Macro and Credit - Harmonic tremor

"Stupidity is an elemental force for which no earthquake is a match." -  Karl Kraus, Austrian Writer

Watching with interest the tragedy unfold on the short gamma crowd through the demise of the short volatility ETN and ETF complex, sending markets into some tailspin and causing additional havocs in bond yields, when it came to choosing our title analogy for our post we reminded ourselves of the meaning of "Harmonic tremor". A "Harmonic tremor is a sustained release of seismic and infrasonic energy typically associated with the underground movement of magma (rising term premiums), the venting of volcanic gases from magma, or both. It's a long-duration release of seismic energy (volatility), with distinct spectral lines, that often precedes or accompanies a volcanic eruption (the demise of some short vol ETNs). More generally, a volcanic tremor is a sustained signal that may or may not possess these harmonic spectral features. Being a long-duration continuous signal from a temporally extended source, a volcanic tremor contrasts distinctly with transient sources of seismic radiation, such as tremors that are typically associated with earthquakes and explosions. Harmonic tremor is part of the four major types of seismograms, the three others being tectonic like earthquakes, shallow volcanic earthquakes and surface events. 

In this week's conversation, we would like to look at the recent sell-off which in effect was triggered by the harmonic tremor coming from the uninterrupted rise of term premiums. This regime change in volatility and the effect of "Who's Afraid of the Big Bad Wolf?" aka "inflation expectations" thanks to rising wage inflation expectations have already claimed the small fishes such as some players in the short volatility leveraged and crowded complex. Like a new grain of sand U.S. Average Hourly Earnings triggered an avalanche as seen in complex systems such as financial markets.

Synopsis:

• Macro and Credit - Rising term premium have already claimed small fishes, when are the whales going to show up?
• Final chart - Increased inflation presents a potential threat to the purchasing power of consumers.

• Macro and Credit - Rising term premium have already claimed small fishes, when are the whales going to show up?

We had hardly pressed the "publish" button for our previous post, that the events taking place in the volatility complex led to the already well publicized demise of some short volatility ETNs. This was bound to happen given the non-linearity aspect one can find in financial markets. 

The events that took place were fascinating as it was indeed yet another confirmation of our musing from February 2016 conversation "The disappearance of MS München" when we were discussing the fascinating destructive effect of "Rogue waves" on man-made "structures". Those waves have a high amplitude and may appear from nowhere and disappear without a trace, or investors in some instances. Generally rogues waves require longer time to form (like "harmonic tremors" led explosions), as their growth rate has a power law rather than an exponential one. While a 12-meter wave in the usual "linear" model has a breaking force of 6 metric tons per square meter (MT/m2), modern ships are designed to tolerate a breaking wave of 15 MT/m2, a rogue wave can dwarf both of these figures with a breaking force of 100 MT/m2. Remember this. 

Once again we would like to come back to our February post of 2016 and quote the book "Credit Crisis" authored by Dr Jochen Felsenheimer (which we quoted on numerous occasions on this very blog for good reasons) and Philip Gisdakis and steer you towards chapter 5 entitled "The Anatomy of a Credit Crisis" page 215 entitled "LTCM: The arbitrage saga" and the issue we have discussing extensively which is our great discomfort with rising positive correlations and large standard deviations move. This amounts to us as increasing rising instability and the potential for "Rogue Waves" to show up in earnest like the one experienced last Monday:

"LTCM's trading strategies generally showed no or almost very little correlation. In normal times or even in crises that are limited to a specific segment, LTCM benefited from this high degree of diversification. Nevertheless, the general flight to liquidity in 1998 caused a jump in global risk premiums, hitting the same direction. All (in normal times less-correlated) positions moved in the same direction. Finally, it is all about correlation! Rising correlations reduces the benefit from diversification, in the end hitting the fund's equity directly. This is similar with CDO investments (ie, mezzanine pieces in CDOs), which also suffer from a high (default) correlation between the underlying assets. Consequently, a major lesson of the LTCM crisis was that the underlying Covariance matrix used in Value-at-Risk (VaR) analysis is not static but changes over time." - source Credit Crises, published in 2008, authored by Dr Jochen Felsenheimer and Philip Gisdakis

You might probably understand by now from our previous sailing analogy (The Vasa ship) and wave analogy (The Ninth Wave) where it will eventually end: Another financial crisis. 

Moving back to the LTCM VaR reference, the Variance-Covariance Method assumes that returns are normally distributed. In other words, it requires that we estimate only two factors - an expected (or average) return and a standard deviation. Value-at-Risk (VaR) calculates the maximum loss expected (or worst case scenario) on an investment, over a given time period and given a specified degree of confidence. This what we had to say about VaR in our May 2015 conversation "Cushing's syndrome" and ties up nicely to our world of rising positive correlations. Your VaR measure doesn't measure today your maximum loss, but could be only measuring your minimum loss on any given day.

Check the recent large standard deviation moves dear readers such as the one experienced on the VIX last Monday and ask yourself if we are anymore in a VaR assumed "normal market" conditions:

"On a side note while enjoying a lunch with a quant fund manager friend of ours, we mused around the ineptness of VaR as a risk model. When interviewing fellow quants for a position within his fund, he has always asked the same question: What does VaR measures? He always get the same answer, namely that VaR measures the maximum loss at any point during the period. VaR is like liquidity, it is a backward-looking yardstick. It does not measure your maximum loss at any point during the period but, in today "positively correlated markets" we think it measures your "minimum loss" at any point during the period as it assumes "normal" markets. We are not in "normal" markets anymore rest assured." - source Macronomics, May 2015

But, last Monday's move while surprising by its velocity, has not been as significant as the move seen in the VIX back in 1987. We are yet to see a spike to 173 seen on the VIX during the October 1987 crash and that was something, really something.

