"Trust my folly then, since it is best for a man truly wise to be thought a fool." - Aeschylus, Prometheus Bound
Watching with interest "risk-off" unfolding with Emerging Markets continuing to bleed thanks to the ratcheting up in the trade war rhetoric, leading to more market turmoil, when it came to selecting our title analogy we decided to go for the Greek tragedy reference namely "Prometheus Unbound", a fragmentary play attributed to 5th century BC Greek tragedian Aeschylus. What is not to like in this title and the tragic element? It seems more and more probable that the United States and China cannot escape the Thucydides Trap being the theory proposed by Graham Allison former director of the Harvard Kennedy School’s Belfer Center for Science and International Affairs and a former U.S. assistant secretary of defense for policy and plans in 2015 who postulates that war between a rising power and an established power is inevitable:
"It was the rise of Athens and the fear that this instilled in Sparta that made war inevitable." Thucydides from "The History of the Peloponnesian War"
"Prometheus Unbound" is as well a four act-lyrical drama by Percy Bysshe Shelleyn Mary Shelly's husband, first published in 1820. In the Act II Asia the daughter of Oceanus and Thetys a water nymph and Prometheus love interest, and her sister Panthea descend in Scene IV in the cave of the Demogorgon, the demon ruler of the underworld. In this scene Demorgorgon promises the rise of a new world, triggered by Prometheus' revolution against Zeus/Jupiter. At the end of this act the revolution is triggered, leading to a new order but we ramble again...
Basically, current face-off between the United States and China, and also with the United States and Europe as well as internal rifts between the European Union mark a tectonic shift and validates our views that we are moving from cooperation to noncooperation. In a certain sense, the old Jupiterian order is being challenged by a new Prometheus. This is of course leading to even more instability and volatility in financial markets.
In this week's conversation, we would like to look again at the evolution of the trade war narrative and the rising instability we are seeing as well as the starts of widening spreads in credit which should put some additional pressure on equities at some point.
Synopsis:
- Macro and Credit - Trade war? It's on like Donkey Kong
- Final charts - Rollover risk? New debt is becoming more expensive...
- Macro and Credit - Trade war? It's on like Donkey Kong
As per our previous conversation, the recent turmoil seen in equities markets is a symptom of the prevailing "mercantilism" and trade war narrative:
"Returning to our core subject of "mercantilism" and trade wars, it is looking more and more likely that in similar fashion to the 1930s, we risk seeing tit for tat reactions from China to additional US sanctions. Obviously equities markets are reacting to this. Emerging Markets were the big beneficiaries of globalization and cooperation. Following NIRP and ZIRP implementation by DM central banks, EM have benefited from the high beta chase and massive inflows into funds. With the QE tide receding thanks to QT and with the escalation of trade war fears, obviously EM are coming under much pressure, hence our reverse macro osmosis theory we have been discussing various times playing out." - source Macronomics, June 2018
As we pointed out as well, any escalation of the trade war rhetoric would lead to some stagflationary effects, higher prices and lower growth. If indeed there are additional measures taken for example against China, Chinese President Xi Jinping said last week he will not hesitate to retaliate against the U.S. on trade. This is indeed something to take note of.
On this very subject we read with interest Deutsche Bank US Economic Notes from the 19th of June entitled "The tit-for-tat trade barbs intensify":
"Trade tensions continue to intensify
We recently noted that the Trump administration has intensified trade uncertainties on a number of fronts. Over the past week, this intensification has focused on trade relations with China, where the administration announce plans for tariffs on $50bn of goods imports. The first tranche of tariffs, covering $34bn of Chinese goods, will begin to be collected on July 6. The second set of products covering $16bn of goods will undergo further review in a public notice and comment process, including a public hearing. In addition, the Administration threatened 10% tariffs on an additional $200bn of goods in response to China’s retaliation, and potentially another $200bn if China retaliates again.
In terms of timing, recall that the Trump Administration announced tariffs on the original $50 billion of goods on March 22. The product list was then released on April 3 with public hearings held in mid-May and the tariffs are expected to be collected starting July 6. We therefore assume that it would take around two to three months for the process to proceed to the implementation stage for the next $200 billion of goods. Given that the annual Chinese party meeting will take place in October and the US midterms in early November, we would not be surprised if tensions remain elevated for the next several months.
