Sunday 17 December 2017

Macro and Credit - Rician fading

"The very concept of objective truth is fading out of the world. Lies will pass into history." -  George Orwell

Watching at the dizzying levels reached by the BitCoin (BTC) mania in true "Orchidelirium" fashion in conjunction with the latest FOMC decision being the final for Janet Yellen with disappointing wage growth on the background still confounding the Phillips curve cult members, when it came to picking up our post title analogy for our final long post of the year we reminded ourselves of "Rician fading" given the weakening central banking support narrative. Rician fading or Ricean fading is a stochastic model for radio propagation anomaly in case you asked and is caused by partial cancellation of a radio signal by itself (Fed's support to financial markets) - the signal arrives at the receiver by several different paths (hence exhibiting multipath interference: ZIRP, QE, NIRP and more), and at least one of the paths is changing (lengthening or shortening). Rician fading occurs when one of the paths, typically a line of sight signal, is much stronger than the others (Fed's tapering then balance sheet reduction). In Rician fading, the amplitude gain is characterized by a Rician distribution. In wireless communications, fading is variation or the attenuation of a signal with various variables. In similar fashion, with forward guidance being the preferred tool of the Fed, the strike level of the put provided by the Fed in recent years is now falling hence our fading analogy with the attenuation of the signal coming from the Fed's balance sheet reduction. 

In this week's conversation, we would like to look at what it entails to navigate in a much "flatter" world courtesy of the Fed, and if indeed yield curves predict recessions in advanced economies or whether it's the "wealth effect".

Synopsis:
  • Macro and Credit - Navigating in a flat world
  • Final charts -  Just "bid'em up".

  • Macro and Credit - Navigating in a flat world
Flat yield curves are most of the time associated with a maturing business cycle, making many investors wondering whether a recession is "imminent" or not. Whereas we pointed out in many of our recent musings that credit would hit the level 11 on the credit amplifier in true spinal tap fashion, many as of late have been pointing out towards the recent dislocations in cross currency basis and the risk of heightened dollar funding crisis and a looming liquidity scarcity. 

For some pundits, higher funding costs would make dollar fixed income assets less palatable for the foreign investing crowd. We pointed out in the past that a large support from US credit markets was "Made in Japan" in our conversation "The Butterfly effect":
"For Japanese investors increasing purchases in foreign credit markets has been an option. Like in 2004-2006 Fed rate hiking cycle, Japanese investors had the option of either increasing exposure to lower rated credit instruments outside Japan or taking on currency risk. During that last cycle they lowered the ratio of currency hedged investments. " - source Macronomics, July 2017
As we pointed out as well recently in our conversation "Stress concentration", when it comes to US Investment Grade and the appetite from foreign investors, it has been a TINA trade (There Is No Alternative):
"As long as the volatility in rates remains subdued, it is still "goldilocks" for credit markets and the fun continues to run "uphill", to the bond market that is. For now our central bankers have managed to tame volatility, and not only in rates." - source Macronomics, November 2017
As we indicated in this conversation also is that the low volatility regime has not only been a VIX or a MOVE index story. It has also been the case in various asset classes. When it comes to credit, everyone is still dancing and playing the "carry trade". Though with the Fed's "Rician fading", some are wondering with growing US funding pressure if indeed some "carry tourists" might decide whether or not to stick around with TINA or simply head home, which would surely no doubt lead to some "repricing" and credit spreads widening. On that matter we read with interest HSBC's take from their "Dollar Drought" note from the 8th of December entitled "Looming liquidity scarcity" in which they argue that dollar fixed income allocations could be impacted on a currency hedged basis for foreign investors:
"Cross border fixed income flows
Assuming our views of heightened dollar borrowing costs were to materialise in the next few quarters, we explore in the following section what would be the medium-term investment implications for global investors. To do this, we look at cross-border FX and duration-hedged dollar assets versus the local equivalent. We find that in most cases, the dollar funding costs are beginning to eat away at the spread pickup that non-US investors have enjoyed while buying dollar-denominated securities. In some cases, a pickup still exists such as USD IG credit FX hedged to JPY while in other scenarios EUR long-end core bonds look be more attractive for US based investors.
Eurozone investors
Using FX hedges alone, USD IG credit still has a small pickup versus EUR IG (figure 9).

However, the pickup is much smaller now than in 2014-15 when euro fixed income outflows were at their largest. In the rates space, the simultaneous flattening and steepening of the long-end of the US and euro-area government bond curves, respectively, means that long-dated euro core bonds provide better currency hedged yields for US investors than local USD government securities. With 6mo Eurodollar deposit rates at 1.65% and 6mo EUR Interbank rates at -0.32%, eurobased real money investors with USD assets are already paying 197bp on their currency hedge (selling USD forward at 1.65% and buying EUR forward at -0.32% to make 197bp). This assumes no basis swap cost. Add that in (3mo basis swaps are currently -43bp) and hedging costs are 240bp. With USD IG13 yielding just 3.20%, there is only 0.80% of yield left.
On an FX and duration-hedged basis, there is not much in it: USD IG spreads trade on top of EUR IG (figure 10), as they have now for a couple of years.

For institutional investors that need to hedge FX – insurers for example, given the regulatory cap – then flows into USD assets should diminish, particularly as Fed rate hikes progress.
Japanese investors
For JPY-based investors, USD IG continues to offer a modest pickup versus EUR when FX hedged back into yen (figure 11).

However, the scale of pickup is smaller than back in 2014-2015, given the four Fed hikes since December 2015. Similarly for JPY-based investors, there is very little pickup when duration is also hedged out (figure 12).