This is also what we argued in February 2016:

If you think diversification is a "solid defense" in a world of "positive correlations", think again, because here what the authors of "Credit Crisis" had to say about LTCM and tail events (Rogue Waves):
"Even if there are arbitrage opportunities in the sense that two positions that trade at different prices right now will definitely converge at a point in the future, there is a risk that the anomaly will become even bigger. However typically a high leverage is used for positions that have a skewed risk-return profile, or a high likelihood of a small profit but a very low risk of a large loss. This equals the risk-and-return profile of credit investments but also the risk that selling far-out-of-the-money puts on equities. In case of a tail event occurs, all risk parameters to manage the overall portfolio are probably worthless, as correlation patterns change dramatically during a crisis. That said, arbitrage trades are not under fire because the crisis has an impact on the long-term-risk-and-return profile of the position. However, a crisis might cause a short-term distortion of capital market leading to immense mark-to-market losses. If the capital adequacy is not strong enough to offset the mark-to-market losses, forced unwinding triggers significant losses in arbitrage portfolios. The same was true for many asset classes during the summer of 2007, when high-quality structures came under pressure, causing significant mark-to-market losses. Many of these structures did not bear default risk but a huge liquidity risk, and therefore many investors were forced to sell." - source Credit Crises, published in 2008, authored by Dr Jochen Felsenheimer and Philip Gisdakis

You probably understand by now why we have raised the "red flag" so many times on our fear in the rise of "positive correlations". They do scare us, because they entail, larger and larger standard deviation moves and potentially trigger "Rogue Waves" which can wipe out even the biggest and most reputable "Investment ships" à la MS München." - source Macronomics, February 2016

Obviously, for us rising term premiums have been like a "harmonic tremor" leading to the buildup that ended with the explosion that occurred in the short volatility space. The catalyst has been rising inflation expectations coming from the latest U.S. Average Hourly Earnings rising 2.9% Y/Y, the most since 2009. This was the little grain of sand that triggered, we think Monday's avalanche. 

If Short Vol ETNs could be compared to a piece of a CDO like structure for the financial markets complex, they could be seen as the "Equity tranche", or the first loss piece of this "capital structure" (The tranche that absorbs the first loss (and thus is the most risky tranche) is often called an equity tranche). Obviously the next question one would like to have answered when looking at his portfolio, is am I "senior" enough in the capital structure and is my attachment point high enough to avoid the pain? 

In relation to the issue of rising term premiums and additional potential pain and "de-risking" coming we read with interest Nomura's Global Markets Navigator note from the 7th of February entitled "Rising term premium claims its first victim":

"According to the NY Fed, the 10yr UST term premium troughed in early December at -0.62bps. As of 5 February, the NY Fed estimates that it has risen 33bps to -0.29bps.

Over the same period, the MOVE Index of implied treasury volatility rose in sympathy from an all-time index low of 48 to today’s 66. In previous reports we have linked low term premia to low implied interest volatility and in turn low implied volatilities in risk assets, e.g., equities.

Front-end VIX started to rise from the first day of 2018 trading, lagging behind the increase in the term premium. It was toward the end of January that both measures began to move faster. The recent few days of trading needs little rehashing.

And yet other risk assets and measures of risk aversion have not gone through similarly violent changes. Equity volatility in the euro area and Japan rose in tandem with the US but didn’t reach the same intraday highs. High yield spreads widened too, but not in a disorderly manner. Investment grade spreads didn't move much. Indeed a simple model of daily VIX changes vs daily S&P changes would have implied that US equities should have been down nearer 11% on 5 February rather than 4%. In other words, implied has move significantly higher than realised.

There are two interpretations of these facts. First, the VIX is sending us a serious message about the outlook for short-term US equity returns that the market needs to pay attention too. Second, US equity volatility products are an asset class in their own right. The existence of levered short-vol carry positions in those markets makes them more exposed to relatively small changes in other asset classes, e.g., rates vol.

To the extent interpretation one is correct, we should see a large increase in long rates hedges and selling pressure in credit and EM. To the extent the second interpretation is correct we should expect, when the affected positions are removed, a return to the status quo ex ante. Feedback from discussions with market participants suggests most people are putting faith in the second interpretation; this was an isolated incident in an important derivatives market.

The evidence supports that case. But there’s a problem. And it's the term premium.

As we wrote last time, the cyclical growth outlook points to higher policy rates and flatter curves, tight spreads and strong equity performance. The secular improvement scenario calls for higher rates too but is ambiguous about the shape of the nominal and real curve. The ambiguity stems from uncertainty about the natural rate and inflation. This makes it hard to anchor long-end rates (without even taking into account net net issuance) – ergo higher realised rates volatilities. This uncertainty is unlikely to go away for some time if we remain in an above-trend positive output gap world.

Thus normalisation – whether cyclical or structural – means a higher term premium and higher rates volatility. Even while the absolute level of yields remains low the past few days tells us two important things: investment strategies habituated to low term premia are likely to struggle in this environment; and higher beta assets can rally as government yields rise so long as their risk premia fall faster than rates increase. If rates rise too quickly, all bets are off.

The speed and breadth of the recent recovery has left central banks with a communication problem – it is tough to forward guide a nervous bond market if inflation is rising and growth is well above trend. If a rising term premia has claimed its first victim, will there be others?

Given the low absolute level of yields there’s little case for a growth driven risk sell off. But it is the speed of the adjustment to normal that will now be critical. Perversely, the best thing for markets now would be more modest growth and comforting downside inflation miss. - source Nomura

There lie the crux of the situation, too much good news could lead to more bad news for risky assets such as the dreaded CPI number coming out soon in the US. This is a very important number for bond yields in particular and asset prices in general. 

As we stated in our previous conversation prior to the VIX bloodbath that ensued, positive correlations matter and matter a lot. On this subject we read with interest Deutsche Bank' Special Report from the 7th of February entitled "The bond risk premium and the equity/yield correlation":

"A few months ago, we noted that the combination of low yields and high equities raised concerns that a sudden rise in bond yields could lead to a material repricing of equities. A week ago, we had a glimpse of a potential shift in the equity/yield correlation (higher yields/lower equities) that has been mainly positive since the late 90s. The risk-off environment of the last couple of days has reasserted the positive correlation between bond yields and equities. How can we explain such shift in correlation?

