Macro and Credit - The Lindemann criterion

"We're just a bubble in a boiling pot." - Jack Johnson

Looking at the acceleration of inflows, with record flows piling into Investment Grade, Emerging Markets (EM) and equities, when it came to selecting this week's title analogy, we reacquainted ourselves with the "Lindemann criterion", being the first theory explaining the mechanism of melting in the bulk. An attempt to predict the bulk melting point of crystalline materials was first made in 1910 by Frederick Lindemann. The idea behind the theory was the observation that the average amplitude of thermal vibrations increases with increasing temperature. Melting initiates when the amplitude of vibration becomes large enough for adjacent atoms to partly occupy the same space. The Lindemann criterion stated that melting is expected when the vibration root mean square amplitude exceeds a threshold value. Quantitative calculations based on the model are not easy, hence Lindemann offered a simple criterion: melting might be expected when the root mean vibration amplitude exceeds a certain threshold value (namely when the amplitude reaches at least 10% of the nearest neighbor distance). In similar fashion, melting up might be expected with such strong inflows into various asset classes. As we have repeated in numerous conversations, you can expect the credit amplifier to reach 11 in true Spinal Tap fashion.  On a side note Lindemann model is based on harmonic forces, which never give way, whereas melting must involve "bond breaking". This is another serious defect of the model. Furthermore numerous experiments carried out at high pressures indicate that the Lindemann model does not estimate adequately the pressure dependence of the melting temperature, but,  the predictive success of the Lindemann melting criterion lent support to the belief that melting could be a gradual process, beginning within the solid at temperatures below the melting point. 

In this week's conversation, we would like to look at cycles in credit in conjunction with what is happening in US Consumer Credit as well as boiling markets and bubbles.

Synopsis:

• Macro and Credit - The credit boiling frog
• Final charts - Fund flows have a tendency to follow total returns

• Macro and Credit - The credit boiling frog

Whereas the heat continues to be "on" when it comes to fund inflows and in particular for Investment Grade credit, one might wonder from a "Lindemann criterion" approach, when could we reach the boiling point in this on-going "melt-up" phase. Some pundits have even called this phase "euphoria" given the strength of the start of the year performance wise in various asset classes including equities. 

As pointed out recently on Twitter by our good friend Tiho Brkan when it comes to the credit boiling frog bullet point analogy, for us there is no real value left in European High Yield to say the least:

"For the first time (possibly ever?) we have European Junk Bonds yielding less than European equities.

You have to hand it to the ECB. They really a did a terrific job bailing out the whole continent. Not a single bankruptcy was allowed!" - source Tiho Brkan - Twitter feed

No wonder flows have been very strong into equities in Europe as pointed out by Bank of America Merrill Lynch in their Follow The Flow note from the 19th of January 2018 entitled "Turbo-boosted flows":

"Inflows into IG, EM and equities continue stronger

The reach for yield continues for another week. IG, equity and EM debt funds have had a great start of the year. IG fund flows have tripled (3wks YoY); flows into equities have moved from marginally flat to $6.4bn, and flows into EM debt have more than doubled. Nonetheless, flows into HY funds have flipped to negative, on the back of rising idiosyncratic risk. With spreads continuing tighter in the IG space and rates vol remaining close to the lows we see no reason for flows to stop in the IG space.

Over the past week…

High grade funds, continued to record inflows their fourth consecutive week. The monthly December data also reflected the calm of the holiday season with the second smallest inflow of 2017 – second to January. High yield fund flows remained on negative territory recording their 10th consecutive week of outflow. Looking into the domicile breakdown for last week, as chart 13 shows, European-focused HY funds were the main source of outflows, but the US-focused and globally-focused funds also recorded withdrawals. Monthly data for December showed a 2nd month of outflows.

Government bond funds recorded an inflow erasing last week’s outflow and a strong inflow during the month of December as well. Overall, Fixed Income funds flows remained sizable – and positive – for the fourth week in a row. As for the month of December, inflows –despite being the smallest in 2017 due to seasonality - confirmed the overall inflow trend that was seen throughout the year.