Japanese life insurers had been less active in the UST market this year, despite the relatively tight level of cross-currency basis swap spreads versus previous years. But given that the long end of the UST curve has flattened meaningfully, there is not much spread left for Japanese investors in the US market. On the contrary, JGB ASW buying remains a key theme for the medium term
UK investors
For GBP investors, the USD IG market has become much more expensive to hedge (it costs around 90bp), making USD credit FX hedged back to GBP unattractive versus 2016 levels. GBP IG was helped by investors returning home from USD credit in 2017. We could see more of the same in 2018.
Foreign ownership of US credit 
Overall, we know that foreign ownership of US credit has reached record levels. In the short run, this has generated positive momentum in USD IG credit amid the global search for yield (figure 13).
However, over the longer term, it leaves a large investor overhang, with USD2.5trn in USD credit from US issuers owned by foreigners. Once Yankees are included (using BIS data), foreign holdings (excluding ABS) rise to USD5.7trn (figure 14), out of the estimated USD13.2trn of USD-denominated corporate bonds.
We know from events such as LTCM in 1998 that capital flows back home can accelerate when volatility rises. In the low volatility regime in 2017, this has not been a problem. However, should volatility rise, cross-border flows may move from the Jekyll of positive momentum, to the Hyde of capital flight.
Looking beyond fixed income markets, we note the US net international investment position (NIIP) has reached unprecedented levels of -USD8.3tn, compared with just -USD4.0tn before the financial crisis (figure 15). 
On the one hand, it is fine for the world’s largest reserve currency to attract foreign inflows from multiple investor types (foreign exchange reserve investors are natural buyers of US Treasuries; multinational pension funds are natural buyers of equity and debt capital market issuers; etc). But what has occurred this cycle in particular has been a flight of fixed income capital from central bank-repressed markets, particularly in JPY, EUR, and CHF, to high-yielding DM markets such as USD fixed income. Given this, movements in front end US rates and cross-currency basis swaps will be closely watched in 2018.
The issuer perspective
For issuers, US corporate supply in front end EUR (reverse Yankees – US names issuing in euros) swapped back to USD remains attractive, in our view. At the longer end, however, the maths no longer work. (What is attractive for US investors is unattractive for US issuers.) This suggests we may see more industrial and auto reverse Yankee issuance rather than utility and telco, given different natural funding maturities.
The funding gap between EUR and USD, from a US issuer’s perspective, is also not as large now as it was in H1 2015 when reverse Yankee issuance was at its peak. Indeed, reverse Yankee new issues as a percentage of all EUR IG issuance ticked down from 25% to 23% between 2015 and 2017" - Source HSBC
While there might be very little pickup when duration is also hedged for Japanese investors during the 2004-2006 Fed rate hiking cycle, Japanese investors reduced their hedging and went for more credit risks. Will it be different this time around? With so much negative yielding bonds around, we find it is difficult  already to turn negative on US credit in true TINA fashion. The question you need to ask yourself when it comes to how 2018 will present itself from a foreign allocation perspective towards US fixed income and in particular for Mrs Watanabe (Mr Watanabe being too busy with BitCoin, but we ramble again...), the GPIF (Government Pension Investment Fund) and their Japanese Lifers friend is whether we are in a bull flattening case or in a bear flattening case: 

  • In a Bull Flattener case, the shape of the yield curve flattens as a result of long term interest rates falling faster than short term interest rates.  This can happen when there is a flight to safety trade and/or a lowering of inflation expectations.  It is called a bull flattener because this change in the yield curve often precedes the Fed lowering short term interest rates, which is bullish for both the economy and the stock market.
  • In a Bear Flattener case short term interest rates are rising faster than long term interest rates.  It is called a bear flattener because this change in the yield curve often precedes the Fed raising short term interest rates, which is  generally seen as bearish for both the economy and the stock market.
But before we look at the predictive recessionary abilities of the yield curve, we would like to look at how our Japanese friends could react to flattening in 2018, given as we pointed out in many conversations, what these guys do matter for flows in financial markets and it matters a lot!

On that subject we read with interest Nomura's Rates note from the 14th of December entitled "How will Lifers address flattening in 2018":
"Lifers’ sensitivity to yield levels: Foreign assets
Even if we narrow the focus to foreign assets, we find a clear preference among lifers for investments offering higher yields. In FY10-FY11 and FY14-FY16, lifers favoured hedged USTs (Figure 7) as they had the highest yields (Figure 6).

In FY12-FY13, OAT yields adjusted to levels comparable with UST yields, and lifers responded by increasing EUR-denominated assets at a similar pace to USD denominated assets. They also increased their EUR-denominated assets in H1 2017.
Lifers’ sensitivity to yield levels: Bear flattening/bull flattening
The flattening in the UST market has received much attention, but the implications differ depending on whether the curve bear flattens or bull flattens (link). In the case of bear flattening, lifers tend to gravitate towards foreign bond investment, but bull flattening encourages lifers to move away from foreign bonds and they are left with no choice but to park their money in yen bonds.
Lifers accelerated their investment in JGBs rather than USTs (Figure 7) in late 2011 to 2012 (a period marked by the European debt crisis and a fight in the US government over raising the debt ceiling; Figure 8), when the UST curve bull flattened.

Japanese investors showed a growing preference for foreign assets from 2010, but they ramped up their JGB investment at one point. We attribute this to the fading appeal of USTs as their yields fell.

At the same time, the trend shifted from bear steepening to bear flattening in 2013-14 (as the taper tantrum subsided), but at the same time lifers accelerated their UST investment. We attribute this to the growing appeal of USTs as their yields rose. Lifers played a part in the flattening trend, in our view. Subsequently, bear flattening turned into bull flattening in 2015, but during this period lifers slowed their UST investment.

- source Nomura

Nomura indicates in their note on the possibility of Lifers showing an increasing preference for domestic and credit instruments. In a Bear Flattener case thanks to the Fed's Rician fading, it is still TINA playing out for the Japanese investor crowd while Mr Watanabe is busy punting BTC. They also made the following point: 
"Lifers tend to reduce their currency-hedge ratio when JPY weakens more than they had expected. In view of this, we would expect lifers to lower their currency hedge ratio and continue investing in USTs if JPY weakens. The median of lifers’ forecast range in their H2 FY17 plans is JPY112 (average for the nine lifers). If JPY strengthens more than this, some lifers could invest in foreign bonds without currency hedges on the assumption that JPY will weaken again, but overall we would expect lifers to show a greater preference for currency hedges."  - source Nomura
Higher yielding bonds for "Bondzilla" the Japanese NIRP monster continue to be the trade du jour. Japanese investors as posited by Nomura tends to be dip buyers ensuring in effect a bear flattening of the US yield curve, that simple.