Our analysis suggests that the equity/yield correlation is related to the bond risk premium. A high bond risk premium coincides with a negative correlation between yields and equities. In the post Volker period, 10Y UST ~3% above r* corresponds to a yield/equity correlation close to zero (on a 12m backward looking basis). Current estimates of r* are between 0 and 50bp, but are expected to rise to 50-75bp in the quarters ahead.

On this basis, assuming that inflation expectations remain broadly anchored, a persistent shift in correlation is likely to occur when 10Y UST is somewhere between 3 and 3.5%. Clearly, the market will notice a shift in correlation on a much shorter time frame: a few days of negative correlation between yields and equities have been enough to generate significant attention. Thus, one would expect the shift in correlation to be felt at lower level of rates.

The relative perception of the risk of high vs low inflation regimes – i.e. 70s stagflation vs. Japanese deflation – is likely to be the underlying driver of the equity/yield correlation. When higher inflation is negative for growth (e.g. the 70s), one would expect bond yields and equities to be negatively correlated. A positive shock to inflation would coincide with a negative shock to growth, leading to higher bond yields and lower equities. Moreover, as higher inflation will coincide with lower growth, bonds will not be a good hedge for an equity portfolio. As a result, they should command a higher risk premium. We would therefore observe a higher risk premium in bond markets and a negative correlation between bond yields and equities.

Conversely, if low inflation is negative for growth (e.g. Japan), then one would expect bond yields and equities to be positively correlated. A negative shock to inflation would coincide with a negative shock to growth. It would lead to lower bond yields and lower equities. Moreover, as lower inflation will coincide with lower growth, bonds will be a good hedge for an equity portfolio, commanding a lower risk premium. We would therefore observe a lower risk premium in bond markets and a positive correlation between bond yields and equities.

From a historical perspective, higher productivity coupled with cheap supply of labour reduced the risk of high inflation since the late 1990s. Technological advances and globalization have therefore likely been instrumental in establishing the low bond risk premium regime in place since the late 1990s.

This enabled central banks to establish their inflation targeting credentials and to be more predictable, thereby reducing interest rate volatility and the bond risk premium. Moreover, the FX regime in place in key EM economies led to  a significant increase in excess savings, which in turn depressed the bond risk premium (the bond market conundrum).

Looking ahead, the focus on inequalities in DM economies coupled with the desire of China in particular to shift towards a more consumption based economy suggest the current political economy trend is conducive to some reversal of the regime in place since the 90s, pointing towards the potential for a higher bond risk premium.

Indeed, these trends should result in (a) greater risk of trade barriers, (b) greater fiscal deficits in DM and (c) reduced current account surpluses in EM. Taken together, this should reduce the savings/investment imbalance and the disinflationary pressures from globalization. The resulting rise in the bond risk premium would put downward pressure on the yield/equity correlation.

The bond risk premium and the equities/yields correlation: the evidence

The bond risk premium (defined as the 10Y yield minus long-term growth and inflation expectations) has proven to be a good indicator of the correlation between equities and yields (defined as the rolling 12m correlation of weekly changes in the S&P and weekly changes in the 10Y yield). In the early 1990s, the bond risk premium was high and the correlation between yields and equities was negative.

Since the late 1990s, the correlation has been mostly positive with temporary exceptions around the end of tightening cycles (early 2000s and 2006) and the taper tantrum (see graphs below, note that the correlation scale is inverted on the left graph).

The focus on the post 1990s period is driven by the data availability: there are no long term growth and inflation surveys prior to this date. This period has been one of relatively stable inflation (especially relative to the 1970s), and as a result a significant portion of the move in yields can be associated to changes in the bond risk premium or a decline in the neutral real rate rather than changes in inflation. Thus, over this period, using the gap between UST10Y and the neutral real rate (as estimated by Holston, Laubach and Williams) is also a good indicator of the correlation between bond yields and equities (see graphs).

By focusing on 10Y UST – r* we can extend the sample to the early 60s. The correlation observed since the 1990s, extends over the whole lower inflation era (1986-2017). During the post Bretton Woods/high inflation/Volker periods, the correlation is weaker which could be ascribe to the fact that the 10Y rates are more impacted by inflation. Also the yield/equity correlation is always negative, which is consistent with the intuition discussed above.

In the 60s, the correlation is somewhat stronger again (see graph below).

In short, the bond risk premium is a decent indicator of the correlation between equities and bond yields. When it is high, the correlation becomes negative, pointing to high bond risk premium weighing on equities.

Assuming that inflation does not vary significantly, the gap between 10Y UST and r* can be used as a reference. Looking at the most recent samples (charts below), a spread between 10Y UST and r* of 3% would be consistent with the bond equity correlation persistently changing sign.

As well as representing a gauge of bond risk premium, the equity/yields correlation can be also used to confirm the evolution of credit risk in the euro area. Indeed, the correlation between the various European bond yields and equity markets was the same pre-crisis. In 2009, Italy switched to a persistently negative correlation which increased back towards zero following the “whatever it takes” statement from Draghi. The correlation between the CAC40 and OAT briefly turned negative when French spreads where widening substantially in 2011/2012. The shift in correlation can be interpreted as the result of a tightening of credit conditions due to a widening of credit spreads." - source Deutsche Bank

If indeed the shift in correlation from positive to negative marks a regime change in the narrative, given how loose financial conditions have been, it also indicates as per Deutsche Bank a tightening of credit conditions. To illustrate further the impact changes in cross-asset volatility thanks to change in correlations we think the below chart from Bank of America Merrill Lynch Cross-Asset Hedging note from the 7th of February entitled "Few signs of X-asset contagion as equity vol bubble finally pops" illustrate even further the "harmonic tremor":

- source Bank of America Merrill Lynch

For us, the most important piece of the puzzle for additional pain would be from the "Big Bad Wolf" aka inflation. This would generate additional pressure on risk premiums and bond yields. What matters as Nomura puts it in their note, is the speed of the adjustment and also from the Fed should it fells it is behind the curve thanks to faster than expected rise in inflation expectations. The risk obviously is on the upside particularly when it comes to rising concerns with inflation expectations. Some see the current situation with the latest US fiscal profligacy as similar to what the US experienced in the 1960s. This is the case for Deutsche Bank which sees similarities as per their Global Fixed Income note from the 9th of February entitled "A structural repricing of bond markets":

"There are some striking similarities between the new policy mix discussed above and the conditions that led to a shift upward in inflation expectations in 1966. In the first half of the 1960s, unemployment was declining rapidly but core inflation remained low and the Phillips Curve was “dead” (right graph below).