US businesses and consumers will be the prisoners of the latest dilemma
As we outlined in a recent note in April, the initial list of $50 billion of goods targeted for tariffs was relatively benign with respect to the potential impact on US consumers. In order to expand the list of goods to $200 billion, the US Administration would be hard-pressed to avoid large sectors such as consumer electronics. As Figure 1 illustrates, computer and electronics products accounted for a substantial portion of Chinese imports last year.
In fact, as Figure 2 outlines, the top 10 product codes in the USITC classification system account for a little over $150 billion of US imports from China.
This is important because as we can see at the top of the list, cell phones are the single largest category of Chinese exports to the US by dollar value, followed by tablets and laptops. These sectors combined represent over $80 billion of imports. It is likely that sourcing these products from other countries is not a near-term option for large US companies such as Apple. Therefore, these firms will then have to choose between absorbing the higher tariff costs via lower profit margins versus passing them onto consumers.
Growth implications
In our initial detailed look at possible trade war scenarios, we concluded that a trade war light scenario in which reciprocal tariffs were enacted that target somewhere between $50 and $100bn in goods, or equivalently around $30bn in tariff revenues, would shave -0.1 to -0.2 percentage points off of US growth. The proposed tariffs target a larger sphere of goods but at a lower weighted average tariff rate than we originally assumed, raising about $32bn in tariff revenue, or about 0.16% of nominal GDP. As such, the growth impact of this scenario is likely to be only slightly larger than we assumed in our original assessment, likely on the order of -0.2 to -0.3 percentage points. There is considerable uncertainty around this conclusion, owing to the difficulty in assessing supply chain disruptions and the response of financial conditions and consumer and business sentiment to the introduction of tariffs on a broad swath of goods.
In addition, our colleagues in China have noted that the cumulative impact of an escalation of the trade dispute would subtract -0.2 to -0.3 percentage points from growth in that region. China’s Ministry of Commerce issued a short statement, which promised to retaliate with "comprehensive quantitative and qualitative measures" if the US imposes more tariffs. As they had previously discussed (reports here and here ), US firms sold US$ 448bn worth of goods and services to China in 2017, $168bn through trade and $280bn through local operation by US subsidiaries in China. China has not threatened officially to target the US firms operating in China, but as the trade tension builds up our Chinese colleagues see that risk rising.
Inflation impact
According to analysis by San Francisco Fed economists, imports account for about 13% of the core PCE price index. The $50bn of goods targeted for a 25% tariff rate accounts for about 1.7% of total imports in 2017, while the $200bn in goods targeted for a 10% tariff rate accounts for about 6.9% of 2017 imports. If we assume that the types of goods targeted by tariffs are broadly representative, we find that core PCE inflation would rise by about 15 basis points (bps) in response to the implementation of these tariffs. Importantly, this would act as a one-time increase in the price level and thus provide a mostly transitory boost to inflation that would largely unwind in the following year.
Fed implications
With growth well above potential, the labor market beyond full employment and inflation near the Fed’s target, the Fed may be willing to tolerate a modest negative impact on growth, and Fed Chair Powell’s press conference indicated that he will likely need to see the negative implications in financial markets or economic data for the Fed to respond. However, with some willingness to tolerate above-target inflation, we think the Fed would very likely look through any rise in core goods inflation as “transitory”. Monetary policy makers would be more concerned about the second order effects on growth. Powell noted reports of companies holding off on making investments and hiring due to increased trade uncertainties when queried on the topic at his post-meeting press conference last week. Indeed, in the Fed's May Beige Book, tariffs were mentioned 22 times and business contacts noted that trade uncertainty was contributing to "supply chain disruptions" and have caused some firms to tap "the brakes on projects in the planning phases." Atlanta Fed President Bostic (voter/neutral) echoed these sentiments yesterday, noting that he has shifted his balance of risks to growth on the back of the recent escalation of trade threats.