European equity funds continued on their positive streak for a third week, with inflows accelerating. Last week inflow was the largest in 36 weeks. However, in December the asset class recorded a large outflow – the largest in 14 months." - source Bank of America Merrill Lynch

As long as rates volatility remains muted, there is no end in sight for the "goldilocks environment" and the Investment Grade TINA (There Is No Alternative) trade supported by the Japanese investment crowd at least when it comes to US credit markets. As we pointed out last week when it comes to US TIPS being enticing and rising risks of inflation surprises, back in October 2017 in our conversation "Who's Afraid of the Big Bad Wolf?" we asked ourselves if indeed the game was turning and if we should switch camp from the "deflationista" towards the "inflationista" camp. Flow wise, what we are seeing is large swath of investors piling into the "inflationista" camp and investors as pointed out by Reuters in their article from the 18th of January entitled "Investors scoop up U.S. inflation-protection bonds":

"Investors on Thursday pounced on $13 billion worth of U.S. bonds that offer them protection from faster wage growth and costlier goods and services as their longer-term inflation outlook rose to their highest in almost a year.

Overall bidding for these 10-year Treasury Inflation Protected Securities was the strongest at an auction in over 3-1/2 years with investors buying nearly 90 percent of the supply.

"The allocation into the (TIPS) asset class has been strong," said Michael Pond, head of global inflation-linked research at Barclays in New York. "It's all related to the pickup in global growth."

Barclays' Pond said the run-up in TIPS may be overdone in the short term. "Evaluation is getting a bit stretched," he said." - source Reuters

TIPS focused funds have reached "boiling point" to the tune of $68.01 billion of inflows in the asset class in the week ending the 10th of January according to Lipper. As the Lindemann melting criterion has shown us, melting is indeed a gradual process. Have we reached the boiling point yet? Could we be on the cusp of a vicious bear market as anticipated daily by the perma bear crowd? One thing for certain, as we pointed out again last week, you need core US CPI to tick up significantly for a bear market to materialize for our equities friends. We do not think we have reached that point yet. Financial conditions continue to remain loose which makes this credit cycle particularly long and extended.

Not all credit cycles are the same and business cycles as well tend to be different across sectors over time. On that subject we read with interest Wells Fargo Interest Rate Weekly note from the 17th of January entitled "Cycles in Credit, Economics and National Finance":

"Business cycle trends differ across major sectors; while some series are mean-reverting, others are shifting over time. Careful analysis leads to a better understanding of ‘normalcy.’

The Bank Credit Cycle: Mean-Reverting

The Federal Reserve reports a bank’s willingness to make loans on a quarterly basis as part of the Senior Loan Officer Opinion Survey. Such a qualitative measure is telling of lender expectations and is often consulted to predict the coming credit conditions in the market. A bank’s willingness to make loans is a stationary series, which contains no structural breaks. That is, the series acts in a predictable nature, allowing one to refer to it as a benchmark to assess lending practices over a business cycle (below chart).

At the beginning of an expansionary period, as banks are very willing to extend credit, the series rises. Once reaching its peak, a bank’s willingness gradually decreases until turning negative during a recessionary period. That said, it would appear that willingness has peaked in the current expansion, which is in line with our assessment of the business cycle being in the late stage.

Elevated Household Debt: Not Back to “Normal” Ratio

Debt as a percentage of GDP serves as a good measurement of indebted households relative to the size of the economy.

However, analyzing averages can be significantly misleading. As seen in the middle chart, after rising exponentially in the prior expansion, household debt as a percent of GDP never returns to its “average” trend, as portrayed in the 1991-2000 expansionary period. Despite decreasing significantly from its peak prior to the Great Recession, household debt still remains quite elevated, stabilizing around 77 percent. As this series is not mean-reverting, it does not follow a cyclical trend, and the behavior of the series in this cycle is distinct compared to prior expansions. The recent deleveraging is likely attributable to increased caution as households acquired large amounts of debt during the previous expansion, which led to a sharp increase in the debt burden.

Unusual Behavior of Profits

Corporate profits as a share of GDP appear to follow a cyclical pattern in which they gradually increase, peaking late in an expansionary period, before steadily declining during a recession. However, pre-tax corporate profits are non-stationary, and thereby are a less predictable series. Although profit growth as a percent of GDP appears to follow a traditional peak to trough trend, there have also been various structural breaks within the series leading to highly irregular and quite misleading judgements based on prior trends. Although this might not be very surprising, the larger decline being in the 2001 recessionary period seems unusual (bottom chart).