In 1966, the US administration significantly increased its fiscal spending on the back of (a) the Vietnam war and (b) the introduction of Medicare and Medicaid. This denoted a turning point in core inflation, which began to rise markedly (graph above). There are competing interpretations as to what drove the pickup in inflation: (a) fiscal stimulus, (b) non-linearities in the Phillips curve as the unemployment rate dipped below 4%, (c) rising healthcare inflation and (d) a Fed that was (ex-post) behind the curve.

Irrespective of the precise drivers, there are some similarities with current conditions. As the US economy is approaching full employment, the US government is implementing a significant fiscal stimulus. The potential introduction of trade barriers creates upside risks to inflation. Finally, the Fed may not intend to be behind the curve today, but if r* rises, as the Fed and our economists expect, current market pricing of the Fed policy will be overly accommodative." - source Deutsche Bank

There are many upside risks we commented in recent conversation, one being the start of a trade war through the implementation of trade barriers (that 30s feeling) which would put indeed some pressure on prices no doubt. This set up of both US profligacy and trade war would obviously reinforce further the bullish case for gold that led, at the end of the Vietnam to the Nixon shock in August 1971 and  the direct international convertibility of the United States dollar to gold. Could the sudden rise in positive correlations be yet another sign of the buildup in "harmonic tremor" that would led to a regime shift in inflation? We wonder.

Given as we pointed out again recently in our conversation "Bracket creep" that bear markets for US equities generally coincide with a significant tick up in core inflation, this the biggest near term concern of markets right now we think. When it comes to the relation between inflation and stock markets we read with interest Nomura's take from their Inflation Insights note from the 6th of February:

"A lesson for the near future

We think the correction in stock prices is connected to the likelihood of a return of wage inflation. We look at the traditional Gordon growth model to find that a scenario of higher wage inflation, higher inflation, higher inflation valuations, higher nominal rates and higher equity prices are very unlikely unless expectations of trend real growth substantially increase from current levels. We note that the US corporate tax cut did not manage to lift these expectations. We also note that the correction in stock prices may be a signal that US monetary policy is maybe being tightened in real terms more rapidly than what profit growth can actually sustain.

A simple framework and a reminder

According to James Bullard, President of the Federal Reserve Bank of Saint Louis, “inflation scare triggered some of the (equity) market sell-off”. There is indeed a very strong theoretical linkage between inflation and equity markets, which is best illustrated in the simple framework provided by the Gordon growth model, also referred to as the Dividend Discount Model. In this very simple model of stock prices, stock prices are discounted dividend expectations so that:

Where P is the stock price, D is future dividends and y is the nominal discount factor. Assuming a constant rate of growth for dividend g, this equation can be re-written so
that:

Yet beyond this simple formula there are various assumptions that connect stock prices to the inflation market. First, y is the discount factor for stock prices, so it is the nominal risk-free rate plus the equity premium (y=r+ep). G is the growth rate for dividends, so it is in fact the nominal growth rate of the economy multiplied by the share of economic growth that goes to profits rather than wages (g=k*gdp, with gdp the growth rate of nominal GDP). Economists call it the sharing of national income.

A key factor affecting the sharing of national income is wage inflation. Average hourly earnings increased to 2.9% year-on-year in the US nonfarm payrolls report for January – their highest growth rate since 2009. This high number was perceived as heralding the return of wage inflation in the US and therefore altering the sharing of national income towards a larger share for wages and a lower share for profits.

Clearly, this framework from 1962 does not capture some other key financial aspects of the determination of stock prices. For example, it is likely that there is some linkage between the sharing of national income and the level of the risk-free rate (k and r). If wage inflation returns, the path of nominal interest rates is likely to be higher: that was also the case recently, resulting in a higher discount factor for dividends negatively affecting stock prices. With the equity premium term already minimal (implied volatility is low), there were few reasons for investors to expect higher risk-free rates to be offset by a lower equity premium.

We view this framework as a good reminder that a scenario of higher wage inflation, higher inflation, higher inflation valuations, higher nominal rates and higher equity prices are very unlikely unless expectations of trend real growth substantially increase from current levels. It is also intriguing that the US corporate tax cut was so promptly followed by the stock price correction – which suggests at least some skepticism about its effectiveness.

A minor caveat

And there is of course a minor caveat to this framework: the risk of an inflation overshoot remains noticeably absent from inflation valuations despite the increase in wages in January. Figure 1 shows that despite an increase in the 5yr inflation swap rate, the price of a hedge against inflation much higher than 2% is not historically high. We are only cautiously long 5yr breakeven rates in the US, it is worth recalling.

Too high, too fast?

Maybe the correction in stock prices – despite the tax cut – is a sign that real rates are getting close to “neutral” levels, for example the level of the natural real rate estimated by Williams and all, is likely to exert a negative effect on growth that may go beyond what inflation conditions currently imply. In other words, absent the risk of an inflation overshoot, real rates may normalise too fast, too soon – which would have an adverse impact on the real risk-free rate, but also on real dividend growth.