It may need to get worse before it gets better
How far the trade tensions escalate will be determined by some combination of political and market pressure that is brought to bear on the issue. At some point, the potential economic damage and uncertainty caused by the trade tensions will precipitate a meaningful pullback in risk asset prices, forcing one or both parties to the negotiating table. How much of a drop in risk may be necessary to force the two parties to a negotiated settlement? As our equity strategist, Binky Chadha has often noted, 3% to 5% pullbacks in the stock market are a frequent occurrence. At the same time, 10% corrections are rare and signal up to a 50% probability of a recession, and the market generally drops 20% or more around actual recessions. We surmise that a correction approaching 10% should be enough to terminate the trade conflict unless it has gathered substantial momentum. A smaller correction could be sufficient if it was clearly induced directly by trade policy actions, though we admit that gauging the Administration’s reaction function is difficult.
Given the sound underlying fundamentals for US output growth over the next year, overall corporate earnings growth may be somewhat impervious to escalating trade tensions with China, but only up to a point. It is difficult to come up with direct effects from a tariff on $250bn of US imports from China plus a similar tariff on China’s $130 billion imports from the US ($160 billion including Hong Kong). This exercise becomes more complicated if regulatory measures are implemented that effectively act as capital controls. The book value of Chinese foreign direct investment in the US was over $27 billion in 2016 while that of US direct investment in China was over $90 billion. Our equity strategists estimate that a 0.2 to 0.3pp drag on GDP growth translates to a 1 to 1.5% hit to S&P 500 earnings growth. In the context of 2018 EPS growth which is on track to hit their forecast of 23% and 2019 growth estimated in double-digits, a 1-1.5% drag is not particularly significant. While the indirect effects of uncertainty and hit to business confidence are harder to gauge, they see corporates responding by putting more of their cash flow towards share buybacks, the primary driver of equities in this cycle from a demand-supply perspective.
In the meantime, China's review of Qualcomm’s $44 billion bid for NXP Semiconductors and the US handling of penalties against ZTE could be a litmus test of how this dispute could play out. A related issue concerns the progress of the Committee on Foreign Investment in the US (CFIUS) Modernization Act of 2018, which has been passed in the House but has not yet been debated in the Senate. This bill would provide a broad expansion of the government’s ability to review foreign investment. Further complicating matters is the Senate’s potential challenge to the Administration's handling of the ZTE penalties. Congress still has a busy schedule ahead of the August recess, which the Senate majority leader has threatened to cancel.
Other areas of potential conflict lie waiting in the wings as well. Soundings from the Administration indicate that tariffs on trade in autos could be in the offing, and important negotiations regarding NAFTA remain well short of a satisfying conclusion. Should the Administration choose to move ahead with auto tariffs and withdraw from NAFTA, the potential for an economic downturn will intensify significantly. In short, there will likely be no rest for the weary among trade warriors and their watchers in the months ahead. - source Deutsche Bank
Both Chinese elections and US elections could be seen as a catalyst for more posturing from both countries in the search for domestic political support. This will no doubt continue to create market jitters and increased volatility in the near term. Therefore we continue to see more pain ahead for Emerging Market (EM) and overall US stock markets should continue to outperform on a relative basis.
From an allocation perspective, we continue to think that any escalation in trade war would be bullish for gold, which recently has been weakening. We also think as well that the US long end is starting to be enticing.
Any escalation in trade war would have various consequences. On this subject we read Bank of America Merrill Lynch's take in their Liquid Insight note from the 20th of June entitled "Implications of trade war on rates and FX":
"Trade wars: near-term lower rates and flatter curve
As trade skirmishes continue to deepen, we see increased risk to our base case year-end outlook of 3.25% 10y rates, while our flatter curve view and wider 5y inflation breakeven view appear less vulnerable. The rates market impact of an escalation of trade tensions will naturally depend on the depth and severity of tariffs and retaliatory actions, but recent rhetoric suggests trade escalation may need to get worse before it gets better.
While views vary across outcome probabilities, our Global and Chinese Economics teams see an increase in trade risks.
Should trade tensions worsen, this would likely contribute to lower levels of nominal and real rates, an initially flatter curve, wider short-dated breakevens, and a stronger USD. The Fed would likely be undeterred from near-term hikes, but increased trade friction would likely force them into a slower medium-term path and lower terminal rate for this cycle. The sharp decline in payer skew (ie, declining cost of higher-rate protection) we think confirms the shifting of market probabilities toward the escalation scenario versus a benign outcome. With 3m-5y payer skew now at its lowest since 2007 (+25bp out vs - 25bp out), contrarian traders could take advantage of historically cheap valuations to position for a near-term benign outcome.
