to generous government support for businesses and the broader economy, commercial bankruptcy filings remain near historic lows. Meanwhile, new US business formation has surged to record highs. The combination of a vaccine and a decline in rents in city centres should persuade some people who were thinking of fleeing to the suburbs to stay put. This will ensure that most urban commercial and residential real estate remains productively engaged. Judging from corporate surveys, capital spending on equipment and intellectual property should continue to rebound. While the pandemic has caused numerous economic dislocations, it has also opened the door to a variety of productivity-enhancing innovations. An open question is whether all the debt that governments have taken on to alleviate the economic damage from the pandemic could in and of itself cause damage down the road. As long as interest rates stay low, this is not a major risk. However, today’s high government debt levels could become a problem if the pool of global savings dries up. Investors should continue to overweight stocks for the time being, while shifting their equity exposure from “pandemic plays” to “reopening plays.” A more cautious stance towards stocks may be appropriate later this decade. The Pandemic’s Potentially Long Shadow In its latest World Economic Outlook, the IMF revised up its growth estimates for this year. Rather than contracting by 4.9%, as it expected in June, the Fund now sees the global economy shrinking by 4.4%. That said, the IMF’s estimates still leave global GDP in 2020 7.5% below where it projected it to be in January. Perhaps even more worrying, the IMF expects the global economy to suffer permanent damage from the pandemic (Chart 1 and Chart 2). It projects that real global GDP will be 5.3% lower in 2024 compared to what it expected last year. In the G7, real GDP is projected to be nearly 3% lower, with most of the shortfall resulting from a downward revision to the level of potential GDP (Chart 3). Chart 1Covid-19: The IMF Expects The Global Economy To Suffer Permanent Damage (Part I) Chart 2Covid-19: The IMF Expects The Global Economy To Suffer Permanent Damage (Part II) The Congressional Budget Office is no less gloomy in its forecast. The CBO expects US real GDP to be 3.7% lower in 2024 than it projected last August. By 2029, it sees US GDP as being 1.8% below what it had expected prior to the pandemic, almost entirely due to slower potential GDP growth (Chart 4). Chart 3G7 Real GDP Growth Projections Have Been Revised Sharply Lower Due To The Pandemic Chart 4A Gloomy Forecast For The US Thanks To Covid-19 The worry that the pandemic will lead to a major permanent loss in output is understandable. That is precisely what happened after the Global Financial Crisis. Nevertheless, as we discuss below, there are good reasons to think that the damage will not be as pervasive as widely believed. The Drivers Of Potential GDP An economy’s potential output is a function of three variables: 1) the number of workers available; 2) the amount of capital those workers have at their disposal; and 3) the efficiency with which this labor and capital can be transformed into output, a concept economists call “total factor productivity.” Let us consider how the pandemic has affected all three variables. The Impact Of The Pandemic On The Labor Market At last count, the pandemic has killed over 1.1 million people worldwide, 222,000 in the US. While the human cost of the virus is immense, the economic cost has been mitigated by the fact that about four-fifths of fatalities have been among those over the age of 65 (Table 1). In the US, less than 7% of the labor force is older than 65. A reasonable estimate is that Covid deaths have reduced the US labor force by 55,000.1 Table 1Pandemic-Related Deaths Are Tilted Towards The Elderly, Who Are The Least Active Participants Of The Labor Force Chart 5The Number Of New Cases Continues To Increase Globally Granted, mortality is not the only way that the disease can impair one’s ability to work. As David Cutler and Larry Summers point out in a recent study, for every single person who dies from Covid-19, seven people will survive but not before manifesting severe or critical symptoms of the disease.2 Based on the experience from past coronavirus epidemics, Ahmed, Patel, Greenwood et al. estimate that about one-third of these survivors will suffer long-term health complications.3 If one assumes that half of these chronically ill survivors are unable to work, this would reduce the US labor force by an additional 65,000.4 Of course, the pandemic is not yet over. The number of new cases continues to rise in the US and globally (Chart 5). The only saving grace is that mortality and morbidity rates are lower than they were earlier this year. Nevertheless, many more people are likely to die or suffer debilitating long-term consequences before a vaccine becomes widely available. Using the US as an example, if the total number of people who end up dying or getting so sick that they are unable to work ends up being twice what it is so far, the pandemic will reduce the labor force by about 240,000. This is not a small number in absolute terms. However, it is less than 0.15% of the overall size of the US labor force, which stood at 164 million on the eve of the pandemic. The impact of the pandemic on the labor forces of other major economies such as Europe, China, and Japan will be even smaller. Labor Market Hysteresis People can drop out of the labor force even if they do not get sick. In fact, 4.4 million have left the US labor force since February, bringing the participation