- source Nomura

When it comes to inflation "expectations", the above reminds us it is all about "implied" and "realized". In similar fashion implied volatility is more simply what the market is expecting (VIX) whereas realized volatility is sometimes deemed more tangible because it reflects the actual daily movement of the S&P 500 Index. The true outlying year in history was 2008 (before 2018's sucker punch), when realized volatility was actually higher than implied volatility - the only such instance over recent years. Even in 2017 the relationship between implied volatility and realized volatility was pretty much "normal". The issue of course is that when central banks are meddling with interest rates and financial repression leads to volatility being subdued for too long, then like a coiled spring, profiting, primarily, from the "volatility premium" (the difference between implied volatility (investors’ forecast of market volatility reflected in options pricing) and realized (actual) market volatility) seems like a "sure bet" for the likes of LJM Fund Management that got beaten up big time with the VIX blowing out à la 2008 (that infamous rogue waves we talk about with a breaking force of 100 MT/m2) . Believing that the spread between implied and realized volatility would persist has indeed been a dangerous proposal with rising positive correlations. In similar fashion believing that "implied inflation" could persist remaining below "realized inflation" could become hazardous in the coming months, particularly with growing geopolitical exogenous risks around. Whereas QE was deflationary, QT could prove to be inflationary but we ramble again...

Make no mistake, inflation is the "Boogeyman" for financial markets.  A sharp pickup in inflation is likely to entail a significant re-pricing and as per our final chart below it represents a serious headwind for the US consumers (Bracket creep aside).

• Final chart - Increased inflation presents a potential threat to the purchasing power of consumers.

 The "Big Bad Wolf" aka inflation would force the hand of central banks and lead to a more rapid pace in rate hikes leading to some significant additional repricing on the way. Inflation is our concern numero uno. Our final chart comes from Wells Fargo Economics Group note from the 8th of February entitled "Is the US Consumer Running on Fumes?" and highlights the relationship between PCE Deflator and Disposable Income:

"The Threat of Higher Inflation and Higher Interest Rates

In general, higher inflation reduces the growth rate of real disposable personal income and vice versa, which is clearly demonstrated in Figure 8.

As income and wealth are affected by fluctuations in inflation, one of the biggest threats over the next several years has to do with the rate of inflation. Markets recently seem to have been spooked by the relatively, and surprisingly, strong report on average hourly earnings, which could be indicating some pressure on prices for the U.S. economy. Higher inflation means higher interest rates, and both factors are clearly negative for the U.S. consumer. Higher inflation reduces the purchasing power of income, while higher interest rates makes purchases of durable goods, which are typically financed, more expensive over time. While for those that have fixed-rate mortgages, it is music to their ears, it is bad news for those that have adjustable rate mortgages.

Although inflation has remained low in this cycle compared to its historical trend, if prices were to accelerate, Americans’ real DPI growth will, once again, slow down and could also lead to a slowing of growth in real PCE, all else equal. Therefore, if we were to see an uptick in inflation, real DPI growth will slow and consumers’ purchasing power, or the amount they could consume based on their current income, would be negatively affected. Although this effect is clearly demonstrated in Figure 8, it is also evident that DPI experiences fluctuations with rather lackluster inflation growth. That is, although higher inflation can directly decrease disposable income growth, it is not the only factor that causes reductions in the rate of growth of income.

Furthermore, increases in interest rates could contribute to a slowdown in real PCE, as consumer purchases might diminish based on increased expense, such as what we have previously mentioned associated with durable goods financing. Another sector of risk for the consumer as well as for the credit market is the tax reform’s change in second mortgages or equity lines of credit. Americans, in some circumstances, can no longer take a credit on their taxes for interest on equity lines of credit and this together with the still-high, relative to the past, delinquency rate for these lines of credit could be signaling problems ahead for the U.S. consumer, as well as for the overall credit market." - source Wells Fargo

One could argue that the only "easing" day was yesterday. Have we seen "peak consumer confidence" in the US? It certainly looks like it so beware of the Big Bad Wolf aka "inflation" because he has already blown apart the "short vol" pig's house made of straw...

"Worry is the interest paid by those who borrow trouble." -  George Washington

Stay tuned ! 

Macro and Credit - The spun-glass theory of the mind

"Success consists of going from failure to failure without loss of enthusiasm." - Winston Churchill

Looking at the results stemming from the Italian referendum in conjunction with the continued gyrations in financial markets on the back of rising FX volatility thanks to "Mack the Knife" aka King Dollar + positive real US interest rates, when it came to selecting our title analogy for this week's musing, we reminded ourselves of "The spun-glass theory of the mind" which is the belief that the human organism is so fragile that minor negative events, such as criticism, rejection, or failure, are bound to cause major trauma to the system (think Brexit, Trump's election). "The spun-glass theory of the mind" is essentially not giving humans, and sometimes patients, enough credit for their resilience and ability to recover, like central banks have been doing, dealing with economic woes with their "overmedication" programs (ZIRP, QE, NIRP, etc). In 1973, the brilliant University of Minnesota clinical psychologist Paul Meehl poked fun at what he called the “spun glass theory” of the mind which is the false notion that most of us are delicate, fragile, and easily damaged creatures that need to be handled with kid gloves. Since then, many researchers have shown that most people are surprisingly resilient even in the face of extreme trauma. Economies are similar, such as the United Kingdom which showed it was more than tremendously resilient while many pundits were predicting "trauma" and disaster should Brexit happens. In similar fashion, Nassim Nicholas Taleb in his book "Antifragile" showed that there's an entire class of other things that do not simply resist stress but actually grow, strengthen, or otherwise gain from unforeseen and otherwise unwelcome stimuli (Iceland). The main underlying concepts of both "The spun-glass theory of the mind" and "Antifragile" is that the majority of causal relationships are nonlinear and so are market movements (hence the relative ineffectiveness of VaR models - Value At Risk we discussed in February in our conversation "The disappearance of MS München"). Typically both have a convex section where the curve rises exponentially upward and is associated with a positive effect (antifragile) and a concave section that declines exponentially downward and has a negative effect (fragile). Think of the dose of a prescription drug. At first, as central banks increase the dose, the health benefits improve (convexity) for financial assets. But, beyond a certain dose side effects and toxicity cause harm (concavity), such as Debt-fueled economies given debt has no flexibility. Therefore highly leveraged economies cannot stand even a slowdown without risking implosion like our current situation, but we ramble again. 