Should trade tensions increase, we would expect rates to initially decline with more severe risk off and the curve to flatten as the Fed likely remains on course for further rate increases later this year. However, over time, we would expect that the Fed will need to react to the weaker global growth backdrop by stopping their tightening cycle and potentially cutting rates, which would allow the curve to steepen. As discussed here, this environment would be supportive of wider short-dated breakevens and we also see risks of lower real rates along with a flatter breakeven curve. We also see risks that a material increase in trade tensions could result in less foreign investor sponsorship of US Treasuries, which would contribute to higher rates and term premiums in the longer run.
Fed’s reaction: flatter curve and lower long-term rates
Overall, we think a trade war escalation is unlikely to deter the Fed from near-term hikes but would likely result in a slower medium-term path and lower terminal rate for this cycle.
We see the Fed’s response to a trade-war escalation as occurring through two phases:
- Near term: the Fed will likely refrain from overreacting to the increase in trade tensions and aim to continue with two additional rate hikes this year. The Fed has repeatedly indicated it does not see a meaningful impact from trade tensions in the data thus far. Chair Powell responded to a question on trade in his most recent press conference by saying that “we don’t see [trade tensions] in the numbers at all. The economy is very strong. The labor market is strong. Growth is strong. We really don’t see [trade tensions] in the numbers, it’s just not there.” That said, Powell and other Reserve Bank Presidents (ex. Atlanta Fed President Bostic) have indicated risks from trade tensions are rising among their business contacts. We expect the Fed will likely refrain from altering the direction of policy until these tensions are more evident in economic data or through a sustained tightening in financial conditions.
- Medium term: the Fed may need to reduce the total number of rate hikes over the course of this cycle and realize a lower terminal rate. Although tariffs will likely place near-term upward pressure on inflation, we expect the Fed will look through this dynamic and instead focus on underlying softening global growth and consumer demand. This would pose risks to the number of hikes the Fed delivers in 2019 or 2020 and result in a lower terminal rate for this cycle. This assumes any near-term tariff-induced increase in the price of goods would be expected to be short lived and not spill over into longer-run inflation expectations, which is our base case view. If trade tensions result in a rapid escalation of tariffs and protectionist policies, a severe trade war could suppress growth and potentially cause the Fed to consider rate cuts.
Rates reaction: lower rates, higher breakevens, flatter B/E curve
US Treasury rates would likely decline with a further material increase in trade tensions. Rate declines would be led by risk-off dynamics and lower growth expectations. Based upon our expectations for the Fed, the initial response from the rates market would be to price in a flattening of the curve as the front end responds to tightening while 10y and 30y rates remain low due to declining growth expectations and a lower pricing of the terminal rate. As the Fed shifts their reaction function, we would expect the Treasury curve to steepen due to lower expectations for rates hikes and potential concerns about overseas demand for Treasuries.
Within the composition of the rate moves, we would expect wider near-dated inflation breakevens, lower real rates, and a flatter breakeven term structure. Based on market reactions to tariffs in recent years (Table 1) including Bush’s steel tariff, Obama’s tire tariff, and Trump’s recent washing machine and aluminum/steel tariffs, the clearest market reaction is higher breakevens and a flatter breakeven curve.
This was particularly pronounced following the 2002 tariffs on steel, and on average still holds for the more recent measures. While we intuitively anticipate a decline in real rates in the face of tariffs, on a historical basis the real rate reaction has been mixed: 2y real rates tend to decline, but otherwise real rate levels increased on average following the announcement
of a new tariff.
Given these expectations, we see upside potential in our 5Y breakeven widener view. We originally recommended this view due to the economic backdrop and positive carry dynamics, but also see this position benefitting from an escalation in trade concerns. Any surprise from OPEC later this week to not expand supply would also benefit higher short-dated breakevens.
Reserve manager impact: upward pressure on rates after the initial risk-off
Trade tensions are likely to result in weaker UST demand from foreign central banks due to slower FX reserve build and less likely due to trade retaliation. From a flow perspective, a key channel between trade tensions and rates is reserve manager demand. Foreign investors accumulated roughly $6.2tn of US Treasuries over two decades as a result of international trade and market intervention, and $4tn of them are held by reserve managers. China and Japan are the biggest creditors of the US government, holding about 7% of overall marketable Treasury securities each. According to the latest TIC survey, foreign central banks hold mostly front-end to intermediate maturities (Chart 1).