rate down from 63.4% to 61.4%. How great is the risk of “hysteresis,” a situation where the skills of laid-off workers atrophy so much that they become unwilling or unable to rejoin the labor force? At least so far, hysteresis has been limited. According to surveys conducted by the Bureau of Labor Statistics, most US workers who have dropped out of the labor force still want a job. The pandemic has made it more difficult for people to work even when they wanted to. During the spring, more than four times as many employees were absent from work due to childcare requirements than at the same time last year. Now that schools are reopening, it will be easier for parents to go back to work. Admittedly, not everyone will have a job to return to. While about a third of US unemployed workers are still on temporary layoff, the number of workers who have suffered permanent job losses has been steadily rising (Chart 6). The good news is that job openings have recovered most of their decline since the start of the year. Unlike in mid-2009, when there were 6.5 unemployed workers for every one job vacancy, today there are only two (Chart 7). Chart 6US: Permanent Job Losses Have Been Rising Steadily... Chart 7...But Job Openings Have Recovered Most Of Their Decline Since The Start Of The Year It is also worth noting that the vast majority of job losses during the pandemic has been among lower-income workers, especially in the retail and hospitality sectors. Most of these jobs do not require highly specialized sector-specific skills. Thus, as long as there is enough demand throughout the economy, unemployed workers will be able to find jobs in other industries. Wither The Capital Stock? The pandemic may end up reducing the value of the capital stock in two ways. First, it could render a portion of the existing capital stock unusable. Second, the pandemic could reduce the pace of new investment, leading to a smaller future capital stock than would otherwise have been the case. Let us explore both possibilities. On the first point, it is certainly true that the pandemic has left a lot of the capital stock idle, ranging from office buildings to shopping malls. However, this could turn out to be a temporary effect. Consider, for example, the case of China. After the pandemic began in Wuhan, China first shut down much of its domestic economy and then implemented an effective mass testing and contact tracing system. The strategy worked insofar as China is now nearly free of the virus. Today, few Chinese wear masks, the restaurants are full again, and domestic air travel is back to last year’s level. Even movie theatre revenue has rebounded. The rest of the world may not be able to replicate China’s success in combating the virus, but then again it won’t need to if an effective vaccine becomes available. Chart 8US Housing Is In A Good Place Even if the pandemic ends up leading to deep and lasting changes in the way people live, work, and shop, the market mechanism will ensure that all but the least desirable parts of the capital stock remain productively employed. As first year economics students learn, if the supply curve is vertical and the demand curve shifts inward, the result will be lower prices rather than diminished output. By the same token, if more companies and workers decide to relocate to the suburbs, urban rents will fall until enough people decide that they are better off staying put. An economy’s productive capacity does not change just because rents go down. What falling demand for urban real estate and increased interest in working from home will do is encourage people to buy larger homes in suburban areas. We have already seen this play out this year. Despite flagging commercial real estate construction in the US, residential construction has boomed. Single-family housing starts were up 24% year-over-year in September. Building permits and home sales have reached new cycle highs. Homebuilder confidence hit a new record in October (Chart 8). The Service Sector Is Not Particularly Capital Intensive Most recessions take a greater toll on the goods-producing sectors of the economy than the service sector. The pandemic, in contrast, has mainly afflicted services. The service sector is the least capital-intensive sector of the economy. This is especially the case when it comes to spending on capital equipment and investment in intellectual property (Chart 9). Chart 9Capex-Intensive Industries Have Let Go Of Less Workers During The Pandemic Chart 10Capex Intentions Have Bounced Back As such, it is not surprising that investment in equipment and IP fell less during this recession than one would have expected based on the historic relationship between investment and GDP growth. According to the Atlanta Fed’s GDPNow model, investment in equipment and IP is set to increase by 23% in the third quarter. The snapback in the Fed’s capex intention surveys suggests that investment spending should continue to rise in the fourth quarter and into next year (Chart 10). Productivity And The Pandemic Just as the impact of the pandemic on the labor supply and the capital stock is likely to be limited, the same is true for the efficiency with which capital and labor is transformed into output. For every person whose productivity is hampered by having to work from home, there is another person who feels liberated from the need to spend an hour commuting to work only to attend a series of pointless meetings. In fact, it is quite possible that the pandemic will nudge society from various “low productivity” equilibria to “high productivity” equilibria. For example, greater use of video conferencing