While many have repeatedly warned over the past year that the record gains in credit are simply too good to stay - especially in Europe where yields and spreads have collapsed largely thanks to the ECB's relentless purchases of corporate debt, with the central bank announcing on Monday it held a record €127.7bn in bonds under its CSPP program - few are as bearish on credit as Morgan Stanley, which today issued ots 2018 US Credit Outlook which is, in a word, "dire."
In the report titled "When the Levee Breaks" strategist Adam Richmond list the three biggest headwinds for credit as follows: "Fed policy should become a material headwind, markets seem very late cycle, and valuations look extremely rich" and details each below:
An unprecedented central bank unwind... We think there is way too much complacency regarding what is a notable and growing shift in central bank policy globally. Remember, monetary policy has been massive in this cycle, and extremely supportive for credit markets. The Fed is now tightening in an untested way, through the balance sheet, while also pushing rates near restrictive territory. Markets expect a seamless unwind. We do not.
...with markets late cycle, and very dependent on ultra-easy liquidity... It is not a coincidence that fundamental problems are becoming more apparent in one sector after the next, as the Fed withdraws liquidity. In fact, we see late-cycle risks popping up all over the place, and as is often the case near a top, these risks are mistakenly (we think) being rationalized as purely 'idiosyncratic' problems. Defaults should remain low in 2018, but that is expected. Credit markets anticipate defaults one year ahead of time, and we think a cycle turn is closer than many believe.
...and valuations very rich: Spreads are near all-time tights, adjusting for the quality deterioration in the indices over time. Yes, the technicals have been strong, but that may change as the Fed's balance sheet shrinks faster. We note, a recession is not necessary to see negative excess returns, especially in the second half of a cycle, and particularly late in a Fed tightening cycle. Credit markets have not experienced three straight years of positive excess returns in over 20 years.
Looking at the technicals, Morgan Stanley echoes what we said last month when he showed the collapse in spreads to 2007 levels, and warns that "credit spreads are very rich nearly any way we slice the data. Spreads adjusted for leverage are back to 2007 levels in high yield, and 1997 levels in IG."
Exhibit 20 shows our fair value model for IG, HY and loans. In short, we estimate that IG, HY and loan spreads are 41bp, 197bp, and 111bp rich to fair value, respectively, using long-term default, downgrade, and risk-premium assumptions. And as we show in Exhibit 21 below, if we adjust for the deterioration in quality of the IG index over time, we find spreads are only 9bp wide of the all-time tights.
One of the main reasons for Richmond's bearishness, is the "complacency" about the Fed's tightening, which of course is applicable to all asset classes. He explains:
More than anything else, we firmly believe that central banks have been THE driver of credit in this cycle, stimulating markets like never before. Now they are attempting to tighten in a completely untested way, and yet credit is pricing in a seamless unwind. At the least, we expect a bumpier 2018, with a tougher setup anyway we slice it. Growth will decelerate, while the Fed continues tightening into a low-inflation environment, driving a completely flat yield curve (per our rates forecasts). Additionally, the year is beginning with booming confidence, as hopes for tax cuts rise, thus the bar to positively surprise is high, while "Goldilocks" is firmly in the price across most risk assets.
We would not rule out the scenario in which financial conditions could tighten materially next year as the Fed withdraws stimulus in this unprecedented way, especially if growth expectations decline at the same time, pushing us from late cycle to end of cycle (though not our economists’ base case). And for those expecting the Fed to come to the rescue any time volatility picks up, remember that, with the balance sheet now effectively set on "auto-pilot," reversing course, in our view, is a last resort.
Taking a step back, per our forecasts, the Fed will hike 3 times in 2018. While gradual on the surface, this rate-hike cycle needs to be put in context. In other words, as we show in Exhibit 3 this time around, the Fed began hiking much later in an expansion, when GDP growth was weaker and corporate leverage higher vs. the start of past rate-hike cycles. In fact, given the drop in the neutral real Fed funds rate over time, monetary policy is already not that far from restrictive territory
As a result, we believe markets can withstand less tightening than a low absolute level of rates might suggest (exhibit 4). And remember, this is a unique rate-hike cycle. One, tightening began not when the Fed first hiked rates in December 2015, but when they began tapering in early 2014. In this regard, the Fed has arguably already tightened policy by a similar amount as in past cycles (Exhibit 5), a point when credit spreads tend to widen on average (Exhibit 6). Two, along the same lines, the Fed is continuing to tighten, not just by hiking rates, but also through reverse QE.
In fact, we believe investors are focused primarily on the "gradual" pace of rate hikes, treating the balance sheet as an afterthought. But the numbers are large. For example, the Fed will shrink its balance sheet by ~$400bn in 2018 alone. In our view, credit investors underestimated the tailwind from QE in this bull market. Similarly, they may now be underestimating the headwind from reverse QE. And while global central banks will still be adding liquidity next year, even they will be doing so at a slower pace, with the ECB cutting their purchases in half in 2018 and likely ending QE altogether around September of next year, while the BOJ hikes their long-term rate target in 3Q18.
We see "quantitative tightening" as a clear catalyst for weaker technicals – i.e., fixed income demand needs to rise to absorb the additional supply or prices have to adjust somewhere (supply/demand 101). Why not expect the opposite of what happened when the Fed was expanding its balance sheet in this cycle (one-way flows into US credit), as the Fed begins its unwind, at least at the margin?
Assessing rate risk, MS says that while the Fed may in fact be successful at threading the needle, an outcome that is likely already priced into markets. However, the bank warns that "at the least we can be certain that as the balance sheet shrinks more rapidly, so will the "liquidity buffer" in markets, which should magnify any negative catalyst that pops up along the way."
Another major risk factor for Morgan Stanley is that the US economy is now very late in the cycle, to wit:
Markets are very late cycle, in our view, and if anything these risks have risen compared to this time last year. That we are in a late-cycle environment is a consensus view, but "late cycle" can mean different things to different people. To be more specific, we think there is a good chance that markets peak for the cycle in 1H18 and price in rising defaults in a bigger way throughout the year. But even if our timing continues to be too early, remember, late-cycle environments are often not great for credit returns regardless, with equities often outperforming. (Note, as we discuss further below, we believe the very late-cycle signal where credit/equities diverge is already happening, focusing on CCC-rated HY credit.) A recession is not necessary for credit spreads to widen late in a cycle. In fact, credit markets have not had three straight years of positive excess returns since 1996.
Here Richmond takes offense with the argument that weak growth for much of this cycle has prevented "excesses" from building, and hence an already long cycle can last even longer. As he says "we disagree and see excesses all over the place, driven in part by years of ultra-low rates." He notes the following specific details:
- Credit markets have grown by 116% in this cycle, and leverage is at unprecedented levels for a non-recessionary environment.
- Low quality BBB issuance was 44% of total IG supply in 2017, a record as far back as we have data, and B rated or below loan issuance is now two thirds of total loan supply.
- LBOs levered over 6x are now a higher percentage of new LBO loans than in 2007. Covenant quality is considerably weaker than pre-crisis, while the debt cushion beneath the average loan is much lower.
- Investors have reached for yield in fixed income in this cycle in a massive way. Foreign flows have flooded into the asset class, arguably treating US credit as a rates product, while liquidity needs have risen, with mutual fund/ETF ownership of credit now over 19% vs. 11% pre-crisis.
- Excesses are apparent even outside of corporate credit, with underwriting quality deteriorating in auto lending in this cycle, while non-mortgage consumer debt is at a high, and CRE prices are ~25% above prior-cycle peaks.
- Stock-buyback activity has been substantial in this cycle, credit valuations have rarely been richer, and consumer confidence has not been this high since 2000.
Summarizing, and "cutting through the details" Morgan Stanley says that it has high conviction in the following two points:
- Excesses have to be out there, given what central banks have done in this cycle – i.e., rates near or below zero for nearly a decade and round after round of QE globally, and
- the excesses are always difficult to spot as markets are rising, and then become obvious after the turn (how did I miss that?). We think this time is no different. To be clear, excesses are not everywhere. For example, credit quality did not deteriorate in places like housing and US financials in this cycle. However, this simply tells us that the problems of the last cycle will not be the same as the problems of the next.
As a result, 2018 is when the critical mass of excesses finally spills over, or, to reuse the title, "the levee finally breaks":
While the excesses may be out there, that has arguably been the case for a while. The difference, we think, is that more cracks are now forming under the surface, which in our view, means a turn is closer than the consensus believes. For example, outside of corporate credit, we have seen signs of weakness and tighter credit conditions in places like commercial real estate. Consumer delinquencies are rising across products (i.e., autos, credit cards, and student loans). And in corporate credit, one sector after the next is exhibiting "idiosyncratic" problems (e.g., Retail, Telecom, and Healthcare to name a few). All of this is consistent with a late-cycle environment where the yield curve is flattening, correlations in markets are dropping, the economy is at (or arguably through) full employment, the Fed is well advanced in its tightening cycle (we think), and equity multiples are expanding.
To Richmond, these dynamics are "late-cycle 101. Problems pop up early on in the areas that experienced the most severe deterioration in fundamentals in the bull market. Investors initially treat those issues as "idiosyncratic." The problems then spread when credit conditions tighten more broadly. And along these lines, we think it is not a coincidence that weaker-quality high yield credits are underperforming, as the Fed is hiking faster and quantitative tightening is now being set in motion."
If that wasn't enough, Morgan Stanley highlights two further risks: one having to do with the incremental impact of tax cuts, should they pass...
And as a side note, tax cuts would not extend the cycle in our view – they risk doing the opposite. Very simply, credit markets will benefit from anything that keeps the cycle going – modest growth and a patient Fed. Tax cuts that come when the unemployment rate is 4.1%, which drives an overheating labor market, forcing a more aggressive Fed, if anything could cut off the cycle sooner.
... and the inevitable rise in default intensity:
We think there is a high likelihood that defaults will start rising again late next year and into 2019. Without going into the details here, in our view, CCC HY bonds are already "sniffing out" these budding default risks with their recent weakness. This should continue as tighter central bank policy exposes the fundamental challenges in the asset class (the problems are easier to hide when markets are flooded with liquidity). And the fundamental issues are broad-based. Not only is leverage high across sectors, but we also estimate that almost 30% of the HY market is either in secular decline or has clear operational challenges (Exhibit 16), with declining revenue growth over the past five years. Thinking about it more quantitatively, as we show in the default section below, based on the lag between when the cycle indicators we track have turned historically and when defaults have subsequently spiked, as well as the status of those metrics today, 2019 could be a year of materially higher defaults.
Wrapping up the above, Morgan Stanley's conclusion is the following:
Adding everything up, we see three key challenges in 2018: 1) Credit markets have been hugely reliant on central banks in this cycle, and now the Fed is withdrawing liquidity in an unprecedented way. We think markets are underestimating the risks of a mistake. 2) This liquidity withdrawal is happening while late-cycle risks (we think) are popping up all over the place. 3) Investors are buying credit at valuations that almost guarantee poor long-term returns, with the assumption that they will be able to time when to get out before the turn.
... or stated even simpler, "get out now."