After an immediate risk-off episode, the medium-term reduction in bilateral trade deficit between the US and China should result in a slowdown in China’s FX reserve build, resulting in lower Treasury demand from China. Lower global trade volumes should result in a slowdown of global reserve accumulation, less central bank demand for USTs and upward pressure on interest higher rates (Chart 2).
We think it is unlikely, however, that central banks would sell Treasuries as a retaliatory measure in trade discussions. It is unclear if Treasury selling would have the desired market impact given the mixed history. From 2015 to 2016, foreign official investors sold over $400bn in Treasuries (per TIC data), yet 5y and 10y UST rallied by 60-100bp. In theory, any sizable selling from CBs may be more concentrated in the shorter duration bucket, where the majority of central bank holdings are located. Given the significant demand in the front end this year (Chart 3, 70% of government mutual fund inflows have been concentrated in the very front end as demand for cash like assets surged), the impact of outright UST selling from CBs may not override the flight to quality bid and the repricing of Fed path, as observed in 2015-16. However, such reserve manager selling would likely serve to tighten swap spreads.
USD: accelerated appreciation
In our view, an escalation of trade war rhetoric would serve to accelerate USD appreciation that is already well underway as a result of cyclical and monetary policy divergence, with the added boost to the dollar essentially a function of the severity of perceived global trade deterioration. Under a scenario of a sharp pickup in trade policy deterioration, we would expect the dollar to sharply appreciate bilaterally against most currencies with the exception of the yen, which despite its substantial export sector tends to be supported in risk-off situations due to its large net foreign asset position. USD returns would likely be highest against the higher beta currencies of economies perceived to be most vulnerable to a slowdown in global and/or US trade, a list traditionally including CAD and AUD among others. Recent sharp deceleration in Euro Area data as a result of rising trade tensions early in the year suggests to us that EUR is also perceived to be included here." - source Bank of America Merrill Lynch
Escalation of trade war would see risks of lower real rates along with a flatter breakeven curve in the near term while obviously a rising US deficit would be negative in the long run for the US yield curve overall as foreign investors might shun purchasing additional US Treasuries.
There indeed mounting signs that in some places global trade and growth are slowing down. The trade war rhetoric is adding to this trend as of late. The risk-off environment can again be ascertained from fund outflows as indicated by Bank of America Merrill Lynch Follow The Flow note from the 22nd of June entitled simply "Cutting risk":
"Trimming longs
With credit spreads been broadly flat over the past couple of weeks, it seems that high grade investors have been trimming risk (chart 1) amid fears that Italian risk will not subside any time soon.
Notably all the major asset classes we track have recorded outflows last week, as trade wars have also supported the risk-off momentum.
Over the past week…
High grade fund flows were negative for a fifth week in a row; note that the majority of the outflow came from two funds.
High yield funds continued to record outflows (32nd consecutive week). Looking into the domicile breakdown, global and to a lesser extend European-focussed funds have recorded the largest outflows, followed closely by the US-focused funds.
Government bond funds recorded a third outflow last week; however the pace of the outflow has more than halved w-o-w.
All in all, Fixed Income fund flows were for a fifth week in negative territory, as the major individual pockets recorded outflows.
European equity funds continued to record outflows for the 15th consecutive week; over that period the asset class has suffered $39bn of outflows.
Global EM debt funds also recorded an outflow last week - the ninth in a row - with the trend deteriorating w-o-w, the dollar strength is heavily weighing on the asset class. Commodity funds recorded sizable outflows, the largest since 2013.
On the duration front, short-term IG funds recorded outflows, reversing last week’s inflow. Mid-term funds recorded their fifth weekly outflow in a row. However, long-term funds have seen inflows after a week of inflows." - source Bank of America Merrill Lynch
What is of course of interest to us from a credit perspective is the developing weakness in credit spreads. We did notice that US Investment Grade has so far clearly underperformed US high beta in the High Yield space, partly because of the Energy sector being a heavyweight in the CCC rating bucket. This disconnect has been highlighted in another Bank of America Merrill Lynch report, their latest Situation Room report from the 25th og June entitled "Let's get back together":
"Let’s get back together
Recently IG underperformance vs. HY in USD credit markets has been remarkable (Figure 1).