could negate the need to take redeye flights to attend business meetings in person. Remote learning could enhance educational opportunities. More widespread use of telemedicine could eliminate the need to waste time waiting in a doctor’s office. Who knows, the pandemic could even fulfill my life-long mission to replace the unhygienic handshake with the much more elegant Thai wai. Granted, disruptive shifts could produce unintended consequences. There is a fine line between creative destruction and uncreative obliteration. If the pandemic forces otherwise viable businesses to close, this could adversely affect resource allocation. Chart 11New Business Applications Have Surged To Record Highs Chart 12Commercial Bankruptcy Filings Remain In Check Fortunately, at least so far, this does not seem to be happening on a large scale. After dropping by 25%, the number of active US small businesses has rebounded to last year’s levels. New business applications have surged to record highs (Chart 11). According to the American Bankruptcy Institute, commercial bankruptcy filings remain near historic lows. While Bloomberg’s count of large-company bankruptcies did spike earlier this year, it has been coming down more recently (Chart 12). Fiscal Stimulus To The Rescue Chart 13Personal Income Jumped Early On In The Pandemic How did so many households and businesses manage to avoid the financial suffering that usually goes along with deep recessions? The answer is that governments provided them with ample income support. In the US, real personal income rose by 11% in the first few months of the pandemic (Chart 13). Small businesses also benefited from the Paycheck Protection Program, which doled out low-cost loans to businesses which they will be able to convert into grants upon confirmation that the money was used to preserve jobs. Similar schemes, such as Germany’s Corona-Schutzschild, Canada’s Emergency Business Account program, and the UK’s Coronavirus Job Retention Scheme were launched elsewhere. The failure of the US Congress to pass a new stimulus bill could undermine the sanguine narrative presented above. Small businesses, in particular, are facing a one-two punch from the expiration of the Paycheck Protection Program and tighter bank lending standards. Ultimately, we think the US Congress will pass a new pandemic relief bill. However, the size of the bill could depend on the outcome of the election. In a blue sweep scenario, the Biden administration will push through a $2.5-to-$3.5 trillion stimulus package early next year, while laying the groundwork for a further 3% of GDP increase in government spending on infrastructure, health care, education, housing, and the environment. A fairly large stimulus bill could also emerge if President Trump manages to hang on to the White House, while the Democrats take control of the Senate. Unlike some Republican senators, Donald Trump is not averse to big increases in government spending. A continuation of the current political configuration in Washington would result in the smallest increase in spending. Nevertheless, some sort of deal is likely to emerge after the election. Even most Republican voters favor a large stimulus bill (Table 2). Table 2Strong Support For Stimulus A Double-Edged Sword? Bountiful fiscal support has undoubtedly lessened the economic scarring from the pandemic. However, could the resulting increase in government debt lead to supply-side problems down the road? The answer depends on what happens to interest rates. As long as interest rates stay below the growth rate of the economy, governments will not need to raise taxes to pay for pandemic relief. In fact, in such a setting, the public debt-to-GDP ratio will return to its original level with absolutely no change in the structural budget deficit (Chart 14). GDP growth in most developed economies has exceeded government borrowing rates for much of the post-war era (Chart 15). Thus, a free lunch scenario where governments never have to pay back the additional debt they incurred for pandemic relief cannot be ruled out. That said, it would not be prudent to bank on such an outcome. If the excess private-sector savings that have kept down borrowing costs run out, interest rates could rise. In a world awash in debt, this could lead to major problems. Thus, while the structural damage to the global economy from the pandemic appears to be limited for now, that could change in the future. Chart 14A Fiscal Free Lunch When r Is Less Than g Chart 15The Rate Of Economic Growth Has Usually Been Higher Than Interest Rates Investors should continue to overweight equities for the time being. With a vaccine on the horizon, it makes sense to shift from favoring “pandemic plays” such as tech and health care stocks to favoring “reopening plays” such as deep cyclicals and banks. A more cautious stance towards stocks will be appropriate later this decade if, as flagged above, a stagflationary environment leads to higher interest rates and slower growth. Peter Berezin Chief Global Strategist peterb@bcaresearch.com Footnotes 1 To estimate the direct impact of Covid-19 on the labor force, we calculate the decline in the labor force by age cohorts using Covid-19 death statistics and labor participation rates. 2 David M. Cutler, and Lawrence H. Summers, “The COVID-19 Pandemic and the $16 Trillion Virus,” JAMA Network, October 12, 2020. 3 Hassaan Ahmed, Kajal Patel, Darren Greenwood, Stephen Halpin, Penny Lewthwaite, Abayomi Salawu, Lorna Eyre, Andrew Breen, Rory O’Connor, Anthony Jones, and Manoj Sivan. “Long-Term Clinical Outcomes In Survivors Of Coronavirus Outbreaks After Hospitalisation Or ICU Admission: A Systematic Review And Meta-Analysis Of Follow-Up Studies,” medRxiv, April 22, 2020. 4 Calculated as 0.5 x (decline in labor force due to Covid-19 deaths) x 7 x (1/3). Global Investment Strategy View Matrix Current MacroQuant Model Scores