Moving on to the subject of the predictability recessionary abilities of the US yield curve, we read with interest CITI's Global Economics View note from the 12th of December entitled "Do Yield Curves Predict Recessions in Advanced Economies":
"Why would the yield curve help to predict recessions?
The slope of the yield curve reflects a number of factors, including expected future interest rates and a number of premia related to inflation risks, interest rate risks as well as supply/demand dynamics. Other things equal, higher future growth and higher inflation should imply higher future nominal interest rates. Usually, higher rates/growth/inflation are also associated, other things equal, with higher term premia. We would therefore expect that low growth would be associated with a shallow slope of the yield curve. Low (growth) expectations can also be self-reinforcing. And indeed, in the US, the 10y-3m spread tends to lead GDP growth by about 1-2 years (Figure 1).

An inverted yield curve could also cause lower growth, as it makes bank lending (and other maturity transformation) less profitable, which could reduce banks’ willingness to lend.
However, the slope of the yield curve is affected by a range of factors which are likely to vary over time. (Growth, rate or inflation) expectations may at times also be unduly low (i.e. wrong). In addition, there is no specific reason to expect an inverted yield curve to be equally informative over time. For example, the level of the yield curve is likely to be relevant. Lower trend growth, lower average inflation and lower neutral rates should be expected to lead to a lower average term premium, which could be unrelated to recession risks over time. Additionally, the yield curve reflects much more information than is captured by the value of the 10y- 3m or 10y-2y spread.
Technical factors and policy can also play a role. For instance, when interest rates were deemed to be at an effective lower bound, and even more so when central banks buy long-duration assets, the slope of the yield curve was ‘artificially’ flattened. Large-scale asset purchases by central banks often were deliberately intended to bring down long-term interest rates, in the case of the BoJ’s yield target explicitly. In addition, higher term spreads can often hinder growth (they do of course imply higher borrowing costs). One would therefore perhaps expect the yield curve to be more useful for ‘safe assets.’
Yield Curve and Recessions: Evidence in the US
Several studies suggest that the yield curve is a relatively reliable predictor of (US) recessions. The slope of the Treasury yield curve between the 3-month or 2- year maturity and the 10-year yield turned negative ahead of each of the last seven US recessions since 1970, and flattened substantially ahead of the 1960 recession (Figure 2).

The average lead between the time the 3m-10y curve inverted and the onset of a recession was 12 months (15 for the 2y-10y spread), with the minimum lead at 6 months and the maximum at 17 months (10 and 23 months for the 10y-2y, respectively, Figure 5).

However, the current slope of the yield curve does not suggest that a recession is imminent. First, according to simple models, the current slope of the US yield curve only suggest a ~10% probability of a recession in 12 months (Figure 4).

However, it is worth noting that a 20% probability threshold – which would imply a 48bp spread in the 3m-10yr and a 15bp spread in the 2y-10y – would capture all US post-war recessions with a reasonable lead, with only two false  signals of recession since the 1960s (see Figure 5).1 Since 1960, the average lead from the time the probability crossed the 20% threshold was 14 months for the 10y- 3m spread and 16 months for the 10y-2y spread (Figure 5). From current levels, it would require slightly more than 50bp of further flattening in the 3m-10y to reach this threshold. Even a flat yield curve (i.e. zero spread) would point at only a 31% probability of a recession in 12 months and estimates based on the 10y-2y spread instead give a similar picture.
Second, much of the recent flattening seems to reflect a reduction in the slope of the term premium. According to the NY Fed’s ACM model, term premia account for 47bp of the 77bp flattening in the 10y-2y spread since the start of the year, with changes in expected interest rates explaining the rest. Meanwhile, rate expectations were historically the main driver of flattening and inversions ahead of US recessions (Figure 6), even though it was less clear for the last two recessions (Figure 7).

Evidence outside of the US
The flattening of the yield curve in the US has been much more extensive than in other AEs this year, where 10y-3m slopes have usually flattened by only 10- 20bp (e.g. in Canada, France, Australia, UK) or in fact slightly steepened (Germany, Italy, Japan, Sweden, Figure 8). Compared to the historical average (since 1997), the yield curve is currently only flatter in the US, Japan, Canada and France, and steeper in the remaining countries. The US stands out alongside Japan in how flat the yield curve is relative to its historical value, and in Japan the BoJ actively manages the 10-year JGB yield. However, it is also worth noting that the average long-term level of the slope of the yield curve also varies quite widely among AEs, from 35bp in New Zealand to 188bp in Italy.
Moreover, the slope of the yield curve is not a reliable and useful predictor for recessions in other AEs. For instance, using an inverted yield curve (10y-3m spread) as the criterion to predict recessions would have yielded 5 false positives in Canada since 1960 (against two in the US), 11 in Australia (since 1969), 7 in Japan (since 1966), and 4 in Italy (since 1973). In addition, the yield curve did not invert ahead of recessions in Sweden (1970, 1975 and 1980), Italy (2007 and 2011) and France (2002, 2008 and 2011), and the Eurozone (2011).
Germany appears to be the only other AE where the yield curve inverted with some regularity ahead of recessions, similar to the US (and also with two false positives). Also, in line with the US, the 10y-2y term spread in Germany leads 10y- 3m spread in predicting recessions. In Germany, the current yield spread (133bp for the 10y-3m and 106bp for the 10y-2y) suggests a 10-15% probability of a recession over the next 12 months, but is likely significantly distorted by the ECB’s asset purchase programme (APP).
One reason why yield curves are less informative in other AEs relative to the US could well be that those yield curves are influenced by US asset prices (and relatively more so than their economies are influenced by the US economy) and therefore are more noisy indicators for recessions." - source CITI
With the Fed's Rician narrative, we also tend to agree with CITI that US asset prices do influence other yield curves. But, the distortion from the strong radio propagation anomaly powered by the central banks and in particular the Fed has rendered the Bear Flattening unavoidable. 