Reasons for IG weakness include first and foremost supply pressures in an environment of reduced demand that began in March and extended through last week, plus the Italian situation, which is about systemic risks running through the global IG financial system. Reasons for HY strength include the lack of supply, etc. Trade war risks are probably more neutral across the credit spectrum, as we saw today. We think IG spreads are somewhat predictable based on seasonal supply volumes and that we increasingly will be able to decouple from the Italian risks. That leaves IG spreads currently at the wide end of the range and with our HY strategist, Oleg Melentyev’s, outlook for wider spreads (see: High Yield Strategy: Scarcity of Yield is Now History 15 June 2018), that suggests decompression." - source Bank of America Merrill Lynch
What is of course of interest to us is that the recent in Investment Grade credit in the US in conjunction to a rising US dollar should seduce again foreign investors, and even Japanese investors, given their yield appetite. This is confirmed by Bank of America Merrill Lynch Credit Market Strategist note from the 22nd of June entitled "The return of foreign investors":
"USD corporate bonds now again offering better hedged yields than EUR
A couple of Fed rate hikes later, some spread widening and the USD corporate bond market is now again offering higher currency hedged yields for foreign investors than its EUR counterpart for most relevant maturities (Figure 6, Figure 7).
Specifically higher hedged yields can be obtained in USD credit inside 7-8 year maturities, which includes more than two-thirds of the EUR market." - source Bank of America Merrill LynchClearly, it appears to us that the divergence between US High Yield and US Investment Grade is not warranted. The recent decoupling to us should correct itself. Although high beta has been performing strongly, we think that in the second part of the year we could see a reversal with Investment Grade outperforming should a deceleration in growth materialize.
As pointed out in our most recent musing, US High Yield has had a much better performance overall than EM High Yield and even against US Investment Grade. But, US High Yield has started to look expensive. On that note we agree with Deutsche Bank's US HY Strategy note from the 26th of June entitled "USD HY looking increasingly expensive":
"In a year that has proved challenging for credit market returns, the USD HY market has proved something of a relative bright spot. In Figure 1 we show YTD total returns across the credit spectrum in local currency terms, ordering performance from best to worst. We can see that USD HY is one of the few parts of the credit market that has provided positive total returns, with CCCs and Bs, the standout performers. This is very much in contrast to the performance of USD IG credit, which has seen the worst of the YTD performance, weighed down by the extra duration and associated impact of rising Treasury yields.
The higher yields available in USD HY, particularly for lower ratings, certainly help to provide higher returns, but as we can see in Figure 2 it is also the only part of the credit market that has seen tighter spreads at an index level so far this year. It is worth highlighting that the strength of this performance is really being driven by CCCs and Bs, with the returns and spread moves for BBs much closer to what we've seen from USD IG credit, even though BBs have still outperformed IG.
Given this relative strength from USD HY we try to assess whether such outperformance makes sense and whether this part of the credit market is becoming excessively expensive.
USD HY broadly in line with volatility
In the left hand chart of Figure 3 we update our analysis comparing USD HY credit spreads against a volatility implied spread series.
(click to enlarge)
We can see that spreads have continued to broadly track the volatility implied series, although it's clear that the reaction of USD HY credit spreads to the spike in volatility earlier this year was limited. As recently as a couple of weeks ago, the two series were in line but HY spreads have generally remained firm, even as we've seen a rise in volatility around trade tensions. The current differential has USD HY spreads around 70bps tighter than the volatility implied level. So this measure is suggesting USD HY is on the expensive side now. Interestingly, a quick comparison of the same measure with IG credit spreads shows that the opposite is currently true, with USD IG credit spreads actually wider (around 10bps) than the volatility implied level, even after the recent move higher in volatility.
HY vs. IG
We've already highlighted in terms of total returns that USD HY has comfortably outperformed USD IG. In Figure 4 we show the relative spread ratio which shows that at the start of 2018 the ratio was as high as it had been since the GFC.
This would arguably suggest that HY looked relatively cheap to IG at this point and therefore it is understandable that HY has outperformed. That said, the ratio is now towards the lower end of the post GFC range so the same indicator is now suggesting USD HY looks expensive compared to USD IG. If you remain bullish on USD HY you may point to the fact that the more domestic based issuers are less trade war sensitive than say IG companies. However if it eventually leads to a recession, these companies will eventually be hit hardest, so it is very hard to say they are a safe place to hide, in our view.