Highlights Our model suggests that more rate hikes are ahead in 2021; we project a less than 50bps increase in the PBoC policy rate from the current level. Chinese stock prices positively correlate with interest rates and bond yields. The relationship has strengthened since 2015. In the next six to nine months, Chinese stock prices will likely trend up alongside a rising policy rate and an accelerating economic growth. Feature China’s policy rate and bond yields have been rising sharply since May and are breaching their pre-COVID 19 levels. Meanwhile, Chinese stock prices have moved sideways since mid-July, despite a steady recovery in the domestic economy. While some commentators view higher interest rates as a harbinger of an impending equity market weakness, our research shows that the relationship between China’s stock prices and short-term rates has been positive since 2015. A rally in Chinese stocks and outperformance of cyclical stocks relative to defensives positively correlate with rising interest rates and bond yields (Chart 1A and 1B). Chart 1ARising Bond Yields Coincide With Ascending Chinese Stock Prices... Chart 1B...And Offshore Cyclicals Chart 2Massive Stimulus In 2020 Will Accelerate Economic Growth Into 1H21 China’s massive stimulus this year generated some self-sustaining momentum that will likely push the nation’s output higher in 1H21(Chart 2). The PBoC may raise the policy rate by as much as 50bps in 2021 from its current level, but strong domestic fundamentals should be able to drive up Chinese stock prices, in both absolute term and relative to global equities in the next six to nine months. PBoC Policy Hikes:Still More Ahead While the PBoC’s policy rate has rebounded sharply, it remains at its lowest level since the Global Financial Crisis. Looking forward, will the central bank bring the policy rate (e.g. 3-month SHIBOR) back to its pre-COVID 19 range of 3 – 3.5% or the pre-trade war level near 5%? The acceleration in China’s economic recovery is expected to continue and would boost China’s annual output growth in 1H21 to two to three percentage points above its trend. Based on these estimates, our interest rate model implies more than 200bps in rate increases in 2021 from the current level1 (Chart 3). Chart 3Rising Odds Of PBoC Rate Hikes In 2021 Historically, our model has successfully captured the major turning points in China's policy rate cycles. This time around, however, the pandemic and the subsequent economic recovery may have complicated the model's predictive power. The model suggests that, in 1H21 the policy rate will return to its pre-trade war range of 4-5%, but we think the rate increases will be capped within 50bps. The model follows a modified version of "Taylor's Rule," in which we assume that the PBoC will target its short-term interest rate based on the deviation between actual and desired inflation rates and the deviation between real GDP growth and China’s trend GDP growth rate. The latest data shows across-the-board strengthening in the economy; most indicators have surprised to the upside, confirming our optimistic assessment.2 However, Taylor's Rule is not able to account for sudden shocks in the economy, such as a pandemic-induced global recession. Thus, the model exaggerates the magnitude of interest rate bumps, based on an economic growth acceleration following a one-off economic shock. In a report earlier this year, we noted that the PBoC has been proactive in normalizing its monetary policy following short-term shocks.3 This is contrary to economic downturns when the PBoC has been a reactive central bank and its decisions often lagged a pickup in economic activity. As such, although interest rates have swiftly rebounded after the pandemic-induced growth contraction in Q1, we expect the pace of rate hikes to be slower in 2021. Chart 4Rapid RMB Appreciation Will Bring Headwinds To Chinese Industrial Profits External factors are accounted for in the model, though they may be underestimated. The US Federal Reserve Bank has decisively shifted its monetary policy to broadly accommodative and will stay behind the inflation curve in the next few years. The collapse in interest rate differentials between the US and China has made RMB-denominated assets attractive, boosting strong inflows of foreign capital and rapidly pushing up the value of the RMB (Chart 4, top panel). While we think Chinese policymakers have pivoted to prefer a strong RMB, the recent countermeasures by the PBoC indicate that the central bank will not allow the RMB to climb too rapidly.4 China's drastic tightening in monetary conditions and the sharp rally in the trade-weighted RMB from 2011 to 2014 led to a prolonged economic downturn (Chart 4, bottom panel). Therefore, in the absence of synchronized policy tightening from other central banks, the magnitude of rate hikes by the PBoC will be measured. Bottom Line: The PBoC will continue to push up the policy rate in 2021, but our baseline view is that the magnitude will be capped below 50bps. Interest Rates And Chinese Stocks Chart 5Chinese Stocks/Bond Yields Correlation Became Much More Positive After 2015 Many investors might think that stock prices tend to react negatively to monetary policy tightening because interest rate upturns and mounting bond yields lead to higher costs of funding for corporations and lower profit growth. However, Chinese stock