How do you "hedge" in such a nonlinear world? The way to do it, we think, is to use a barbell strategy that positively captures the optionality of the variable (being long in the convex area and short in the concave area).  If indeed, we live in a nonlinear world and given correlations are less and less static and change more and more frequently, leading to larger and larger standard deviation moves such as typically going way up during downturns, it therefore eliminates Markowitz portfolio theory of diversification benefit. Just when you think your diversification will render your portfolio "antifragile" it brings instability and "fragility" to it. A barbell strategy should render your portfolio more "antifragile". Why is so? Markowitz portfolio theory causes investors to "over allocate" to risky asset classes such as "High Yield" and/or Emerging Markets and play the same crowded "beta" game. In similar fashion, "The spun-glass theory of the mind" cause central bankers to "overmedicate". One could conclude that "Overmedication" leads to "Over allocation". 

In this week's conversation we would like to look again at the importance of flows versus stocks from a macro and credit perspective, taking into account "overmedication" and "over allocation".

Synopsis:

• Macro and Credit -  It's a question of flows versus stocks
• Final chart - Could Japanese equities be antifragile in 2017?

• Macro and Credit -  It's a question of flows versus stocks

Our core thought process relating to credit and economic growth is solely based around a very important concept namely the accounting principles of "stocks" versus "flows". We have used this core principle in the past when assessing the issues plaguing Europe versus the United States as per our September 2012 conversation "Zemblanity":

"We mentioned the problem of stocks and flows and the difference between the ECB and the Fed in our conversation "The European issue of circularity", given that while the Fed has been financing "stocks" (mortgages), while the ECB is financing "flows" (deficits). We do not know when European deficits will end, until a clear reduction of the deficits is seen, therefore the ECB liabilities will have to depreciate."

Before we delve more into the nitty-gritty, it is important, we think to remind our readers of what is behind our thought process of the "stocks" versus "flows" macro approach.

We encountered previously through our readings an essential post dealing with our core concept of "stocks versus "flows" from Mr Michael Biggs and Mr Thomas Mayer on voxeu.org entitled - How central banks contributed to the financial crisis which explains precisely why both Friedman, Keynes and the central banks have been behind the curve in preventing the previous financial crisis and potentially the next one: 

"We have argued at some length in the past that because credit growth is a stock variable and domestic demand is a flow variable, the conventional approach of comparing credit growth with demand growth is flawed (see for example Biggs et al. 2010a, 2010b).To see this, assume that all spending is credit financed. Then total spending in a year would be equal to total new borrowing. Debt in any year changes by the amount of new borrowing, which means that spending is equal to the change in debt. And if spending is equal to the change in debt, then the change in spending is equal to the change in the change in debt (i.e. the second derivative of the development of debt). Spending growth, in other words, should be related not to credit growth, but rather the change in credit growth. 

We have called the change in debt (or the change in credit growth) the 'credit impulse'. The credit impulse is effectively the private sector equivalent of the fiscal impulse, and the analogy might make the reasoning clearer. The measure of fiscal policy used to estimate the impact on spending growth is not new borrowing (the budget deficit), but rather the change in new borrowing (the fiscal impulse). We argue that this is equally true for private sector credit." - Mr Michael Biggs and Mr Thomas Mayer on voxeu.org

We have always wondered in relation to the global rounds of quantitative easings the following:

"Does the end (lowering unemployment levels) justify the means (increasing M) or do the means justify the end (deflationary bust)?"

Credit dynamic is based on Growth. No growth or weak growth can lead to defaults and asset deflation. The change in credit growth is a flow variable and so is domestic and global demand!

The big failure of QE on the real economy at least in Europe has been in "impulsing" spending growth via the second derivative of the development of debt, namely the change in credit growth.

As we have argued before QE in Europe is not sufficient enough on its own to offset the lack of Aggregate Demand (AD) we think.

As a reminder, in our part 2 conversation "Availability heuristic" from September 2015, the liabilities structure of industrial countries is mainly made up of debt (they are “short debt”), in particular in Japan, the US and the UK. In contrast, the international balance sheet structure of emerging markets is typically composed of equity liabilities (“short equity”), which is the counterpart of strong FDI inflows that contributed to improve emerging markets’ external profile in the last decade. With a rising US dollar, what has been playing out is a reverse of these imbalances hence our "macro reverse osmosis" discussed again recently relating to violent rotations in flows. 

From that perspective, we read with interest Citi Research note from November entitled "How does active fund management survive in 2017?" where as well they tackle the very important point of stock versus flows:

"Is it the stock or the flow of QE that matters?

Essentially, central banks tend to think in "stock" terms

 “Reduce the quantity available to investors & prices will lift permanently”

 To us, QE flows seem very important empirically

Reduce the QE flow by just ~1/3 & markets are quite likely to fall

That makes an asset not priced to fundamentals but to policy …

… prone to non-linear reactions when perceptions of policy change" - source CITI

As we have seen in recent weeks, and as we have remarked in our conversation  "Critical threshold", higher yields leads to material fund outflows in the short-term as indicated by Bank of America Merrill Lynch Follow the Flow note from the 2nd of December entitled "Where's the money going?":

"High grade funds had their fourth week of outflows, and the third that exceeds the $1bn mark. High yield funds recorded their fifth week of outflows in a row; so far this year they have lost more than $10bn. As chart 13 below shows, the outflows this time came mainly from European and global funds, where on the other hand US high yield funds recorded an inflow.

Government bond funds had yet another outflow, the eighth in a row, reflecting rising QE-tapering risks. Money market weekly fund flows were relatively subdued recording a negative flow. Overall, fixed income funds flows remain largely negative, and over the past four weeks almost $20bn has flown out of European domiciled funds.

European equity funds flows switched aggressively to the negative side again, post a short stint of inflows. Last week the asset class had its biggest outflow in eleven weeks. Outflows so far this year are just shy of $100bn.