USD vs. EUR HY
Last month we published a more detailed note considering the relative value between EUR and USD HY ( link ), and here we provide an update of the simplest part of that analysis comparing broad index spreads for USD and EUR HY. At the time USD HY spreads had been tightening relative to EUR HY spreads but at an index level were still wider. As we can see in Figure 5 that is no longer the case with USD spreads now around 20bps tighter than EUR spreads.
(click to enlarge)
That said, the outperformance of USD HY can, in part, be justified by the fact that on this measure in Q4 2017 EUR HY was about as tight to USD HY as it had been since the GFC. However with this differential now back in positive territory and towards the upper end of the post-GFC range, USD HY is now looking fairly expensive on this measure and this does not even take into account the lower credit quality of the USD HY market compared to EUR HY.
We've shown that USD HY has compared favourably, performance wise, to other parts of the main corporate bond markets and that this outperformance has not been entirely unjustified. However there is certainly evidence that it is looking increasingly expensive. We now compare with some other parts of the broader credit spectrum; starting with EM.
DM vs. EM (USD HY vs. USD EM HY)
It's certainly interesting to look at how these two asset classes have performed relative to one another this year. YTD total returns for EM credit are currently less than -5%. This is not only weak compared to the positive total returns produced by USD HY but also compares unfavourably with other parts of the credit spectrum we have already used for comparison.
Interestingly, through Q1, total returns for both HY and EM were around -1%, so all of the divergence has really occurred during Q2. This is fairly well highlighted in Figure 7 which looks at the relative spreads of the two asset classes.
(click to enlarge)
The spread differential had fluctuated in the 20-80bps range from late 2015 to the end of Q1 2018. However over the course of the last few months this has widened from less than 30bps to more than 150bps. Whether this says something about the relative attractiveness of USD HY now is probably a little more complicated. It can be argued that these relative moves are justified by the implications of Fed actions of late. USD HY should benefit from the same growth dynamics that have allowed the Fed to hike rates and therefore it's reasonable for spreads to keep tightening or at least remain tight. At the same time higher rates in the US can have negative implications for EM currencies and expose potential economic weakness and therefore lead to spread widening for EM sovereign credits. We have seen this recently for countries such as Argentina, Brazil and Turkey. We can also look to 2013 when there was also a notable divergence. This occurred around the time of the taper tantrum with the potential for the removal of monetary accommodation a likely factor here too. The EM-DM differential went from around 0bps in Q1 2013 to nearly +200bps by the end of September and as much as +280bps during Q1 2014 before hitting a peak differential of more than 400bps in early 2015. The differential only closed at the height of the recent energy/commodity crisis.
Given the likely drivers of the divergence, we're not sure in isolation whether the underperformance of EM in recent months is a particularly strong barometer. That said, given that other measures are suggesting USD HY looks expensive then it only adds to the story.
Bonds vs. loans
One comparison that is not so flattering for USD HY is the performance compared to the loan market. Figure 8 shows YTD total returns for bonds against loans and highlights that so far in 2018 loans have outperformed bonds by more than 1.5%.
So based on this comparison with broadly similar rated securities, there is nothing exceptional about the performance of USD HY. Clearly the floating rate nature of loans has helped here but we can see that loan returns are also much less susceptible to mark-to-market volatility and have therefore provided a more steady accumulation of carry this year. Looking at relative spreads in Figure 9 we can see that despite the underperformance in returns, spreads for USD HY do not look attractive on a relative basis.
In fact, current bond spreads are tighter than loan spreads at an index level and whilst this has largely been the case for the past 18 months or so, it is actually a fairly rare event historically. So in this instance, the performance of USD HY does not standout but it still looks relatively expensive on this measure.
Based on all of the simple relationships we have analysed, USD HY looks quite expensive. Our expectation for the remainder of the year, as we highlighted in our recent outlook update , is that USD HY is likely to underperform other parts of the credit universe." - source Deutsche Bank
Simply put, there is potential for a reversal of fortune for US Investment Grade credit versus US High Yield particularly in the light of the performance in High Yield being in the high beta bucket with CCCs taking the lead. As liquidity is withdrawn by central banks and financial conditions start tightening, it will become more and more expensive to issue debt for the weakest players. Some might say that the lack of issuance will therefore continue to provide a technical bid to the asset class, given the very high correlation with the S&P 500, we do not think US High Yield will be spared as it was in the second part of 2018.