The greatest decline in corporate profits would be assumed to have occurred during the Great Recession, as we saw dramatic movements in a bank’s willingness to lend as well as households acquiring record amounts of debt. That is, one might suspect a similar case of dramatics in the decline of profits during the Great Recession; however, the drop was more significant in 2001." - source Wells Fargo

When it comes to the US economy, US consumer credit matters a lot. We continue to monitor that space given any weakness in US consumer credit could be an additional sign the US economy is reaching a turning point. Also in our October 2017 conversation "Who's Afraid of the Big Bad Wolf?" we asked ourselves what could make the credit cycle turn once we have reached boiling point:

"When it comes to credit and in particular the credit cycle, the growth of private credit matters a lot. If indeed there are signs that the US consumer is getting "maxed out", then there is a chance the credit cycle will turn in earnest, because of too much debt being raised as well for the US consumer. But for now financial conditions are pretty loose. For the credit music to stop, a return of the Big Bad Wolf aka inflation would end the rally still going strong towards eleven in true Spinal Tap fashion. " - source Macronomics, October 2017

From a Lindermann criterion perspective, measuring the level of indebted households matters and in particular the use of Consumer Credit and in particular non-revolving credit we think. In regards to the results from big US banks we read with interest the Financial Times article from the 21st of January indicating a surge in credit card losses:

"US banks suffer 20% jump in credit card losses

Rising soured debts raise concerns about the financial health of middle America

The big four US retail banks sustained a near 20 per cent jump in losses from credit cards in 2017, raising doubts about the ability of consumers to fuel economic expansion.

“People are using their cards to get from pay cheque to pay cheque,” said Charles Peabody, managing director at the Washington-based investment group Compass Point. “There’s an underlying deterioration in the ability of the consumer to keep up with their debt service burden.”

Recently disclosed results showed Citigroup, JPMorgan Chase, Bank of America and Wells Fargo took a combined $12.5bn hit from soured card loans last year, about $2bn more than a year ago.

Banks have ramped up lending, wooing customers with air miles, cashback deals and other offers. The number of open credit card accounts in the US is forecast to reach 488m this year, according to Mercator Advisory Group, a rise of 108m from post-crisis lows in 2010.

Yet borrower delinquencies are outpacing rising balances. While still less than half crisis-era levels, the consultancy forecasts soured credit card loans will reach almost 4.5 per cent of receivables this year, up from 2.92 per cent in 2015."  - source Financial Times

Low inflation and loose central bank policies have indeed played a critical role in moving asset prices towards the boiling point and the ongoing "melt-up" but the narrative is slowly but surely changing as we pointed out on numerous occasions. 

Nonetheless, the most predictive variable for default rates remains credit availability. Senior officer loan survey leads default rates (SLOOs) so it is important to track them. When it comes to consumer credit, as the credit cycle matures, it is becoming essential to monitor what is happening for Credit card ABS we think.  As displayed by Wells Fargo in the below chart in their Credit Card Surveillance Report from the 19th of January, credit card ABS issuance has been increasing in 2017 but it is still relatively tame compared to the years prior to the Great Financial Crisis (GFC):

- source Wells Fargo

While securitization of US consumer credit is comparatively tame compared to what happened before the GFC, US consumer debt surged by the most in over 2 years to $3.8 trillion and jumped by 8.8% in November, the most in two years, to $3.83 trillion, according to the Federal Reserve. Clearly in the coming months US Consumer Credit should be on everyone's radar in conjunction with SLOOs we think.

As per our bullet point the credit frog is slowly but surely heating up for sure as per the Lindemann criterion but, we do not think yet we have reached the boiling point yet in credit we think.

For our final charts we have always thought tracking flows matter, particularly this year the ones coming from Japan given their support for US Credit as a whole. 

• Final charts - Fund flows have a tendency to follow total returns

When it comes to flows the most cited linkage according to CITI is trailing returns. Our final charts come from CITI's European Credit Weekly note from the 19th of January entitled "How durable is the fund flow frenzy" and display fund inflows as a percentage of AUM (Assets Under Management) versus total returns for both US Investment Grade and US High Yield:

"Fund flows have a tendency to follow total returns, both on the way up and on the way down. The relationship holds in IG (Figure 7), in HY (Figure 8) and in equities, and while sometimes coincident, the return series quite often seem to lead at turning points.