The current high level of consumer confidence and recent positive spins coming from US macro data is masking the fact that US consumers are resorting to releveraging with the help of consumer credit. US wages  have finally lead one sell-side pundit such as Bank of America Merrill Lynch to throw in the towel in their Situation Room note from the 13th of December simply entitled "The Phillips curve is dead". Take that Norwegian Blue Parrot! For us, the US consumer is showing growing signs of strain in a US economy plagued by "fixed income" (lack of wage growth) and "floating expenses" (healthcare and rents). 

It appears to us that the Rician fading is finally showing us that 2018, (and we think the second part ) could clearly be the start in the change of narrative and marking the last inning of a very long cycle, leading us to believe that one should become more and more defensive and active when it comes to 2018. (We like Gold Miners at the moment, they have cheapened a lot). 

In relation to our favorite Norwegian Blue Parrot and the bear flattening narrative, we read with interest Bank of America Merrill Lynch Securitization Weekly Overview note from the 15th of December entitled "The wheel is turning - cycle talk":
"As 2017 winds down, we turn philosophical and consider the role of cycles in our lives, starting with the fixed length physical cycle of a year, defined by the earth’s rotation around the sun, and extending that to related fixed length economic and political cycles such as a tax or fiscal year and political terms (for example, 4-year presidential terms). Currently, for example, 2017 is drawing to a close and Republicans have a sense of urgency to complete a tax deal: the calendar year is interacting with the political cycle to encourage political progress.
We consider how these fixed length economic and political cycles interact to form variable length economic and market cycles and where we might be in the current cycle. At a minimum, we think that success on tax reform means the chances the up wave in the current economic/market cycle extends out to 2020 improve. This bodes well for securitized products, for as we noted in our Year Ahead Outlook, while tax reform is sold as a GDP/income boost story, the big gains will come in wealth, particularly for those long risky assets, including securitized products.
Translating this into economic numbers, unemployment is probably all that really matters when it comes to thinking about economic and market cycles, including the current one. Chart 1 shows the headline unemployment rate back to 1965.

The wave-like cyclical pattern is evident, even if the length and the minimum and maximum of each cycle are variable. Importantly, with the unemployment rate currently at 4.1%, it is at the third lowest cyclical level of the past 50 years; prior cycle lows came in 1968 (3.2%) and 2000 (3.8%). Surely, given this, we must be nearing a bottom, and the economic – and market – cycle is coming to an end.
As a reminder, this bottoming and economic/market cycle process usually happens because the Fed thinks low unemployment will create inflation and raises rates, bringing an end to the cycle (Chart 2, Fed Funds v unemployment rate).

With another rate hike from the Fed this week, it appears as if that’s how things are playing out once again; eventually, the hikes will matter and unemployment will move up in response to rate hikes. But, as some of our colleagues have suggested (see “The search for missing inflation” and “The Philips curve is dead”), this time is a little different: problematic inflation has failed to surface in spite of the collapse in the unemployment rate over the past 8 years. Is inflation the looming problem the Fed’s actions seem to imply? If it’s  not, perhaps unemployment still has more downside than Chart 1 suggests. Perhaps the Fed will change its tune.
An alternative view on unemployment, and the economic cycle: the profound impact of income inequality
Our primary theory for why inflation is missing and the Philips curve is dead is the rise in income inequality (Chart 3) over the past 50 years.

Wage pressures at the low end of the income spectrum just aren’t there, due to globalization, demographics, union weakness, technology displacement, etc. It appears as if wage growth at the high end of the income spectrum can only do so much (not a lot) to drive overall inflation. We also think the drop in labor force participation (Chart 4) is a contributing factor, with so many leaving the labor force for various reasons since 2000: hopelessness, drug addiction, retirement, disability, etc.

The unemployment rate is understated by virtue of many dropping out of the labor force; as a result, we may well not at an inflationary-inducing level of unemployment. The cyclical view of Chart 1 may not be as scary relative to inflation as it may look; as a result, unemployment may actually have room to move quite a bit lower.
To account for these factors, we compute an alternative measure of unemployment, U’, as follows (U is the headline inflation rate):
U’ = U * exp(income inequality)^2 / labor participation rate
We normalize the income inequality and labor force participation data to the low in unemployment in 1968. Chart 5 shows the history of this alternative measure, along with the standard unemployment rate. The cyclical nature of the data is preserved, but it provides a somewhat different picture of the minima and maxima.
Specifically, it shows a very different picture on the all-time peak level of unemployment. The massive spike in unemployment associated with the Great Financial Crisis looks far worse than the comparable level of unemployment that was observed in the early 1980s, and more accurately captures the unemployment zeitgeists of the respective periods. In this framework, 2008-2009 shows as the catastrophe that it was.
Similarly, the current level of unemployment does not appear to be as low as the nominal expression of 4.1% unemployment, which says that times right now are about as good as they have ever been over the past 50 years. We think it’s fair to say that many do not feel that way. The alternative measure we compute has the adjusted unemployment rate at 5.6%. In this view, there has been significant progress made on employment over the past eight years, but the inequality and participation adjusted unemployment rate is still fairly elevated by historical standards and is well above the prior cycle lows of 2000 (4.8%) and 1968 (3.2%). This data seems to jive with zeitgeist a little more than the headline number, in our view.
When we do a Philips curve view of unemployment and inflation (Chart 6, PCE core inflation v 1/U’), we see that the current level of adjusted unemployment has not done much to trigger higher inflation.

Perhaps it’s missing inflation, or perhaps unemployment has just not dropped enough to trigger inflation. We’ll let readers decide. We believe that, eventually, declining unemployment will trigger inflation and that, eventually, it will be appropriate for the Fed to hike aggressively; we’re just not there yet. According to this measure, though, we need significantly lower unemployment to get us there.
Consider some math to assess how much lower. Assuming there won’t be much change in inequality or labor participation in the next few years, this means most of the decline will have to be driven by declining headline unemployment. Using the above formula, to get us back to the 2000 level of 4.8% adjusted unemployment rate, the headline unemployment number needs to drop to 3.45%. To get us back to the 1968 level of 3.2% adjusted unemployment (note that BofAML economist Ethan Harris thinks the  current inflation cycle may be similar to the late 1960s, so that may be the more relevant data point), this cycle would need to see the headline unemployment rate hit 2.35%. Obviously, these are extraordinarily low numbers and it would take a while to get there, but we think they may be useful in thinking about how long the current cycle might last.
Low unemployment and wealth
In our Year Ahead Outlook, we noted that the true goal of tax reform is wealth accumulation. Reform is sold as a GDP/income boost story, but the big gains will come in wealth, particularly for those long risky assets, including securitized products. With that in mind, we republish the chart showing household net worth as a % of disposable income per capita versus the inverse of the headline unemployment rate (Chart 7, quarterly data).