Global EM debt fund flows remained negative for the fourth week in a row; nonetheless we note a significant improvement in the trend, with the latest outflow being significantly smaller than that of the previous two weeks. Commodities funds also were in negative territory for a third week, as higher inflation expectations support reflation trades.
On the duration front, short-term IG funds flows remained negative for a second week. Mid-term funds had their fourth outflow in a row but riding an improving trend; while long-term funds recorded a marginal inflow." source Bank of America Merrill Lynch

Whereas central banks tend to focus their attention on "stocks", we'd rather focus ours on "flows", from a macro perspective as well as from a fund perspective. Evidently the consequences of "Mack the Knife" aka King Dollar + positive real US interest rates can also be seen in the "Great Rotation" from Europe and Emerging Markets towards the US hence validating our "macro osmosis process":

"In a normal "macro" osmosis process, the investors naturally move from an area of low solvency concentration (High Default Perceived Potential), through capital flows, to an area of high solvency concentration (Low Default Perceived Potential). The movement of the investor is driven to reduce the pressure from negative interest rates on returns by pouring capital on high yielding assets courtesy of low rates volatility and putting on significant carry trades, generating osmotic pressure and "positive asset correlations" in the process. Applying an external pressure to reverse the natural flow of capital with US rates moving back into positive real interest rates territory, thus, is reverse "macro" osmosis we think. Positive US real rates therefore lead to a hypertonic surrounding in our "macro" reverse osmosis process, therefore preventing Emerging Markets in stemming capital outflows at the moment." - Macronomics, August 2013.

As a reminder, more liquidity = greater economic instability once QE ends for Emerging Markets. Now if indeed "flows" matter, we took a keen interest in reading the impact  "Mack the Knife" has had on Emerging Markets as indicated by Bank of America Merrill Lynch in their GEMs Flow Talk note from the 1st of December 2016 entitled "Reported foreign holdings of local debt dropped 6% in November so far":

"EPFR weekly fund flows
• EPFR measures mutual funds and ETFs (AUM about $225bn)
• This week’s outflows were less than last week which was less than the prior week.
• EPFR outflows in the 3 weeks since the election were:
-4.0% for EXD
-3.7% for LDM
• Part of that is ETFs, whose outflows since the election as a % of ETF AUM were:
-8.1% for EXD
-7.3% for LDM

2013 lessons were learned well, take fast action
• Outflows have been faster than in 2013 (Chart 1)

• Largest 3 week outflow in 2013 was only -4.2% for all currency funds.

Foreign holdings of local debt sharp drop – $40bn implied
• We track $640bn of foreign holdings of local debt
• We observed 6.2% outflows for the month of November in 4 countries (India,
Indonesia, Hungary and Turkey) with foreign holdings of $107bn.
• If all countries had the same average outflow of 6.2%, then applied to the $640bn
of foreign holdings, this implies we could have had a $40bn total outflow in Nov.
Since the election – $35bn potential outflow
• We observed 5.5% outflow since the election.
• If we apply this rate to the $640bn foreign holdings, it implies a $35bn total
outflow in the few weeks since the election." - source Bank of America Merrill Lynch

We recently pointed out that "macro tourists" and levered carry players alike did play the second half of 2016 more aggressively hence the extension of their credit risk and duration risk leading to faster deleveraging and consequently outflows and pain inflicted. Obviously we guess that your next question is going to be how much more additional outflows could be expected particularly from the impact of a rising US dollar is having on Asia for example given capital outflows matter and matter a lot, so does a US dollar shortage as well although Stanley Fischer from the Fed thinks as of late there is no liquidity issue. Regarding this matter we read yet another Bank of America Merrill Lynch note from their Connecting Asia series entitled "Flow and flight":

"We examine two of the most pressing issues facing Asia investors in the Post-Trump election victory world.
The first, is how much capital reversal and outflow in Asia is yet to run amid USD strength, rising yield differential with the US and CNY depreciation. Thus far, some USD24bn in foreign bond and equity portfolio flows are being unwound and compares with the maximum drawdown of USD55bn during the GFC – see front page chart.

The countries that we find are at the sharp end of this outflow are Korea, India, Indonesia and Malaysia.
The second is how investors should evaluate Asia FX and rates vulnerability to the tail risk of rising trade protectionism and confrontation. The brief answer to this is that China, Korea, Taiwan and Thailand appear most vulnerable, while Indonesia and India may be the least.
Ultimately, the combination of capital outflows and rising trade protectionism discussed in this report, suggests that FX risk premiums and volatility for CNY and KRW will rise. From an FX hedging strategy perspective, we continue to recommend low delta 6M USD call, CNH puts such as our year-ahead top trade recommendation for USD/CNH 7.60 strikes.

Portfolio drawdowns – how much more to go?We highlight the maximum drawdown Asian markets have previously seen in terms of outflows. This gives us a sense of the worst case scenario as far as outflows are concerned relative to FX reserves. The bottom line is:
• The largest outflow Asia saw on a cumulative basis was ~USD 55bn during 2008. In comparison to this, Asia has seen about USD 24bn of outflows since October 2016 (see Chart 1 for cumulative outflows during other risk off periods).
• Equity outflows so far have been around 9.9bn USD since October 2016, led by. India, Taiwan and Thailand. When compared to historical drawdowns, the larger equity markets of Korea, Taiwan and India stand to lose the most, although only in Korea’s case does the worst case scenario account for a substantial portion of FX reserves (19%, see Table 1).

• Bond outflows, so far have been about 14bn USD since October 2016, with most seeing outflows of at least 2bn USD. When compared to historical drawdowns, Malaysia stands to lose the most, with the worst case scenario representing about 6.3% of current FX reserves (see Table 2).

- source Bank of America Merrill Lynch

While obviously the biggest "known unknown" lies in the foreign trade policies which will be adopted by the new US administration. As we pointed out in our last musing, measures that would restrict global trade would no doubt be bullish for gold in that particular case. For now, in relation to gold we still remain neutral.