This relationship is clearly sending a positive signal at the moment. While it left us very worried in early 2016 (after investors had lost money on almost every asset class in 2015), 2017 returns were almost universally positive. If investors simply chase trailing returns, there is very little reason for them to withdraw their money at present.

Admittedly there are a few wrinkles. In such a risk-on environment seems odd that investors are quite so keen on IG (both outright and relative to trailing returns), and quite so cautious on HY (likewise). It might just be an overhang from the losses of 2015, when investors clearly ended up regretting their enthusiasm for HY in 2010-14.

Something similar is apparent in equities: last year’s $200bn inflow to mutual funds and ETFs – a mere 1% of total holdings – seems relatively paltry when markets are returning 15+ percent. Again, it may just be that investors have been unable to regain the love of the asset class they displayed during the 1990s (Figure 9).

But we would be much more convinced of private investors’ ability to take over from central banks as the main driver of risk assets this year if we saw more elevated and sustained inflows to both HY and equities. Perhaps those are on the verge of showing up, but there is an alternative interpretation which points in the opposite direction.

Cash is king

Just as last year’s outflows from HY funds did not seem to have much negative impact on their performance – perhaps because IG or other investors were crossing over into HY – so we find we get the best understanding of fund flows when we consider the big picture, rather than looking at each asset class individually.

Specifically, we like to look at inflows to all risky (or “long-term”) assets together – IG, HY, govies, EM, equities and alternatives, in mutual funds and ETFs – relative to total inflows into money market funds and deposits. How much of people’s saving is going into risk assets relative to what they a simply trying to keep safely in cash?

This combined flow also exhibits a strong relationship with equity returns, with recent strong inflows at first sight almost exactly where they “should be” (Figure 10).

When risk assets are performing well, investors do most of their saving in risky assets, and keep relatively little in cash.

But look more closely, and another pattern is visible. In each cycle, as the economy emerges from recession, with risk assets priced cheaply and deposit rates low, investors start by doing the vast majority of their saving in risk assets (1992-3, 2003-4, 2009-10). Then as the cycle matures, risk assets become more expensive and deposit rates rise, they do steadily more of their saving in safe assets. Finally as risk assets start to wobble they try and withdraw some money and do all of their saving in cash, precipitating a sell-off, before the cycle starts again (Figure 11).

In addition, both the red and blue lines have exhibited a long-term downward trend. The move in the blue line could just be the relative growth in deposits across cycles. But an alternative interpretation – consistent with investors’ relative reluctance to buy both HY and equities – is that as markets have become more obviously expensive, it has taken steadily lower deposit rates just to prevent them from hoarding cash.

One way or another, the past year’s shift into risk assets, so late in the cycle, is extremely unusual. Even the late 1990s sees nothing like it. It is the exact opposite of what one would normally expect as deposit rates rise. And yet this year needs to see an even bigger such shift if private investors are to absorb the $1tn global hole left as central banks reduce their purchases.

Perhaps the strong start to the year will encourage just such buying on the part of investors. And in €, at least, negative deposit rates will continue to be an incentive for a good while yet. But despite the recent drop in correlations, our instinct remains that markets remain deeply interlinked across both assets and across geographies – meaning that the Fed’s rate rises have significance even in €, and the ECB’s and BoJ’s reductions in purchases have relevance for $ investors alongside the Fed. So while the signal from recent fund flows is undeniably positive, it will take more than just a good few weeks at the start of the year to convince us that 2018 as a whole will see a continuation of this raging bull market." - source CITI

As we posited at the start of our conversation, from a Lindemann criterion perspective, melting up might be expected with such strong inflows into various asset classes, particularly in credit. That simple. Melting is indeed a gradual process, particularly for our credit frog but it seems our central banker are using a pressure cooker, which is interesting because by pressurizing the vessel, they are able to produce much more heat than the boiling point of water, and still have water (inflows) present but we ramble again...

"When the water starts boiling it is foolish to turn off the heat." - Nelson Mandela

Stay tuned !