We also recreate that chart using the adjusted unemployment rate (Chart 8, annual data); Chart 8 suggests significant additional upside is possible, as the unemployment gauge is not particularly stretched.

Our view is that this measure of household wealth will not peak until about the time the unemployment rate hits the cycle low and turns higher; perhaps it will lead by a quarter or two, but it will be roughly coincidental. In other words, the length of the up wave in cyclical wealth accumulation that began in 2009, around the time unemployment peaked, will ultimately be determined by how low unemployment can go and how long it will take to get there." - source Bank of America Merrill Lynch
It reminds us about the Laffer Curve which used to mean the following: "Too Much Tax Kills the Tax". With the upcoming Tax deals mostly benefiting the 1%, we would opine that "Too Much Wealth Effects Kills the Wealth" but we would digress...

For our final chart below and point for the year, we indicated in our recent musings that once the tax deal is a done deal, US corporates with more clarity ahead could resume/start some M&A typical in the late stage of the credit cycle game in the first part of 2018.


  • Final charts -  Just "bid'em up".
As a tongue in cheek reference to the mighty Bruce Wasserstein aka "Bid 'Em Up Bruce", the M&A legend, our final charts comes from Wells Fargo from their Credit Spotlight note from the 27th of November entitled "Potential for M&A Boom in 2018", and displays US corporates $2 trillion "dry powder" as well as the S&P 500 P/E and deal premium and implied accretive cash deal multiples. If you think some valuations are getting "silly", we think it is about to get "sillier":
Large-cap Non-Financial U.S. companies have $2 trillion of cash, much of which may be freed up for use of M&A in 2018. With the expected certainty of tax policy, combined with the potential of this cash being freed up, next year M&A may rise to a new high. Not only would this allow for M&A to return to a more normalized level given the level of equity prices, but we could see some make-up for the lack of M&A this year. With bond yields remaining low, companies can afford to pay high multiples and still have the acquisition be accretive to earnings. The gap between what the average company would cost to buy compared to what the average IG company can afford to pay and have the deal be accretive to earnings if financed with 100% debt remains quite wide. This gulf has been a hallmark of this cycle and is contributing to the desire for IG companies to engage in debt-financed M&A, despite elevated company valuations. Currently, a company making a fully debt-financed acquisition and funding at the average of the IG market can pay up to 47x earnings for a company and still have the deal be accretive to earnings, even without accounting for synergies. This compares to only 20x for much of the 1990s, which is below where the average company was trading." - source Wells Fargo
If "Rician fading" is a stochastic model for a radio propagation anomaly, then again $2 trillion in cash in conjunction with the U.S. corporate tax rate being cut to 21 percent and cut taxes for wealthy Americans should enable more "wealth effect" with an acceleration of debt-financed M&A à la 2007. Make sure you have your LBO screener at hand in 2018. Just a friendly "credit" advice.

Before we return in 2018 we wish you dear readers a Merry Christmas and a Happy New Year to all your family and friends. 

Don't hesitate to interact more with us either via comments, through twitter, LinkedIn, or on our facebook page in 2018. As a reminder you can as well subscribe to our musings by registering for e-mail delivery via the blog.

"Old soldiers never die; they just fade away." -  Douglas MacArthur
Stay tuned !

Sunday 10 December 2017

Macro and Credit - Volition

"The true test of a leader is whether his followers will adhere to his cause from their own volition, enduring the most arduous hardships without being forced to do so, and remaining steadfast in the moments of greatest peril." - Xenophon

Looking at the dizzy heights being reached in equities with uninterrupted flows into US Investment Grade with yet another warning coming from the wise wizard from the BIS, namely Claudio Borio in the latest quarterly report, when it came to selecting our title analogy, we reminded ourselves of the term "Volition" from psychology given our fondness for behavioral economics. Volition or will is the cognitive process by which an individual decides on and commits to a particular course of action such as the normalization process undertaken by the Fed. It is defined as purposive striving and is one of the primary human psychological functions. Volition means the power to make your own choices or decisions. As an example of the classical concept of volition, comes from Eliezer Yudkowsky while discussing friendly AI development and "Coherent Extrapolated Volition" in 2004. The author came up with a simple thought experiment: imagine you’re facing two boxes, A and B. One of these boxes, and only one, has a diamond in it – box B. You are now asked to make a guess, whether to choose box A or B, and you chose to open box A. It was your decision to take box A, but your volition was to choose box B, since you wanted the diamond in the first place. Now imagine someone else – Fred – is faced with the same task and you want to help him in his decision by giving the box he chose, box A. Since you know where the diamond is, simply handling him the box isn’t helping. As such, you mentally extrapolate a volition for Fred, based on a version of him that knows where the diamond is, and imagine he actually wants box B. In developing friendly AI, one acting for our best interests, we would have to take care that it would have implemented, from the beginning. The question one might rightly ask is relating to the volition of the Fed, is it really for our best interests? One might wonder.

In this week's conversation, we would like to look at the volition of the Fed in its normalization process and the potential upcoming impacts in 2018 should the Fed, once again be behind the increasing flattish curve.

Synopsis:
  • Macro and Credit - When the going gets tough
  • Final chart -  The Fed can't escape Newtonian gravity

  • Macro and Credit - When the going gets tough
Back in April 2017 in our conversation "Narrative paradigm" we argued that when it comes to credit market the only easy day was yesterday. Given the significant role of beta thanks to cheap credit and the "carry play" supported by low rates volatility, it becomes increasingly difficult for us to be supportive of at least the European beta play with the level touched by European High Yield. We also indicated recently that we were expecting a significant pick-up in M&A activity in 2018. This means that some investment grade issuers could experience some sucker punches in the form of blowing out credit spreads in 2018 hence the need of reaching out for your LBO screener à la 2007. A raft of M&A transactions in 2018 would clearly reinforce the view of the lateness in the credit cycle. In terms of change of narrative and in continuation of the "synchronicity" we mentioned in our previous post, the Fed's volition should not be taken lightly.