prices started moving in the same direction with policy rates and bond yields following the burst of the 2014/15 stock market bubble (Chart 5 and Chart 1A and 1B on Page 4 and 2). In general, when China’s economic and profit growth accelerates, share prices can rise with higher interest rates. Share prices can still climb with cuts in interest rates even when economic growth slows but profit growth rate remains in positive territory. However, when profit growth is expected to drop below zero, share prices will drop even if rates are falling (Chart 6A and 6B). In this vein, the most pertinent reason for Chinese stocks to move in tandem with bond yields is that Chinese stocks are increasingly driven by economic fundamentals, which are supported by the volume of total credit creation (measured by total social financing) rather than the price of money in China. Furthermore, the reverse relationship between the volume and price of money in China broke down after 2015; China’s credit creation has become less sensitive to changes in interest rates. Chart 6AWhen Interest Rates Rise... Chart 6B...Economic Growth Holds The Key For Stock Performance Since 2015, the PBOC shifted its policy to target interest rates instead of the quantity of money supply (Chart 7). In order to effectively manage the official interbank rates (the 7-day interbank repo rate), the central bank uses tools such as reserve requirement ratio cuts and liquidity injections in the interbank system (Chart 8). In other words, the central bank has forgone its control of the volume of money. Moreover, since late 2016, rather than direct interest rate hikes, the PBoC has been taking monetary policy tightening measures through changes in its macro-prudential assessment (MPA). The changes in the MPA are evident in the 3-month / 1-week repo spread.5 As such, an increase in the 3-month interbank repo rate (and SHIBOR) is often intended to curb shadow-banking activities rather than depress aggregate credit creation and business activities (Chart 9). Chart 7Monetary Policy Regime Shifted In 2015 Chart 8More Open Market Operations Chart 9Most Monetary Tightening Has Been Carried Out Through MPA Since 2016 Another idiosyncrasy is China’s fiscal stimulus, which has become a more relevant driver of total social financing since the onset of the 2014/15 economic downcycle (Chart 10). The amount of government bond issuance is specified by the People’s Congress in March each year and is not affected by changes in interest rates or bond yields. Therefore, growth in total social financing can still accelerate despite a higher price of money (Chart 11). Chart 10Fiscal Lever Has Become More Prominent In Driving Business Cycles Since 2015 Chart 11Changes In Interest Rates Have Little Impact On Fiscal And Quasi-Fiscal Borrowing By the same token, a rising 3-month SHIBOR can also be the result of rapid fiscal and quasi-fiscal expansions, as seen in Q3 this year. A flood of central and local government bond issuance drained liquidities from commercial banks, boosting the banks’ needs to borrow money from the interbank system. Nevertheless, the market’s appetite for risk assets increases because fiscal stimulus provides an imminent and powerful reflationary force in China’s business cycles. Chart 12Bank Lending Rates Can Still Trend Downwards Against A Rising Policy Rate Rising policy rates typically push up corporate bond yields. However, bond yields in China play a relatively small role in driving corporate financing costs on an aggregate level, since commercial banks are still dominant in China’s debt market. Commercial banks' average lending rates closely track the PBoC’s policy rate on a cyclical basis, but Chinese authorities periodically use window guidance to target the Loan Prime Rate (LPR), a reformed bank lending rate. Hence, the direction in both the LPR and the average lending rate can temporarily diverge from the policy rate. These measures can boost bank loan growth even in a rising interest rate environment (Chart 12). Bottom Line: The key driver of Chinese stock performance is the country’s domestic credit, business, and corporate profit growth cycles. Since the 2014/15 cycle, the policy rate has not been the determinant of China’s economic or credit growth. Investment Conclusions We expect that this year’s massive monetary and fiscal stimulus to accelerate the country’s economic recovery into 1H21. Therefore, even if interest rates and bond yields advance, Chinese stock prices can still trend upward. Chinese cyclical stocks should also continue to outperform defensives, in both the onshore and offshore markets (Chart 13A and 13B). Chart 13AStay Invested In Chinese Stocks Chart 13BCyclicals Still Have Upside Potentials Rates will begin to climb and fiscal policy will also become more restrictive if China’s output moves above trend growth through 1H21. Government bond quotas and fiscal budget will be determined at the National People’s Congress in March. If the economy is strong, odds are that fiscal stimulus will be scaled back. At that point, investors should start to look for a peak in China’s business cycle linked to monetary and fiscal policy tightening. As growth expectations start to downshift in the equity market, yields on long-dated government bonds will start to decline while yields on the short end will not drop. Additionally, the small-cap ChiNext market has been considered as a speculative segment of the domestic financial market with higher multiples and greater volatility than large-cap A shares. The bourse's trailing price-to-earnings ratio and price-to-book ratio are extremely elevated at 79 and 8.6, respectively, much higher than for broader onshore and offshore Chinese stocks. As such, this market will remain the most vulnerable to domestic liquidity tightening. Jing Sima China Strategist jings@bcaresearch.com Footnotes 1 based on our estimates for 1h21: 7.5-8.0% GDP growth, 2.5-2.8% headline CPI, 6.5-6.7 USD/CNY, and the fed holding current fund rate unchanged. 