Economy

Highlights Duration: The current strength of the U.S. economy suggests that, as was the case in February, the sell-off in risk assets may not result in much of a drop in Treasury yields. Weak foreign economic growth still presents a risk, but it should be hedged by adopting a more defensive stance on credit, not by increasing portfolio duration. Corporate Bonds: Weak foreign economic growth will impact U.S. corporate profits and investment spending before hitting U.S. consumer spending and overall GDP. This means it is better to hedge the risk from weak foreign growth by scaling back exposure to corporate bonds, while maintaining below-benchmark duration. Credit Curve: Favoring the long-end of the credit curve is a way to increase the average spread of a bond portfolio without taking on extra credit risk. Rather, the long-end of the credit curve outperforms when Treasury yields rise and underperforms when they fall. In the current environment we prefer going after the extra spread at the long-end of the credit curve, while maintaining an up-in-quality bias and only a neutral allocation to corporate bonds. Feature We're not out of the woods yet. Risk assets continued their decline last week, the VIX remains elevated and the 10-year yield has fallen off its highs (Chart 1). In the context of our Fed Policy Loop, lower bond yields are a positive sign for risk assets. Chart 1How Much Worse Will It Get? In our Fed Policy Loop framework, higher yields and the perception of increasingly hawkish Fed policy cause credit spreads to widen and stock prices to fall. Then, tighter financial conditions eventually lead to perceptions of more dovish Fed policy and lower bond yields. At some point, yields fall far enough to put a floor under risk assets (Chart 2). Chart 2The Fed Policy Loop We are now at the stage of the loop where we must determine how large a decline in Treasury yields will be necessary to halt the slide in risk assets. To make that determination, it is helpful to think about why risk assets are falling. Is it a simple correction driven by investors re-assessing appropriate valuations? Or is the market sniffing out a future slowdown in economic growth? Chart 3 shows why the difference is meaningful. In February 2018, a sharp increase in Treasury yields caused the stock-to-bond total return ratio to decline. However, the ratio quickly recovered once investor sentiment toward the stock market became somewhat less bullish. Importantly, Treasury yields did not need to fall to support a rebound in risk assets, they only needed to level-off for a time. Chart 3Lower Yields Required? In contrast, a meaningful decline in Treasury yields was required to arrest the drop in the stock-bond ratio that occurred in late-2015/early-2016. The difference between that period and the February 2018 period is obvious. In late-2015/early-2016, the U.S. Manufacturing PMI had just dipped below the 50 boom/bust line. This year the PMI has been closer to 60 (Chart 3, bottom panel). The current strength of the U.S. economy suggests that, as was the case in February, the sell-off in risk assets may not result in much of a drop in Treasury yields. With the market only priced for 54 bps of rate hikes during the next 12 months, and no signs of softening in the U.S. economic data, we are reluctant to abandon our cyclical below-benchmark portfolio duration stance at this time. That is not to say there are no risks on the horizon. In past reports we flagged the risk that slowing foreign economic growth will eventually impact the U.S. economy, causing it to slow as we head into next year.1 However, we think it makes more sense to hedge this risk by adopting a more defensive allocation to corporate credit versus Treasuries, rather than by shifting portfolio duration to look for lower yields. Hedge Economic Risk In Credit, Not Duration As was stated above, U.S. economic growth remains strong and the biggest risk on the horizon is that weak foreign growth eventually migrates stateside via a stronger dollar. Last week's third quarter GDP report confirmed that overall growth is solid, but also showed some evidence of weak foreign growth impacting the U.S. figures. Overall, real GDP grew by a healthy 3.5% (annualized) in the third quarter, supported mostly by consumer spending which contributed 2.7% to overall growth, the most since Q4 2014 (Chart 4). However, weakness was found in nonresidential investment spending which contributed only 0.1% to real growth, down from 1.2% in the prior quarter (Chart 4, bottom panel). Chart 4Parallels With Early 2015 This distribution of growth between consumer spending and investment is identical to what occurred in 2015, the last time that weak foreign growth infiltrated the U.S. economy. The more globally-exposed investment sector contributed almost nothing to growth in the first two quarters of 2015, while overall GDP growth stayed elevated, driven by strong consumer spending. Eventually, consumer spending also weakened and GDP growth plunged in the second half of the year, but the warning sign that weak foreign growth was negatively impacting the U.S. economy came from investment spending in the first half of 2015. We draw the distinction between U.S. investment spending and U.S. consumer spending for two reasons. The first is that investment spending is more influenced by global factors than the U.S. consumer. The second is that investment spending is tightly linked to corporate profit growth (Chart 5). In other words, weak foreign economic growth is likely to negatively impact U.S. corporate profits before it hits overall U.S. GDP. This makes credit