Also, regardless of the economic narrative put forward by many, including November payrolls rising by a seasonally adjusted 228K and beating expectations of 200K, leaving the unemployment rate in the US to 4.1%, rising wage pressure remains relatively absent. On top of that, surging consumer confidence and modest income growth should trigger much stronger loan demand in our "credit book". One could argue that the credit impulse in the US is tepid at best, no offense to the volition of the Fed. This is indicated by Wells Fargo in their Interest Weekly note from November 29th entitled "Mixed Credit Trends Among Businesses and Consumers":
"In the ninth year of the current economic expansion, credit standards point to a varied, but stable, outlook. Recent data suggest business lending demand has fallen, while consumer demand has seen modest gains.
Banks Optimistic (Yet Cautious) Approach to Lending
Banks continue to relax lending standards on business, commercial and industrial (C&I) loans and mortgages, as banks’ willingness to make loans has gained some stability. A slowdown in banks’ willingness to lend is customary with late cycle expansion. As seen in the below graph, banks’ willingness to extend credit follows a cycle-like trend.

At the start of an economic expansion, banks appear very willing to extend credit. But as the cycle matures, they become less willing to extend credit, and, in turn, tighten their standards in a cautionary nature.
Banks have reported increased competition from other bank and nonbank lenders, which in part, has led to the easing of standards. Continued loosening of lending standards of C&I and mortgage loans points to sustained confidence in the current state of the expansion. However, banks continue to tighten standards on credit card and auto loans, which may signal some caution in the consumer sector as the economic cycle ages.
Slowly Growing Consumer Demand for Loans
With an unemployment rate of just 4.1 percent, and in an environment of modest income growth and surging consumer confidence, theory would suggest robust loan demand. Loan demand, however, remains muted, pointing to a change in consumer sentiment towards debt. Credit card demand has remained unchanged, while we have seen a recent uptick in auto loan demand (below graph).

The rise in demand for auto loans is likely attributable to rebound effects in auto-sales due to damage from the recent hurricanes in Texas and Florida. Consumer demand for mortgage loans has slowed, which is expected to reverse as existing home sales edge higher off a recent slowdown.
Without an increase in income growth, loan demand should continue to increase modestly. Our forecast calls for an uptick in disposable personal income in Q2 2018, due to the effects of the proposed tax reform, followed by a slowdown in disposable personal income growth through the last year of our forecast (2019). Such slowing in income growth suggests consumers may increasingly turn to borrowing to fund their consumption habits in the future.
Demand for business loans remains weak, yet increased strength in business investment suggests that firms have turned to other sources of funding to fuel capital spending (below graph).

The loosening of lending standards, coupled with our positive outlook for business investment, should drive C&I loan demand higher in the near future. However, as the cycle continues to mature, we project equipment spending to slow, likely resulting in a reduction in business demand looking further ahead. The current credit climate insinuates stability within lending practices as growth continues in the ninth year of the current economic expansion."  - source Wells Fargo
While the Fed's volition is to continue with its hiking cycle, financial conditions remain loose and the lack of pressure on wages means that consumers are increasing their leverage through consumer loans in an environment where there is increased competition for business from other players. There are two ways we think the first part of 2018 could play out, "Goldilocks" or "Stagflation" with a sudden rise in inflation expectations that would provide support for a bond bear market narrative with rising interest rates volatility. This is of course without taking into account rising geopolitical risk aka the dreaded "exogenous" factors. We sidestepped various political "exogenous" risks in the first part of 2017 with various elections taking place in Europe in particular. It remains to be seen what will be the trigger of the return of volatility which has so badly wounded many large global macro funds in recent years thanks to its absence. As far as interest volatility is concerned it is at the lowest level in more than 20 years as displayed in the below Wells Fargo chart from their 2018 US Corporate Credit Outlook published on the first of December:
- source Wells Fargo

It won't last rest assured as it is getting late in the game we think and we are already seeing a "synchronized" change in the central banking narrative. Yet some "beta" players continue to be oblivious to the change in the central banking rhetoric. It doesn't mean that the "Goldilocks" environment won't last a little bit more in 2018 for credit, but, we think you should start thinking about playing "defense". This is why back in July we recommended to tactically going for duration again particularly in credit. This has been paying off nicely in 2017 (MDGA - Make Duration Great Again) as indicated by Bank of America Merrill Lynch Credit Market Strategist note from the 8th of December entitled "Year of Duration":
"Year of Duration
That 30-year corporate bonds, which we define in the following as 15+ years, have performed well in 2017 (+11.48% YtD) is no surprise. After all we expected “equity-like” 8-9% total returns in our 2017 outlook piece (see: 2017 US high grade outlook: Let’s do the twist 21 November 2016) on the back of a 27bps decline in 30-year corporate yields (actual: -48bps YtD, Figure 1).

However the surprise has been that credit spreads and Treasury yields were almost equally responsible for this decline in yields (25bps tightening in spreads and 30ps decline in 30-year Treasury yields as of yesterday, Figure 2).

This positive correlation between credit spreads and Treasury yields for such large moves is noteworthy and counterintuitive, but not unseen in the post-crisis years.
How did duration become king?
There are many reasons corporate yield curves are flattening so much including first and foremost that the Fed plans to hike rates actively in an environment of weak inflation data (Figure 3, Figure 4).


This was highlighted by today’s mixed jobs report for November where, although headline nonfarm payrolls were strong at +228k (vs. +195k consensus, Figure 5), average hourly earnings grew only 2.5% YoY (vs. 2.7% expected, Figure 5).