But moving back to credit and nonlinearity, one way of "mitigating" dwindling policy support given recent talks from the ECB in tapering its stance would be to "embrace" a barbell strategy as pointed out by Citi Research note from November entitled "How does active fund management survive in 2017?":

"Barbell when repression is at risk of being wound down

Belly of the credit curve holds disproportionate amount of unpaid β" - source CITI

What would be an interesting barbell credit wise in our opinion? We would favor US credit markets obviously from a flow and currency perspective. We would go long US investment grade credit than European investment grade and even selectively long European non-financial High Yield issuers due to lower leverage than their US peers. But should you want to play the beta game from a "barbell" perspective, then again US High Yield via its derivatives US CDX High Yield, given that it is less sensitive to convexity and interest rate risks and less exposed to CCCs (10% versus 16% weight in cash index), should the risk-on environment persist on the back of favorable macro data. 

From an allocation perspective, we are already seeing once again decoupling between US credit and Europe, because, as we stated on many occasions, the Fed tackled earlier one "stocks" issues on banks balance sheet, which in effect, enabled a stronger credit income and better economic growth relative to Europe, whereas the ECB has in no way alleviated the burden of "stocks" plaguing peripheral banks in the form of nonperforming loans, therefore in no way repairing the broken credit mechanism that stills explain the on-going "japanification" process and much weaker growth prospects. To that effect, if indeed the US reflation story is playing out, then again it makes sense to "over allocate" to US credit as once again decoupling could be on the menu between both regions. This is clearly illustrated by Bank of America Merrill Lynch in their European Credit Strategist note from the 2nd of December entitled "The Italian job":

"The last month has presented something rather rare: a truly decoupled global credit market. For instance, US high-grade spreads went 2bp tighter in November, while Euro high-grade spreads went 16bp wider. And the phenomenon seeped into high-yield credit too: US spreads tightened by 24bp in November, while European spreads widened by 47bp. After the moves of the last month, headline IG spreads in Europe are now wider than US high-grade spreads out to almost 10yrs in maturity.

“Politics” has been the undoing of European credit lately. Italy heads to the polls on Sunday amid a climate of rising global populism. And ECB tapering noises have driven a pattern of rising rates and wider spreads in Europe over the last month (note, though, BofAML base case is for an €80bn QE extension until Sep-17).
Yet, even with all the political hiccups that Europe has encountered over the last 5yrs, genuine decoupling of credit markets has not been common. Chart 2 shows that there have only been 3 periods over the last 10yrs when European and US credit spreads went in different directions.

Decoupling – the new norm for ‘17

We think decoupling between US and European credit will be a lot more common in 2017 though. In fact, our US credit strategy colleagues forecast US high-grade spreads to tighten by 20bp next year. In Europe, we expect high-grade spreads to widen by 20bp.

Much of the divergence in views is simply down to technicals – which shouldn’t come as a surprise given how technicals have been the be-all and end-all of credit markets for the last few years. In the US, we expect Republican tax proposals to lead to a big drop in US high-grade net supply next year. But in Europe, we expect shareholder-friendly activity (M&A, in particular) to broaden out in 2017, and contribute to a further jump in supply. This should leave the Euro market with too many bonds and not enough buyers.

Get in quick…

In fact, lingering QE tapering noises may just coax European companies to speed up their spending and issuance plans, for fear of missing the (low-yield) boat. This means the risk of supply being front-loaded in the first half of next year, leading to some heavy indigestion for the Euro credit market.
Recall that this is not too dissimilar to what US companies did in 2014 as the Fed drew closer to their first interest rate hike. As chart 3 shows, US M&A volumes were brisk in the middle of 2014. Then, as better US data pushed yields higher, issuers moved quickly to fund the M&A backlog, leading to some big months of US high-grade supply in September and November 2014.
This also caused a big decoupling in credit markets: US high-grade spreads widened by almost 40bp in the second half of 2014, while European high-grade spreads went slightly tighter. We fear the same bad technicals could be at work in Europe next year if companies rush to issue ahead of any potential QE tapering." - Bank of America Merrill Lynch

If indeed we get this "reflationary" case playing out, then again we might have a situation where US credit continues to outperform European credit in 2017.

Going forward, when it comes to following a potential deterioration in US High Yield we encourage you to keep an eye on a possible flattening of the CDX HY curve, being the derivatives proxy for the US High Yield market. As per Credit Market Analysis (CMA) latest chart, it is worth following the trend to see if indeed the CDX HY series 27 will be getting flatter as we move towards 2017. We noticed that one year protection has started to move upwards albeit very slowly, while it is yet a meaningful move for the moment contrary to what we had back in November last year where 1 year was only 50 bps apart from 3 year, you should keep an eye on the shape of the curve:

- source Credit Market Analysis

Right now, it is hard to be as sanguine as we were in November regarding US High Yield given at the time of the fast flattening movement we were seeing at the end of 2015. We continue to see a risk-on environment for the time being, although as we pointed out last week when discussing credit conditions in the US for Commercial Real Estate, it does appears to us that some segments including our CCC credit canary are already experiencing tightening financial conditions. The most important indicator to track, risk wise is "Mack the Knife" aka King Dollar + positive real US interest rates. 

Finally for our final chart and if the "spun-glass theory of the mind" is correct, then indeed we think if the US dollar continues to shine, then it makes sense to "over allocate" to Japan given earnings are higher there.

• Final chart - Could Japanese equities be antifragile in 2017?

While the risk-on mode is still prevailing thanks to the strong beliefs in the reflation story playing out, from an equity perspective, corporate earnings and payouts remain the principal drivers of equity returns. To that extent, our final chart comes from Barclays Global equity and cross-asset strategy note from the 28th of November entitled "Reassessing the rotations" and displays earnings in Japan relative to the US and Europe:

- source Barclays

While it remains to be seen how long the "reflationary story" continues to play out, for now it is indeed "risk-on", but no doubt there are many political events lining up in 2017 that could put a spanner in the works. One thing is clear to us though, is that when it comes to markets commentators and some members of the sell-side, 2016 has proven with both Brexit and the US election that the spun-glass theory of the mind is alive and well...

"Success breeds complacency. Complacency breeds failure. Only the paranoid survive." -  Andy Grove, former CEO of Intel.

Stay tuned!