2Please see China Investment Strategy Weekly Report "China Macro And Market Review," dated October 7, 2020, available at cis.bcaresearch.com 3Please see China Investment Strategy Weekly Report "Don’t Chase China’s Bond Yields Lower," dated February 19, 2020, available at cis.bcaresearch.com 4On October 12, the PBoC removed financial institutions’ Forex reserve ratio of 20%, making betting against the RMB cheaper. 5Please see China Investment Strategy Special Report "Seven Questions About Chinese Monetary Policy," dated February 22, 2018, available at cis.bcaresearch.com Cyclical Investment Stance Equity Sector Recommendations
Highlights Duration: Prospects for more pre-election fiscal stimulus are slim. But with the Democrats gaining ground in the polls, the bond market will stay focused on rising odds of a blue sweep election and greater fiscal stimulus in early 2021. Municipal Bonds: Municipal bonds offer exceptional value relative to both US Treasuries and corporate credit. Not only that, but rising odds of a blue sweep election make state & local government fiscal relief increasingly likely. Investors should overweight municipal bonds in US fixed income portfolios. Economy: The economic recovery continues to roll on, but it will be some time before the output gap is closed and inflation starts to rise. Slow consumer and corporate credit growth suggest that animal spirits have not yet taken hold. Meanwhile, the falling unemployment rate masks a persistent uptrend in the number of permanently unemployed. Feature Chart 1Breakout After having been lulled to sleep by several months of stagnant yields, bond investors experienced a minor shockwave in early October. The 10-year Treasury yield and 2/10 slope both broke out of well-established trading ranges and implied interest rate volatility bounced off all-time lows to reach its highest level since June (Chart 1). We suspect this might turn out to be just the first small tremor in a tumultuous month leading up to the US election. Specifically, there are two main political risks that will be resolved within the next month. Both have major implications for the bond market. Bond-Bullish Risk: No More Stimulus Before The Election The first risk is the possibility that the current Congress will not deliver any more fiscal stimulus. This increasingly looks like less of a possibility and more of a likelihood, especially after the president tweeted that he is halting negotiations with House Democrats. While he partially walked those comments back the next day, the fact remains that there is very little time between now and November 3rd, and the two sides remain at loggerheads. We have argued that more household income support from Congress is necessary. Otherwise, consumer spending will massively disappoint during the next year.1 However, it could take a few more months before this becomes apparent in the consumer spending data. Real consumer spending still rose in August, though much less quickly than it did in June and July (Chart 2). Meanwhile, August disposable income remained above pre-COVID levels, as it continued to receive a boost from facilities related to the CARES act (Chart 2, bottom panel). This boost will fade as the CARES act’s money is doled out, pushing spending lower. That is, unless Congress enacts a follow-up bill. There are two main political risks that will be resolved within the next month and both have major implications for the bond market. It looks less and less likely that a bill will be passed this month but, depending on the election outcome, a follow-up stimulus bill could become more likely in January. If consumer spending can hang in for the next couple of months, then the bond market might look past Congress’ near-term failure. This appears to be what is happening so far. The stock market fell 1.4% last Tuesday after Trump tweeted about halting negotiations. The 10-year Treasury yield, however, dropped only 2 bps on the day. More generally, long-dated bond yields rose during the past month, even as stocks sold off and prospects for immediate fiscal relief dimmed (Chart 3). Chart 2September's Consumer Spending Report Is Critical Chart 3Bonds Ignore Stock ##br##Market... With all that in mind, we think September’s consumer spending data – the last month of data we will see before the election – are very important. If spending collapses, it might re-focus the market’s attention on Congress’ failure, sending bond yields down. However, we think the market would see through a modest drop in spending, especially if the election looks poised to bring us a larger bill in 2021. Bond-Bearish Risk: A Blue Sweep Election Chart 4...Take Cues From Election Odds This brings us to the second big political risk that could influence bond yields during the next month: The possibility of a “blue sweep” election where the Democrats win control of the House, Senate and White House. This would clearly be a bearish outcome for bonds, as an unimpeded Democratic party would enact a large stimulus package – likely worth $2.5 to $3.5 trillion – shortly after inauguration. It appears that the bond market is already tentatively pricing-in this outcome. While the recent increase in bond yields is hard to square with weak equity prices and souring expectations for