If indeed financial conditions are tightening and liquidity is being withdrawn, then again high beta such as EM will continue to suffer as per our final charts, this mean debt is becoming more expensive.
- Final charts - Rollover risk? New debt is becoming more expensive...
Many pundits have pointed out the correlation between liquidity being withdrawn and tightening financial conditions. Obviously the credit spigot is being gradually closed by the Fed and it remains to be seen when corporate America's debt binge to finance buybacks will turn. On this subject our final charts come from Wells Fargo Economics Group report from the 20th of June entitled "Corporate America's Debt Binge: An Issue When Rates Rise?":
"Corporate debt has risen to 45 percent of GDP, equal to the peak of the previous expansion. Low interest rates and longer payment periods have kept corporate debt cheap, but these trends are beginning to reverse.
Nonfinancial Corporations Load on Debt
The overall U.S. economy has become less indebted since the Great Recession, with domestic debt outstanding as a percent of gross domestic product (GDP) falling to 330 percent in Q1 from more than 370 percent at the start of 2009. Shrinking debt relative to GDP is mainly due to deleveraging in the household and financial sectors, while debt has grown faster than GDP for government and nonfinancial business (below chart).
Aside from government, the largest debt gains have occurred in the nonfinancial corporate business sector. Corporate debt growth turned positive in 2011 and has averaged 5.8 percent a year since, propelled higher by low interest rates and robust investor appetite for fixed income securities. As a percent of GDP, corporate debt is currently sitting at its highest level of the cycle (45 percent) and is equal to the peak of the previous expansion.
Lower Interest Rates and Longer Payment Periods
The interest rates corporations are paying on their debt remain far below historical levels, which has helped to keep debt servicing costs low relative to the amount of debt corporations are holding on their balance sheets. We look at interest expense as a share of total short-term and long-term debt as a proxy for the average interest rate paid by corporations (below chart).
This measure fell substantially after 2009 as the Federal Reserve cut the benchmark rate, and remains below the lows of the 2001-2007 cycle for retail trade and all major manufacturing industry categories. Also limiting the corporate debt burden is a shift toward more long-term debt, which means lower principal payments due in each period (bottom chart).
Short-term debt (due in less than one year) represented 9.4 percent of debt held by manufacturing and retail corporations in Q1. This compares to over 13 percent at the end of the previous expansion.
In Q1, manufacturing corporations owed $505 billion on their debt in the coming year (short-term debt plus installments due on long-term debt), while large retail corporations owed $73 billion. These obligations amount to 19.1 percent and 12.6 percent of current assets, respectively. Even though interest rates and the short-term share of debt are lower than during the 2001-2007 expansion, high debt loads mean that debt payments coming due relative to current assets have far surpassed levels of the previous expansion.
But… Debt is Becoming More Expensive
In the past year, the short-term share of debt and the interest rate proxy have risen for corporations, reversing previous trends. Therefore, debt is set to become more expensive. Given strong economic growth and profits, leverage is not overly concerning at present. However, should investor appetite for debt wane or benchmark interest rates go up quickly, funding new debt could fast become more onerous for U.S. corporations." - source Wells Fargo.
Looks like Prometheus is indeed "unbound" and that there could be some "revolutionary" changes down the line for financial markets. For many years the Fed had your back. Not anymore. If indeed there is an escalation in trade war leading to a stagflationary outcome, then one would expect US long Treasuries and long duration Investment Grade to start outperforming again in the near term. In similar fashion to Demorgorgon promising the rise of a new world, triggered by Prometheus' revolution against Zeus/Jupiter, liquidity withdrawal should be taken very seriously , particularly for EM. As we stated before, liquidity matters more than fundamentals in many instances as our reverse macro osmosis thesis discussed recently played out which is an illustration of "Sudden Stops: A Primer on Balance-of-Payments Crises" as we were reminded by this excellent article (H/T Alibey Tuncer). Things are starting to get interesting...
"The story of 'Prometheus' is the idea that if you're given a gift from the gods, do not abuse it, and do not think you can compete." - Ridley Scott
Stay tuned !
Posts