spreads more exposed to global weakness than Treasury yields, which take their cues from overall GDP growth. Chart 5Investment Spending And Profits Are Linked While we think that weak foreign growth will weigh on corporate profits in the coming quarters, presenting a clear negative for corporate bond spreads. We must also consider that spread widening during the past two weeks means that valuation has improved. Our measure of the excess spread available in the High-Yield index after accounting for expected default losses now stands at 274 bps, up from 212 bps a month ago and slightly above the historical average (Chart 6). However, we must also point out that our calculation embeds expected default losses of only 1.04% for the next 12 months. This low default loss expectation, which is derived from Moody's baseline default rate forecast and our own forecast of the recovery rate, means that there is a high risk that default losses surprise investors to the upside during the next 12 months (Chart 6, bottom panel). Any moderation in profit growth would make such an upside surprise even more likely. Chart 6Junk Value Has Improved... Another way to think about our default-adjusted high-yield spread is that if we assume that default losses occur in line with our forecast and that junk spreads remain flat at current levels, then junk bonds will outperform duration-matched Treasuries by 274 bps during the next 12 months. If spreads tighten by enough to bring the default-adjusted spread back to its historical average of 247 bps, then junk will outperform duration-matched Treasuries by 380 bps. However, at the current juncture we are more worried about spread widening during the next 6-12 months than spread tightening. Chart 7 shows that junk spreads tend to predict changes in capacity utilization. At current spread levels, this means we should expect capacity utilization to rise back to the 80% level during the next six months. If weak foreign economic growth starts to weigh on U.S. corporate profits, then such a large gain is very much in doubt (Chart 7, bottom panel). Chart 7...But Spreads Embed Strong IP Growth Bottom Line: Weak foreign economic growth will impact U.S. corporate profits and investment spending before hitting U.S. consumer spending and overall GDP. This means it is better to hedge the risk from weak foreign growth by scaling back exposure to corporate bonds, while maintaining below-benchmark duration. Extend Maturity In Credit, Not Treasuries Given the risk to corporate profits that is posed by weak foreign economic growth, we recommend investors maintain only a neutral allocation to corporate bonds and also maintain an up-in-quality bias across credit tiers.2 In the current environment we think the best way to pick-up spread within a neutral allocation to corporate bonds is to favor the long-end of the maturity spectrum. This will need to be offset by maintaining very low duration within your Treasury allocation to ensure that overall portfolio duration stays below benchmark. The rationale for favoring the long-end of the corporate credit curve is twofold. First, there is extra spread available at the long-end of the credit curve compared to the short-end. In fact, the long-maturity investment grade corporate bond index carries an average option-adjusted spread that is 107 bps greater than that of the intermediate-maturity index (Chart 8). Second, the extra spread available at the long-end of the credit curve is purely compensation for the extra duration risk. The bottom panel of Chart 8 shows that there is no spread advantage at the long-end on a "per unit of duration" basis. Chart 8Favor The Long-End Of The Corporate Credit Curve The fact that the extra spread at the long-end of the credit curve is purely compensation for duration is important because it means that when Treasury yields rise and average index duration falls, investors should demand less compensation for the extra duration risk at the long-end of the curve. In other words, rising Treasury yield environments should coincide with spread compression at the long-end of the credit curve versus the short end. We tested this idea empirically by looking at monthly excess returns in long-maturity corporate bonds versus short-maturity corporate bonds. Using a sample of monthly returns going back to 2000, we divide the months based on whether the Treasury curve bear-steepened, bear-flattened, bull-steepened or bull-flattened (Table 1). The results show that while changes in the slope of the yield curve don't have much impact, the long-end of the credit curve outperforms when Treasury yields rise and underperforms when they fall. Table 1Monthly Excess Return In Long Maturity Vs. Short Maturity Corporate Bonds (2000-Present) Bottom Line: Favoring the long-end of the credit curve is a way to increase the average spread of a bond portfolio without taking on extra credit risk. Rather, the long-end of the credit curve outperforms when Treasury yields rise and underperforms when they fall. In the current environment we prefer going after the extra spread at the long-end of the credit curve, while maintaining an up-in-quality bias and only a neutral allocation to corporate bonds. Ryan Swift, Vice President U.S. Bond Strategy rswift@bcaresearch.com 1 Please see U.S. Bond Strategy Weekly Report, "Waiting For Peak Divergence", dated October 23, 2018, available at usbs.bcaresearch.com 2 Please see U.S. Bond Strategy Weekly Report, "Rigidly Defined Areas of Doubt And Uncertainty", dated June 19, 2018, available at usbs.bcaresearch.com Fixed Income Sector Performance Recommended Portfolio Specification