immediate stimulus, it is consistent with rising betting market odds of a blue sweep election (Chart 4). To underscore the bond bearishness of this potential election outcome, consider that not only would a unified Congress be able to quickly deliver another fiscal relief bill, but Joe Biden’s platform calls for even more spending on infrastructure, healthcare, education and other Democratic priorities. In total, Biden is proposing new spending of around 3% of GDP, only about half of which will be offset by tax increases (Table 1). Table 1ABiden Would Raise $4 Trillion In Revenue Over Ten Years Table 1BBiden Would Spend $7 Trillion In Programs Over Ten Years How likely is a “blue sweep” election? It is our Geopolitical Strategy service’s base case.2 Also, fivethirtyeight.com’s poll-based forecasting model sees a 68% chance that Democrats win the Senate, a 94% chance that they win the House and an 85% chance that Joe Biden wins the presidency. Investment Strategy These two political risks appear to put bond investors in a bit of a conundrum. On the one hand, if no stimulus bill is passed this month and September’s consumer spending data are weak, then bond yields could fall in the near-term. However, we are inclined to think that if all that occurs against the back-drop of rising odds of a blue sweep election outcome, the bond market will look beyond the near-term and yields will move higher on expectations of larger stimulus coming in January. As such, we retain our relatively pro-reflation investment stance. We recommend owning nominal and real yield curve steepeners, inflation curve flatteners and maintaining an overweight position in TIPS versus nominal Treasuries. All these positions are designed to profit from a rising yield environment.3 Municipal bonds look extremely cheap compared to other US fixed income sectors. We retain an “at benchmark” portfolio duration stance for now, for two reasons. First, while a blue sweep election outcome looks like the most likely scenario, it is not a guarantee. Second, even against the backdrop of greater government stimulus and continued economic recovery, the US economy will still be dealing with a large output gap next year that will temper inflationary pressures. This will keep the Fed on hold, limiting the upside in bond yields. That being said, the odds of another significant downleg in bond yields look increasingly slim. We will likely shift to a more aggressive “below-benchmark” duration stance this month, if our conviction in a blue sweep election outcome continues to rise. A Rare Buying Opportunity In Municipal Bonds No matter how you slice it, municipal bonds look extremely cheap compared to other US fixed income sectors. First, we can look at the spread between Aaa-rated munis and maturity-matched US Treasury yields (Chart 5). When we do this, we find that 2-year and 5-year municipal bonds trade at about the same yields as their Treasury counterparts. This is despite municipal debt’s tax-exempt status. Munis look even more attractive further out the curve, with 10-year and 30-year bonds trading at a before-tax premium relative to Treasuries. Chart 5Aaa Munis Versus ##br##Treasuries Table 2Muni/Corporate Breakeven Effective Tax Rates (%) Next, we can look at how municipal bonds stack up compared to corporates. We do this in a couple different ways. In Table 2, we start with the Bloomberg Barclays Investment Grade Corporate Index split by credit tier. We then find the General Obligation (GO) municipal bond that matches each corporate index’s credit rating and maturity and calculate the breakeven effective tax rate between the two yields. The breakeven effective tax rate is the effective tax rate that would make an investor indifferent between owning the municipal bond and the corporate bond. For example, if an investor faces an effective tax rate of 7%, they will observe the same after-tax yield in a 12-year A-rated GO municipal bond as they do in a 12-year A-rated corporate bond. If their effective tax rate is more than 7%, the muni offers an after-tax yield advantage. Alternatively, we can look at the relative value between munis and credit using the Bloomberg Barclays Municipal Indexes. In Chart 6A, we start with the average yield on the Bloomberg Barclays General Obligation indexes by maturity. We then find the US Credit index that matches the credit rating and duration of the municipal index and calculate the yield differential.4 We find that in all cases, for GO bonds ranging from 6 years to maturity and higher, the muni offers a before-tax yield advantage compared to the Credit Index. This is also true when we perform the same exercise using municipal revenue bonds instead of GOs (Chart 6B). Chart 6AGO Munis Versus Credit Chart 6BRevenue Munis Versus Credit You may notice that municipal bonds trade at a before-tax premium to credit in Charts 6A and 6B, but at a discount in Table 2. This is because we compare bonds by maturity in Table 2 and by duration in Charts 6A and 6B. Unlike investment grade corporates, municipal bonds often carry call options making them negatively convex and giving them a duration that is much shorter than their maturity. Cheap For A Reason, Or Just Plain Cheap? Chart 7State & Local Balance Sheets Will Weather The Storm We have effectively demonstrated that municipal bonds offer value relative to both Treasuries and corporate credit. But attractive value is not enough to warrant an overweight allocation. Ideally, we would also like some degree of confidence that wide spreads won’t eventually be justified by a wave of downgrades and defaults. While state & local government balance sheets are