There are many other reasons for flattening yield curves including Treasury’s refunding announcement, where they committed to meet any increased issuance needs using shorter maturities (see: On funding and refunding), the ongoing shift in Europe out of cash and way out the curves (see: QE is dead, long live NIRP), macro risks, etc.
Re-iterate 10s/30s spread curve flatteners
Unusually - and a testament to extreme investor need for both yield and duration – we have seen the 10s/30s corporate spread curve (non-fin, commodities) flatten 8-9bps this year, despite the fact that the 10s/30s Treasury curve has flattened 22bps. More broadly this environment of both flattening Treasury and spread curves has been in place since spreads began tightening in early 2016 (Figure 7). However, the past six months has seen little flattening of the spread curve – but the lack of steepening is actually very impressive as the Treasury curve has flattened 27bps over the same period of time. Needless to say if the Treasury curve continues to bull flatten our recommended non-banks spread curve flattener will not work. We are probably even now reaching levels where further bull flattening actually leads to steeper spread curves. Our house view (see: Global Rates Year Ahead) remains that the Treasury curve is too flat and eventually will re-steepen, which should be very favorable for our spread curve flattener trade. The trade also works with higher interest rates as long as the curve does not flatten.
- source Bank of America Merrill Lynch

As we concluded back in early June in our conversation "Voltage spike", the MDGA trade (Make Duration Great Again) has made a very good come back as indicated by the ETF ZROZ we follow, which delivered a 10.53% return so far this year. For 2018, we continue to favor style over substance, quality that is, over yield chasing from a tactical perspective. As well in another conversation of ours in June this year entitled "Goldilocks principle" we indicated:
"The relentless flattening of the US yield curve shows that in the current inning of the credit cycle, and with a Fed determined in continuing with its hiking path, from a risk-reward perspective, we believe long duration Investment Grade still offers support to the asset class and not only from a fund flows perspective with retail joining late the credit party. On another note the "Trumpflation" narrative has now truly faded to the extent that the deflation trade du jour, long US Treasuries (the long end that is) is back with a vengeance, while inflation expectations has been dwindling on the back of weaker oil prices (after all they still remain "expectations" from our central bankers perspective). " - source Macronomics, June 2017
Our call has been vindicated as well as our contrarian stance against the "bullish" dollar crowd. We continue to believe we will see further weakness ahead for the US dollar in 2018. If we do see additional pressure on the leverage play thanks to the carry trade due to central banking's volition, then one should expect a rise in the Japanese yen we think. As well we continue to see value in the long end of the US curve. Also while gold prices have been weaker recently, the recent pullback as for goldminers looks to us enticing, particularly in the light of a growing risk in "exogenous" factors. If credit options are cheap, gold/gold miners options are "cheaper" as well from a convexity reward perspective in 2018. When it comes to the bull run in credit spreads it has not yet ended and probably will last longer than many might expect, until we hit 11 that is on the "credit amplifier" in true Spinal Tap fashion. We do not see any value left (apart from playing the "dispersion" game with active credit management) in European High Yield at these levels. What is left is "carry" and clearly not enticing enough for us from a risk reward perspective in 2018. We would rather continue playing the US Investment Grade long duration play when it comes to credit allocation. What about US equities being priced to perfection and US High Yield? Given the strong correlations of both asset classes (close to 1 regardless of what some pundits would like to tell us), it is all a matter of "earnings" and they have been so far holding fairly well.

The big risk out there is, as we pointed out the return of the "Big Bad Wolf" aka inflation. This would force the hand of central banks and lead to a more rapid pace in rate hikes leading to some significant repricing on the way. Inflation is our concern numero uno and this would we think be the trigger for higher volatility. This is as well the view of Deutsche Bank from their Asset Allocation note from the 4th of December entitled "What to make of volatility at 50-years lows":
"In our view the leading candidate for a shock that would lead to a sustained increase in vol is a sharp increase in inflation
As noted above, exogenous shocks have historically played a significant role in increasing and sustaining vol at higher levels. Shocks in general of course tend to be inherently unpredictable. Our economists forecast is for a gradual rise in inflation. But in the current context a sharp increase in inflation sticks out as a leading candidate for a shock that raises volatility in a sustained manner in 2018 given the extent to which it is priced in and the likely reaction of monetary policy:
  • Slow inflation priced in. The slowdown in US inflation beginning in March this year had large impacts across asset classes and looks to be priced in (The Growth-Inflation Split, Sep 2017).
  • Few expect a sharp pickup in inflation. The market and FOMC narrative around low inflation with widespread buying into structural declines and a breakdown of traditional relationships looks to have gotten carried away (Six Myths About Inflation, Oct 2017). This suggests few are expecting a sharp pickup in inflation.
  • Four fundamental reasons to expect inflation to move up. First, the lagged impacts of inflation to the growth slowdown during the dollar and oil shocks points to a pick up. Second, the labor market continues to tighten and in our reading there is no reason to believe the traditional Phillips curve has broken down, just swamped by other factors. Third, the direct drag from the past appreciation of the dollar should begin to pass through. Fourth, idiosyncratic factors together have had a strong negative run but tend to mean revert over time. Acting in concert, the four factors clearly have the potential to create a sharp move higher in inflation.
  • A sharp pickup in inflation is likely to be interpreted as a sign of the economy overheating and the Fed embarking on hiking until it ends the cycle. Fed hiking to keep inflation in check has been a—if not the —leading driver of recessions historically. Market expectations of Fed hikes are currently of course far below the Fed’s guidance and a sharp pickup in inflation is likely to entail a significant re-pricing. We don’t see Fed rate hikes from current low levels as ending the cycle any time soon (Is Unprecedented Monetary Policy Easing Creating Secular Stagnation? Jul 2016). But we do see faster rate hikes against the backdrop of a sharp pickup in inflation as having the potential of raising and sustaining vol at higher levels as concerns about the end of the cycle grow." - source Deutsche Bank
Make no mistake, inflation is the "Boogeyman" for financial markets. From the long interesting report from Deutsche Banks, three graph stand out when it comes to heightened risk for large "sigma" events in 2018 given how coiled the volatility spring is:
- source Deutsche Bank

No matter how high your "interval of confidence" is, your VaR model is looking/asking for trouble, particularly your "liquidity" hypothesis. It is time to build some cheap "convexity" defenses as we posited above, not to mention the need for your LBO screener with rising M&A risk and the sucker punches they can deliver to your Investment Grade portfolio à la 2007. Just some thoughts for 2018.

When it comes to the relentless flattening of the US Yield curve as a harbinger for rising recession risk in a classical macro way, we read with interest Deutsche Bank's take from their Global Market Strategy note from the 1st of December entitled "The Fed, the Curve and Risk Assets in the Great Rate Normalization about the recession risk transmission:
"Recession risk transmission
The sharp curve flattening has given renewed life to worries that recession risks are on the rise. Given the flattening, our recession probability metrics – one using the outright slope of the 1s10s yield curve, and one that adjusts the 1s10s curve for the level of 1s – both suggest that November’s flattening was worth a ~6pp increase of the probability that a recession will begin in the next 12 months.

Holding all other inputs equal (the u-rate vs nairu, aggregate hours worked growth, core CPI ex-shelter, and the change in oil prices) puts the probability of a recession beginning in the next 12 months at 16.4% on the unadjusted model, and 26.4% when we adjust the curve for the level of front-end rates. This is still low.
The parallels between the current environment and the Greenspan conundrum are inevitable given the apparent insensitivity of long-end rates to the Fed’s hiking cycle – particularly in the context of a balance sheet unwind that many expected to put sustained steepening pressure on the curve. There is no “conundrum” in our mind – the Fed is simply hiking into a very low neutral rate, with no evidence that inflation is set to materialize to allow them to move faster. This cycle is following a somewhat similar path, however, to the mid 2000’s cycle from the perspective of recession probability.

Now 2 years into the current cycle, the recession probability is at a similar spot to mid-2006 (which was just months away from being within the 12 month window to the start of the great recession).
The risks presented by a flattening yield curve have a clear transmission mechanism to broader conditions as the Fed’s balance sheet shrinks. Before the Fed began to pare its balance sheet we discussed why a steep curve is important to a successful QE unwind – if the curve flattens too much, banks will have no incentive to hold securities over cash earning IOER, meaning non-banks will likely be the marginal buyers of securities no longer owned by the Fed, and deposits will leave the system. This in turn carries meaningful risks to loan growth, which had caught a fair amount of attention given the slowing earlier in the year. Our financial conditions index suggests that C&I growth should be accelerating, though that has not really materialized yet.

The lagged relationship between C&I loans and the Senior Loan Officer survey points to a pick-up in loan growth, but only into the ~5% y/y range, not the double digit range of years past. This has failed to materialize, however, and is a going risk amid the Fed’s balance sheet wind down.
It is still sufficiently early into the Fed’s balance sheet unwind that banks’ reaction function is hard to gauge – securities holdings have risen since the end of September, while cash holdings have been more or less stable and deposits have shown some recent volatility but might still be on an upward trend. So the limited evidence does not yet suggest that this is posing an imminent risk to the economy, but the flatter the curve gets, the more likely banks are to eschew securities for cash, increasing the risk that deposits leave the system, and banks have to pull back on lending."  - source Deutsche Bank

There you have it, no matter how strong the "volition" of the Fed is, the exercise is difficult to execute. While there is no imminent threat to the positive narrative from a fundamentals perspective, the relentless flattening of the Us yield curve is difficult to ignore on top of rising geopolitical exogenous factors. We could easily entertain a blow out of oil prices on the back of exogenous factor in an already tensed Middle East, which would no doubt spill into the inflation expectations surprises on the upside and lead to some repricing and heightened volatility. As we pointed out in our bullet point, when the going gets tough...volition or not.

In our last conversation, we pointed out that "volatility" conundrum was not a paranormal phenomenon, and no matter our strong the Fed's volition and "Forward Guidance" (or imprudence), what the US yield curve is telling us is that no matter how strong the Fed's volition is, they cannot escape Newtonian gravity:
"When it comes to the paranormal phenomena of the low volatility regime instigated by our central bankers, no offense to their narrative" but modern physics still works and normalisation of interest rates should lead to some repricing and a less repressed volatility in conjunction to a fall in the "free put" strike price set up by our central planners in 2018. You probably do not want to hold on too long on "illiquid parts" of your portfolio going forward, given, as many knows, liquidity is indeed a coward." - source Macronomics, November 2017 
Our final chart below points towards the gravitational pull the Fed is facing.

  • Final chart -  The Fed can't escape Newtonian gravity
No matter how strong the spells of the "wizards" at the Fed have been in recent years as pointed out by the wise wizard of the BIS aka Claudio Borio, the Fed's volition is one thing, Newtonian gravity is another. Our final chart comes from Bank of America Merrill Lynch's Weekly Securitization Overview from the 8th of December entitled "The gravity of the yield curve" and displays the 2Y10Y versus the US unemployment rate:
"Given the magnitude of recent yield curve flattening, we revisit our framework where we look at the 2y-10y spreads relative to the unemployment rate back to 1989 (Chart 1).

The flattener has moved more slowly than we anticipated back in 2014, dropping by about 50 bps per year over the last 4 years. Given where unemployment already is, and expectations that it will move lower in the year ahead, we see a strong gravitational pull lower on the curve. Moreover, given that the Fed likely has 4-5 rate hikes in store over the next year (including December 2017), while inflation readings remain low, dropping an additional 50 bps on the curve over the next year seems very achievable, particularly after the 30 bps drop in the 2y-10y spread in the past six weeks.
We note that flattening is not the house call: BofAML rates strategists believe the curve will steepen due to easier fiscal policy, higher deficits, and a higher inflation expectation and the Fed will require higher 5y-10y yields as a precondition to flattening or inverting the curve. We recognize that the flattening process has already taken longer than we expected a few years back, due to multiple “dovish” Fed hikes or pauses, and we acknowledge the potential for what we think would be steepening detours along the way.
Newton's law of gravitation resembles Coulomb's law of electrical forces, which is used to calculate the magnitude of the electrical force arising between two charged bodies. Both are inverse-square laws, where force is inversely proportional to the square of the distance between the bodies. In similar fashion, the relationship between the US 2Y10Y and the US unemployment rate could be seen as inverse-square laws when it comes to the ongoing flattening stance of the US yield curve but we ramble again...

"A man of genius makes no mistakes; his errors are volitional and are the portals of discovery." -  James Joyce
Stay tuned!
 
View My Stats