"The pessimist complains about the wind; the optimist expects it to change; the realist adjusts the sails."
- William Arthur Ward, American writer
While looking at the numerous "sucker punches" delivered on Friday due to the "Brexit" results, we reminded ourselves for our title analogy for a subject we already touched back in January 2012, namely the "Optimism bias" which we touched in our conversation "Bayesian thoughts". Following the dismay of so many of our friends relating to the outcome and as well to both markets and bookmakers being all wrong at the same time, we reminded our friends and ourselves the following on this occasion:
"Brexit analysis simply explained: Optimism bias: One of the most consistent, prevalent, and robust biases documented in psychology and behavioral economics. Many tend to overestimate the likelihood of positive events, and underestimate the likelihood of negative events. For example, many underrate our chances of getting divorced, being in a car accident, or suffering from disease and overestimate probability on extreme long-shots such as winning the lottery."
- Source: Macronomics
Last week, when it came to assessing the potential outcome for the much dreaded referendum, we also indicated the following:
"For our take on "Brexit, we will keep it simple for our readers: From a game theory perspective and prisoner's dilemma, the only possible Nash equilibrium is to always defect. The United Kingdom "defecting" could mean, we think, taking business (and profits) from other European Union members in the long run. First mover advantage? Maybe..."
- Source: Macronomics, June 2016
While assisting in Paris to the "Brexit conference" set up by our friends at Saxo Bank, one of the members of the audience during the Q&A session pointed out the "accuracy" of the bookmakers for the remain to "prevail". We could not resist but intervene to rebuke that statement by using as an illustration how bookmakers got it so wrong when offering 5000/1 odds at the beginning of the season for FC Leicester to clinch the British football Premier League and still having the odds at 500/1 around October. The biggest liabilities for the bookmakers were accrued at around 100-1 to 500-1. To quote Mike Tyson: "Everyone has a plan 'till they get punched in the mouth". Since that "FC Leicester punch", the longest odds that can now be placed on any event will be 1,000-1 to ensure that the betting company Ladbrokes is less exposed in future to 'black swan' events. We reminded also the Saxo crowd of the Nash equilibrium concept, us playing on this occasion the "Devil's advocate". In fact, not a single time did the bookmakers anticipated a victory for "Brexit" - yet another display of the "Optimism bias" as seen in the below chart from the following The Telegraph article "Why the 'experts' failed yet again to call which way Britons would vote" from the 25th of June:
Source: The Telegraph
Parsing through the markets various "reactions" on Friday akin to a "deer caught in the headlights" kind of moment, us being contrarians, we were ready for the "sucker punch" on our "gold miners" exposure (which we have been advocating for a while for those who follow us...). It looks like indeed our "pessimism bias" was this time around the "lucky approach". But at this juncture, dear readers, before we go into the "nitty gritty" of this week's conversations, we think it is is important for us to "illustrate" our core philosophy which we have nurtured in recent years by asking a simple question: Why people get lured into making inaccurate conclusions?
"What affect one's judgment in today's world markets?", one might rightly ask.
We think the below four points resume our core contrarian "philosophy":
- Herd mentality: People are influenced by their peers to adopt certain behaviors, follow trends, and/or purchase items. Examples of the herd mentality include stock market trends and fashion.
- Information cascade: One of the topics of behavioral economics often seen in financial markets, where they can feed speculation and create cumulative and excessive price moves, either for the whole market (market bubble...) or a specific asset, like a stock that becomes overly popular among investors.
- Dunning-Kruger effect: A cognitive bias in which unskilled and inexperienced individuals suffer from illusory superiority, mistakenly rating their ability much higher than average.
- Optimism bias: One of the most consistent, prevalent, and robust biases documented in psychology and behavioral economics. Many tend to overestimate the likelihood of positive events, and underestimate the likelihood of negative events. For example, many underrate our chances of getting divorced, being in a car accident, or suffering from disease and overestimate probability on extreme long-shots such as winning the lottery.
Of course, Friday's shock result was clearly illustrated by the last point, hence our title analogy, given we have already used as title analogies most of the above points in previous musings. Therefore, challenging the "consensus" is like betting in a race horse on the outsider - the returns, if achieved, will be significantly higher.
For instance we have told you before that Government bonds were always correlated to nominal GDP growth, regardless if one looks at it using "old GDP data" or "new GDP data". So, if indeed GDP growth continues to be weak, then one should not expect yields to rise anytime soon. As a matter of fact, in January 2014, we recommended and bought very long duration exposure on US Government bonds using PIMCO ETF ZROZ, when nearly all experts were calling for higher US yields by the end of 2014. Very few such as us, Dr. Lacy Hunt and Jeff Gundlach, called it differently. The ETF ZROZ finished the year 2014 as the best-performing ETF in the Fixed Income space, gaining more than 44% in US dollar terms, but that's another story...
While everyone is still reeling on the "Black Swan" outcome from the Brexit vote, we watched with interest US Durable Goods Orders cratering further by 2.2% in May (versus -0.5% expected). Not only are U.S. firms already cutting back on their capital expenditures, but we would also expect weaker nonfarm payrolls going forward, neutering, in effect, the "recovery" story and stopping in its tracks the hiking process of the Fed, which is reinforcing even more our appetite for US long-duration exposure and, of course, validating our gold miners exposure.
In this week's conversation, we will focus on the productivity puzzle and its impact on economic growth leading to the ongoing "secular stagnation", we will as well look at additional pointers on deterioration of credit metrics and why the trend, in the medium term, is not your friend.
- Macro and Credit - Secular stagnation? It's the productivity, stupid!
- Macro and Credit - Our "pessimism bias" make us continue to prefer the safety of US High Grade
- Final chart: Global Non-Financial Corporate Debt to GDP is "off the charts"
Macro and Credit - Secular stagnation? It's the productivity, stupid!
On numerous occasions on this very blog, we have pointed out our lack of "optimism bias" towards the much vaunted "recovery story" being sold mostly by pundits due to the lack of "wage growth", as indicated in our conversation "Perpetual Motion" from July 2014. We argued that real wage growth was indeed the "most important" piece of the puzzle the Fed has so far been struggling to "generate":
"Unless there is an acceleration in real wage growth we cannot yet conclude that the US economy has indeed reached the escape velocity level given the economic "recovery" much vaunted has so far been much slower than expected. But if the economy accelerates and wages finally grow in real terms, the Fed would be forced to tighten more aggressively."
- Source: Macronomics, July 2014
But, there is more to it when it comes to the theory of "secular stagnation", and this has all to do with the lack of "productivity", it seems. On that very subject, we read with interest Bank of America Merrill Lynch's take from their Global Economic Weekly note from 3rd June, entitled "The global productivity puzzle":
"The global productivity puzzle
• Labor productivity growth has slowed significantly across developed markets over the past half-century; it currently is at record-low levels.
• Weak total factor productivity (TFP) growth is the main reason; low capital investment or declining labor quality only account for a small portion of the slowdown on average.
• There are many potential explanations for weak productivity but much of the decline remains a puzzle, with no clear policy offset.
Pronounced, and potentially persistent
Growth in most developed markets (DM) has been disappointingly slow, while emerging markets have seen their growth rates decelerate each of the past five years. Many observers have gradually come to the conclusion that the supply side of the global economy appears to have been damaged - perhaps permanently. The clearest evidence of this phenomenon is the sharp reduction in labor productivity - total output divided by total hours of labor input - in most developed markets (and many emerging markets) over the past several years. In fact, DM labor productivity has converged to historically low, sub-1% growth rates (see Chart 1 and Chart 2 for G10 country data). More significantly, the decline appears to have begun before the global financial crisis hit.
These two observations have important implications. First, it strongly suggests that common global factors may be responsible for the worldwide decline in productivity growth. Second, the global financial crisis is unlikely to be the main, let alone the only, explanation - although persistent spill-overs or hysteresis (cyclical shortfalls that become structural) from the crisis may be playing some role in holding back the supply side globally. Indeed, one of the troubling aspects of this analysis is that the causes of the productivity slowdown remain elusive, despite a myriad of potential interpretations.
Notably, some of these are much more persistent than others, and some are more amenable to policy fixes. Together, these imply the global economy isn't doomed to low productivity forever, but it could drag on for quite some time.
Lower everywhere you look
The simplest and most direct measure of productivity is output divided by hours worked. The more output that can be produced with a given amount of labor input, the more productive each hour of labor is. Conversely, if hours are growing faster than output, labor productivity must be slowing down. Such slowing has been a common occurrence over the past few decades among developed economies (and more recently among emerging markets). Chart 1 plots a smoothed measure of labor productivity growth (estimated by locally weighted regressions) for each of the G4 economies since 1951. All are appreciably lower during the last 5 to 10 years than at any time in the prior 55 to 60 years. Chart 2 is a similar plot for several other European economies.
In the US, productivity growth slowed notably in the 1973-1995 period, then accelerated for about a decade before slowing down more significantly from around 2004 until today. Japan and Germany had notably higher labor productivity growth in the aftermath of the Second World War but recently have converged toward US levels. Britain started slower but its labor productivity surpassed the US pace from the late 1960s until around 2000, before diving below zero recently. Chart 2 shows a similar pattern for other DM economies, with labor productivity growth generally peaking at some point in the 1960s and declining ever since; Sweden and Switzerland experienced some stability in the 1990s and early 2000s before declining further. Like the UK, Italy's measured labor productivity has turned negative on average recently.
Accounting for the growth slowdown
A capital concern
To understand the factors that have slowed productivity growth, economists often engage in "growth accounting." One way to increase labor productivity is to give workers more or better capital (tools, equipment, computers, etc.) to work with - so called "capital deepening." As investment spending has been weak in the post-crisis period, that seems like an obvious place to look for an explanation for the slowdown in labor productivity. Chart 3 illustrates how much a decline in capital deepening has impacted labor productivity growth across the eleven G10 DM economies. Note that due to data limitations, we analyze annual data from 1994 to 2014.
All eleven economies experienced a decline in productivity growth in the 2005-2014 period relative to the prior decade, by nearly 1.1pp on average. And, in most of them, the contribution from capital deepening (the sum of the orange and green bars) declined as well - but only by less than 0.3pp on average. Japan experienced the largest decline (60% of the fall in labor productivity growth can be attributed to less capital deepening), followed by the UK and the US (although only about 40% of the productivity slowdown); in Canada capital deepening actually grew slightly. All told, only a relatively small proportion - about 25% - of the slowdown across DM in labor productivity growth over the past two decades is due to slower growth in the amount of capital per worker.
A related hypothesis is that investment in information and communication technology (ICT) capital boosted labor productivity in the late 1990s and early 2000s, but that impact has since faded. Chart 3 also separates capital deepening into ICT (green) and non-ICT (orange) components. Some countries experienced measurable slowdowns in ICT capital deepening - most notably France, the Netherlands and the UK - while several others - Germany, Canada, Sweden and Belgium - actually saw a rise. Thus there is no systematic relationship between ICT investment and the decline in labor productivity among the G10 economies since the mid-1990s.
We don't need no education?
A second underlying factor that could change labor productivity would be changes in the quality of the labor force. A more skilled or educated labor force is likely to be more productive, all else equal. Chart 3 additionally contains estimates of "labor quality" from the Conference Board. Some analysts have speculated that more skilled workers are hired early in a recovery, and as it progresses the average skill level of newly hired workers goes down. That dynamic might help account for lower labor productivity recently versus a few years ago, but cannot readily explain the decadal differences in Chart 3. That said, with the exception of the UK, the decline over time has only shaved 0.1 to 0.2pp from labor productivity growth. Hence, changes in the composition of the labor force are not a significant factor for lower trend productivity." - source Bank of America Merrill Lynch.
What we find of interest on the above statement relating to "education" is clearly that the "student debt growing bubble" has not translated, we think, in an increase in the quality of the labor force. This can be seen on a monthly basis through the BLS data relating to employment components and education, as per our below updated chart:
Source: Macronomics / BLS - June 2016 update
So, you will excuse our "pessimism bias", because when it comes to "student loans debt", it's "much ado about nothing" in our book, hence our lack of belief in education translating in "productivity labor growth". It hasn't happen and will not happen.
When it comes to explaining the "productivity" conundrum, Bank of America Merrill Lynch, in their note, try to put forward several possible explanations on the subject:
The part of labor productivity growth that cannot be explained by factor inputs (as above) is called "total factor productivity" (TFP). Conceptually, TFP can take a variety of forms, including technical innovations, gains in efficiency of operations, management style changes, etc. Practically, TFP is also known as the "Solow residual," named after the prominent macroeconomist Robert Solow, the father of growth accounting. As the name implies, TFP is what is "left over" after other observable supply-side factors are accounted for. Significantly, the decline in measured TFP accounts for most of the decline in labor productivity for the DM countries in Chart 3. More generally, variation in TFP tends to account for most of the variation in labor productivity both over time and across countries. What has led to a decline in TFP is the key question.
There are a few big theories about why TFP growth has slowed over time across DM, listed in order from the most persistent - ie, the most likely to result in long-run low
TFP growth - to the least:
• No more "low-hanging fruit." The big economic innovations have happened already according to this argument, and inventions today are simply not as significant as (say) electrification or penicillin or the internal combustion engine.
This is a structural story that has an "end of history" flavor to it: the internet is just a fancy telegraph; nothing transformative here. Growth will stay low in this view; better get used to it.
• Consumption over production. In this argument, innovation has shifted from supporting more efficient production to more intense consumption. Mobile phones, ubiquitous cellular service - these are used for casual gaming and sharing cat videos, rather than pushing out the production possibilities frontier. That doesn't necessarily preclude a more productive use down the road. But to the extent this too is a structural shift, it likely means TFP growth remains lower for longer.
• Diffusion dynamics. Important technological development is still happening; it just takes time for its effects to show up on the shop floor and in the data. Firms are still learning how to most effectively utilize mobile, cloud, sharing, etc, to raise TFP. Robotics, genetic engineering, the "internet of things" - these innovations will raise productivity (and with it trend growth) at some point in the future.
• Mismeasurement. Productivity growth already is high, for all the reasons mentioned above - it just isn't measured correctly. Output excludes many free services because they aren't priced; significant quality improvements aren't fully captured. In this view, the statistics will eventually catch up to reality. Meanwhile, in this view real output is higher and inflation lower than commonly reported. The future's so bright, you've got to wear shades.
The first two more persistent slow growth stories may have some ring of truth, but history is littered with prior assertions that productivity growth had ended - only to be proven unduly pessimistic. However, if TFP is embedded in the capital stock and investment remains low, a self-reinforcing adverse feedback loop could arise: weak productivity creates few incentives to invest which perpetuates weak productivity. Indeed, there is a risk that such a feedback loop is already in place. The third explanation does have recent history on its side: the US (as well as some other DM economies) experienced a significant slowdown in TFP growth in the 1980s as the personal computer age began, but then saw more rapid growth from the mid-1990s until the mid-2000s as the returns to ICT investment were realized. Economic historians have found related evidence that the introduction of the steam engine into manufacturing did not materially improve productivity until the production processes were altered to better take advantage of this new invention. This diffusion process took several decades in that case. Something similar may be occurring with today's latest technological innovations, such as mobile and cloud computing.
The mismeasurement story, on the other hand, has some optimistic appeal but is hard to reconcile with the data. A recent in-depth study finds evidence of mismeasurement of technological innovation, but this has not changed over the past few decades while,the share of such products in domestic production has declined thanks to offshoring. As a result, mismeasurement actually exacerbates the productivity slowdown rather than explains it. Meanwhile, the combination of low wage and price inflation suggests that weak demand is as much at play as weak supply. Strong productivity growth historically has produced strong real wage growth, yet real wages are generally depressed in DM."
- Source: Bank of America Merrill Lynch
What we have seen with "the rise of the robots" and the very aggressive "cost cutting exercise since the Great Financial Crisis undertaken by many large corporations is that the productivity slowdown has indeed been exacerbated, leading to non-existing wage growth and depressed "real wages" à la Japan, as discussed in our previous Macro and Credit conversation.
So what are the implications for policymakers around the world given the continued rising of populism? Bank of America Merrill Lynch, in their note, give us some sobering indications:
Implications for policy
Weak productivity growth translates into weaker long-run economic growth, which means it is harder to grow out of budgetary shortfalls or debt burdens. Coupled with slowing population growth and declining worker-to-retiree ratios, it means more stress on social safety nets. For monetary policy, slower productivity growth will tend to mean output gaps close more quickly for any given pace of recovery, which will force central banks to normalize policy sooner or risk overshooting on consumer and/or asset prices. It also means a slow, drawn out recovery cannot be remedied merely with low rates and large-scale asset purchases.
Fiscal policy may be able to help jumpstart faster productivity growth by encouraging innovation through product and/or labor market reforms, but these take time and tend to be politically divisive. Investment in public infrastructure also may be supportive. Evidence strongly supports the idea that global competition via open trade boosts productivity. Unfortunately, around the world both the volume of trade, and the political support for promoting it, appears to be waning. This, too, could be an avoidable factor that is helping to push down productivity growth across countries. But without a clear diagnosis of the reasons for the decline in productivity, it is difficult to suggest a set of feasible policies to counteract it."
- Source: Bank of America Merrill Lynch
Furthermore, for the "Optimists" crowd, we think the weak productivity situation should not be taken lightly. This is clearly reiterated in the most recent BIS annual report published in June:
"Less comforting is the longer-term context - a "risky trinity" of conditions: productivity growth that is unusually low, global debt levels that are historically high, and room for policy manoeuvre that is remarkably narrow. A key sign of these discomforting conditions is the persistence of exceptionally low interest rates, which have actually fallen further since last year."
- Source: BIS, 86th Annual Report
In our book, "secular stagnation" is not only due to the burden of high global debt levels, but as well by the evident slowdown in productivity labor growth, which is clearly impacted by the "rise of the robots". This does not bode well for the stability of the "social fabric" and with rising populism in many parts of the world.
Back in November 2014, in our conversation "Chekhov's Gun", we argued the following:
"Our take on QE in Europe can be summarized as follows: Current European equation: QE + austerity = road to growth disillusion/social tensions, but ironically, still short-term road to heaven for financial assets (goldilocks period for credit)…before the inevitable longer-term violent social wake-up calls (populist parties access to power, rise of protectionism, the 30's model…).
"Hopeful" equation: QE + fiscal boost/Investment push/reform mix = better odds of self-sustaining economic model / preservation of social cohesion. Less short-term fuel for financial assets, but a safer road longer-term?
When it comes to the Current European equation, we note with interest that civil unrest is a rising global trend."
- Source: Macronomics, November 2014
Obviously, our "Hopeful equation" suffered from "Optimism bias", and as we argued at the time:
"Our "Hopeful" equation has a very low probability of success given the "whatever it takes" moment from our "Generous Gambler" aka Mario Draghi which has in some instance "postponed" for some, the urgent need for reforms, as indicated by the complete lack of structural reforms in France thanks to the budgetary benefits coming from lower interest charges in the French budget, once again based on phony growth outlook (+1% for 2015) "
- Source: Macronomics, November 2014
Increasingly, it looks to us that we are moving towards the inevitable longer-term violent social wake-up calls (populist parties' access to power, rise of protectionism, the '30s model…). That's our take and our "pessimism bias" for now.
When it comes to "asset allocation" in the current "risk-off" environment, we continue to like long-duration, high-quality credit, such as US High Grade. It benefits from the rally in the "greenback" (US dollar) as well as higher yields than its European peers. More on this in our next bullet point.
Macro and Credit - Our "pessimism bias" make us continue to prefer the safety of US High Grade
While we have long advocated going for "quality" and indicated that US Investment Grade credit was the only "game in town", particularly for Japanese investors, the current "risk-off" environment is favoring the safety of the US Dollar Index given today, as we write our latest musing, we have seen the largest two-day increase since 1992 - a "cool" 5-sigma event. This is what you get with central banks meddling with asset prices for too long: rising positive correlations leading to significantly large standard deviation moves. This was highlighted in our "lengthy" but nonetheless must-read February post on risk and VaR (Value at Risk) conversation, "The Disappearance Of MS München".
What is currently being priced in the US government bond market contrary to the Fed's recent stance is more "easing" rather than "hiking", we think, hence the relative safety and comfort in US Investment Grade credit from an allocation perspective. On this subject, we agree with Bank of America Merrill Lynch's take from their Credit Market Strategist note from 24th June, entitled "European divorce":
"More monetary policy easing
Behind today's plunge in global yields lies - in addition to lower global growth - the expectation that Brexit will prompt more monetary policy easing from global central banks. Included in this is that we now expect the Bank of England (BOE) to lower rates and start engaging in QE this summer. Brexit is also expected to dampen the Fed's rate hiking cycle - case in point right now the Fed funds futures market is pricing in equal probabilities of a rate hike and a rate cut at the next FOMC meeting in July (12%).
However, clearly since foreign countries lead the decline in economic growth the net result is incrementally more dovishness abroad than in the US.
That explain why US yields remain high relative to European yields, even though the absolute level of US yields is coming down (Figure 9), which should continue to attract European buying of US corporate bonds.
Moreover, on the Japanese side US corporate bonds are now more than ever the only game in town. Specifically Japanese corporate yields are now 0.13% and the situation is not much better in 30-year JGBs yielding 0.15%. However, the yield on a fully currency hedged basis for a Japanese investor buying a 10-year US senior bank bond is 0.80%, or 5-6 times as high (Figure 10)
Corporate yields can decline
A defining aspect of the big sell-off in the beginning of the year was the stability of corporate yields - hence as Treasury yields plummeted spreads blew out almost in the relation of one-to-one. That happened as both domestic and foreign investors stepped to the sidelines with the simultaneous decline in US yields and yields relative to foreign fixed income (Figure 9). However, because the weakness this time around is driven more by expected European weakness, as explained above, our market is not becoming less attractive to foreign investors. This means that, even though yield sensitive domestic investors may now become very defensive we should continue to attract significant foreign demand. We think that the Brexit surprise means that US corporate yields can now decline further - which is what we saw in today's post-Brexit reaction - both in the USD and EUR markets.
Credit outperforming on Brexit
Given the low Brexit probability priced by the market by the end of Thursday the reaction on Friday was severe across most markets. In high grade credit, on the other hand, the move wider in spreads was rather underwhelming. In particular banks and some industrial spreads are actually tighter today following the actual vote for the "leave" camp than the wides reached last week in response to just the risk of Brexit.
Moreover, highlighting the strength in credit and unlike February, Friday's widening in high grade credit spreads was less than the decline in rates, pushing corporate yields lower in both in the US and in Europe.
On Friday, following the results of the UK EU referendum, the Sterling was down 8.2% (a 15 standard deviation move for the period Jan 2010 to the present), European stocks sold off 8.6% and iTraxx Main closed 18bps wider. Naturally the shock spilled into US markets as well, pushing the yield on the 10-year Treasury down 19bps to 1.57% and stocks down 3.6% on the S&P 500, including 5.4% for the financial sector (Figure 1).
- Source: Bank of America Merrill Lynch
Having exposure to US High Grade Credit in this ongoing "Japanification" process can indeed dampen the volatility experienced in various asset classes, while US credit still boasts more favorable carry thanks to higher yield and roll-down compared to European credit, which has tightened even more dramatically thanks to ECB meddling with Corporate credit as of late. Furthermore US Investment Grade credit still benefits from favorable flows, unlike equities and High Yield.
While we have been, in recent months, describing the slow deterioration in the credit cycle, "Brexit" or not being the catalyst for the recent bout of "sell-off", it was our belief in the US economy being weaker than expected that has enticed us again this year to play, à la 2014, the US long duration play (partly once more via ETF ZROZ). We keep telling you that we are indeed in the last inning of the credit game, and as per our last conversation, we'd rather focus on the big picture being tighter financial conditions and deterioration in macro data rather than on the "political fallout" stemming from "Brexit". Once again, like any behavioral psychologist, we tend to focus on the process rather than on the content. When it comes to advising you to reach for quality in 2016 credit-wise, we also agree with Bank of America Merrill Lynch's High Yield team when it comes to assessing the credit picture. For instance, we read with interest their High Yield Wire note from 20th June entitled "Bifurcate or break: the investment conundrum of a rolling blackout":
"Laelaps and the Teumessian fox
In Greek mythology the Teumessian fox's great power was to always be able to escape its hunter. Laelaps the dog, on the other hand, charged with catching the fox, had an equally great power: to always catch its prey. And with such simplicity a great paradox was created: how can one who catches everything catch something that can never be caught? The answer? It can't. In the case of Laelaps and the Teumessian fox, Zeus turned both into stone. We find great meaning in the story … what happens to an economy that can't grow or generate inflation when it is met with a central bank that, although has made mistakes in the past, seems to always right the economy eventually? Like Zeus, who turned our main characters to stone, do the modern day fox and dog effectively become paralyzed in their current states of low growth and accommodation? And if so, what does this mean for markets?
We believe the credit cycle is currently in its latter innings as the growth process has arguably stalled if not begun to slip backwards. Although Draghi, Kuroda and Yellen are doing their best, monetary policy can only be as useful as fiscal policy allows. And although our belief is that eventually deteriorating corporate earnings will lead to layoffs, a collapse in consumer confidence and spending and ultimately a recession, we also believe that an increasingly compelling case can be made that in between bouts of volatility our rolling blackout scenario may become the norm for much longer than any of us can anticipate. Under such circumstances the most levered companies with the poorest earning potential, those disrupted by technological innovation, and issuers relying on constantly open capital markets would trade at distressed-type levels. Meanwhile, those companies with consistent capital market access, strong management structures and who are financially nimble effectively would trade at some liquidity premium and very little credit risk premium, likely becoming targets for IG M&A. Effectively, valuations in high yield could become even more binary than they are today: trading at distressed levels or like an investment grade credit with a bit of extra liquidity premium.
Under such a universe, we think the case can be made that bigger is not better. In fact, we would argue that smaller high yield, high quality corporates that are attractive M&A candidates for IG companies would be compelling vs. a large BB and high quality single B credit that has a massive balance sheet and is susceptible to shocks. For those who must play high yield, believe in our barbell strategy, and are looking for index-level yields and returns, own small-cap BBs (along with a cash position in treasuries) with smaller CCC risk in issuers that have "already realized" their event.
The bigger the tree the harder the fall
With our view that bigger companies are susceptible to risks and shocks, and that default risk has the potential to be a constant presence - even if a subdued one - for a long period of time, we are often told that the market is safer today because it's more a "real" market relative to pre-crisis. "Real", often defined as a bigger universe of names, larger issuers, more recognizable brands and higher EBITDA is a term we would hesitate to use. And although there is no doubt that the high yield market today is not only larger than it has ever been (it has grown in size by 90% since 2008), we would challenge the assertion that the high yield market is any safer because of growth. We prefer to judge markets more on their "maturity" and would focus more on the increase in the base of investors, breadth of industries, as well as the degree of price transparency and liquidity than on size. Clearly the ability for capital markets to structure deals in difficult times is also crucial to defining credit market maturity and having investors up and down the capital structure is also important. When considering this definition, we find that today's market is not significantly more "real" than it has been in the past. Although we do benefit from increased price transparency because of the documentation of TRACE activity, the high yield market today still shows a relatively small investor base and high concentration among the top 3 industries (table 2):
Mutual fund ownership peaked in 1998 and - absent a brief increase in 2012 - has been on a consistent decline ever since, allowing institutional funds to gain greater market share and concentrate ownership into the hands of a few large investors (Chart 2).
The same story holds true as it relates to sectors, where in today's market Energy, Healthcare, and Telecom make up 37% of US HY by face value. This is not all that different from Telecom, Media, and Materials' 42% representation in 2000 and suggests the lack of industry breadth present in today's market. We do not deny that the development of TRACE in 2005 has led to significantly greater price transparency and was the harbinger of future technological developments, including our own Instinct® Loans e-platform that was recently launched. However, these financial innovations have been coupled with increased regulatory burdens and more balance sheet restrictions, and we have actually seen lower trading volumes (Chart 3) and little change in bid/ask spreads since 2008. In fact, some would argue TRACE has hurt liquidity, as the transparency into small transactions makes it difficult to execute larger transactions discreetly.
Additionally, we believe the ghosts of markets past coupled with indebtedness never before seen will likely cause financial burdens that are not fully appreciated. Our view is that central banks have effectively obfuscated the true health of the global corporate markets and absent a significant increase in growth - something we question accommodative policy's ability to spark - this cycle is unlikely to look meaningfully different than past cycles when the business cycle ultimately turns. In fact, when coupled with the growth in commodity issuance in the post-crisis years, the lingering impact of 2006 and 2007's LBOs and the lack of earnings power outside of just a few sectors, we argue that in real terms, the US high yield market today looks remarkably similar to how it looked like during the Clinton and Bush years.
This is in contrast to the view that more mature capital markets and larger companies will cushion the next cycle. In fact, we think we could make just the opposite case - that large serial issuers, fueled by cheap debt and poor organic earnings prospects will likely prove a problem when the cycle turns, as history suggests size is not a good determinant for defining credit or default risk. Additionally, the biggest high yield issuers are unlikely to spur any targeted M&A, sponsor or strategic deals just given their size and bloated balance sheets. To this end, high yield is no more a "real" market than it has ever been. Have some aspects matured? Yes. Do traditional size-based measures indicate a healthier investing landscape? No."
- Source: Bank of America Merrill Lynch
While some might infer with their "Optimism bias" that eventually things will turn out alright, we beg to disagree. In this long credit cycle fueled by overzealous central bankers, we believe that when the cycle will turn in earnest with US slowly grinding towards recession, the default rate will be much larger and recoveries will, of course, be much lower. That's a given.
What makes us have this "Pessimism bias", you might rightly ask? How about our final chart below?
Final chart: Global Non-Financial Corporate Debt to GDP is "off the charts"
Our "Pessimism bias" stems from the very high leverage of the non-financial sector on a global scale, and, as pointed out before, a lot of Emerging Markets corporate debt has been issued is US dollar-denominated. We believe this is a recipe for disaster, as previously highlighted by the BIS as well as by our friends from Rcube. Our final chart comes from Deutsche Bank's Credit Bites note from 8th June, entitled "A no excess cycle? Not true in corporates":
"Figure 1 looks at global non-financial corporate debt/GDP based on the BIS dataset and definitions. Whilst it's not always easy to categorise debt into various buckets and whilst we have some issues with slightly different results across different data providers it's fair to say that the BIS data is the best way of looking at the global debt picture.
So non-financial debt has increased significantly in this cycle. On this measure the increase is of the magnitude of around $15 trillion since the lows after the financial crisis."
- Source: Deutsche Bank
So, in a world plagued by low productivity labor growth and high level of debt, we wonder how some pundits can continue with their "Optimism bias". We'd rather stick to our "Pessimism bias and play the "minimax principle":
"A principle for decision-making by which, when presented with two various and conflicting strategies, one should, by the use of logic, determine and use the strategy that will minimize the maximum losses that could occur. This financial and business strategy strives to attain results that will cause the least amount of regret, should the strategy fail."
- Source: businessdictionary.com
It is isn't a game of capital appreciation, but it certainly becoming one of capital preservation, we think...
"A pessimist is a man who tells the truth prematurely."
- Cyrano de Bergerac