certainly stressed, we see strong odds that the muni market will emerge from the COVID recession relatively unscathed. For starters, state & local governments were experiencing strong revenue growth prior to the pandemic (Chart 7, top panel). This allowed them to build rainy day funds up to all-time highs (Chart 7, panel 4). Second, income support for households from the CARES act helped prop up state & local income tax revenues in the second quarter (Chart 7, panel 2), though sales tax revenues took a significant hit (Chart 7, panel 3). Going forward, a blue sweep election scenario would not only provide more income support for households – helping income tax revenues – but a Democratic controlled Congress would also quickly deliver fiscal aid directly to state & local governments. In fact, it is this aid for state & local governments that is currently the key sticking point in fiscal negotiations. In the meantime, state & local governments will continue to clamp down on spending. This can already be seen in the massive drop in state & local government employment (Chart 7, bottom panel). This is obviously a drag on economic growth, but the combination of austerity measures and high rainy day fund balances will help municipal bonds avoid downgrades and defaults, at least until a fiscal relief bill is passed next year. While state & local government balance sheets are certainly stressed, we see strong odds that the muni market will emerge from the COVID recession relatively unscathed. Bottom Line: Municipal bonds offer exceptional value relative to both US Treasuries and corporate credit. Not only that, but rising odds of a blue sweep election make state & local government fiscal relief increasingly likely. Investors should overweight municipal bonds in US fixed income portfolios. Economy: Credit Growth & The Labor Market Credit Growth Slowing Chart 8No Animal Spirits Of notable economic data releases during the past two weeks, we find it particularly interesting that both consumer credit and Commercial & Industrial (C&I) bank lending continue to slow (Chart 8). On the consumer side, massive income support from the CARES act and few spending opportunities caused households to pay down debt this spring. Then, after two months of modest gains, consumer credit fell again in August (Chart 8, top panel). This strongly suggests that, even as lockdown restrictions have eased, consumers aren’t yet ready to open up the spending taps. On the corporate side, firms received much less of a direct cash injection from Congress and were forced to take on massive amounts of debt to get through the spring and early summer months. But as of the second quarter, we recently observed that nonfinancial corporate retained earnings now exceed capital expenditures.5 This strongly suggests that firms have taken out enough new debt and that C&I bank lending will remain slow in the coming months. Cracks Showing In The Labor Market Chart 9Far From Full Employment Finally, we should mention September’s employment report that was released two weeks ago (Chart 9). It is certainly positive that the unemployment rate continues to fall, but the main takeaway for bond investors should be that the US economy remains far from full employment, and therefore far away from generating meaningful inflationary pressure. While the unemployment rate fell for the fifth consecutive month, it is now dropping much less quickly than it did early in the summer (Chart 9, panel 2). Also, we continue to note that labor market gains are entirely concentrated in temporarily unemployed people returning to work. The number of permanently unemployed continues to rise (Chart 9, bottom panel). Bottom Line: The economic recovery continues to roll on, but it will be some time before the output gap is closed and inflation starts to rise. Slow consumer and corporate credit growth suggest that animal spirits have not yet taken hold. Meanwhile, the falling unemployment rate masks a persistent uptrend in the number of permanently unemployed. Appendix The Fed rolled out a number of aggressive lending facilities on March 23. These facilities focused on different specific sectors of the US bond market. The fact that the Fed has decided to support some parts of the market and not others has caused some traditional bond market correlations to break down. It has also led us to adopt of a strategy of “Buy What The Fed Is Buying”. That is, we favor those sectors that offer attractive spreads and that benefit from Fed support. The below Table tracks the performance of different bond sectors since the March 23 announcement. We will use this to monitor bond market correlations and evaluate our strategy’s success. Table 3Performance Since March 23 Announcement Of Emergency Fed Facilities Ryan Swift US Bond Strategist rswift@bcaresearch.com Footnotes 1 Please see US Bond Strategy Weekly Report, “More Stimulus Needed”, dated September 15, 2020, available at usbs.bcaresearch.com 2 Please see Geopolitical Strategy Weekly Report, “It Ain’t Over Till It’s Over”, dated October 9, 2020, available at gps.bcaresearch.com 3 For more details on these recommended positions please see US Bond Strategy Weekly Report, “Positioning For Reflation And Avoiding Deflation”, dated August 11, 2020, available at usbs.bcaresearch.com 4 Note that we use the US Credit Index in Charts 6A and 6B. This index includes the entire US corporate bond index but also some non-corporate credit sectors like Sovereigns and Foreign Agency bonds. 5 Please see US Bond Strategy Weekly Report, “Out Of Bullets”, dated September 29, 2020, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification