Loan Losses and Rumors

We want to briefly comment on recent news about a rise in loan loss provisions at US banks and rumors that have lately made waves in this context.

The iceberg – an excellent simile for what we know and what we don’t know… or rather, what we don’t know just 

yet
Image credit: Ralph A. Clevenger

First though, here is a look at the Philadelphia Bank Index (BKX) as well as its ratio to the S&P 500:

Investors seem increasingly worried about the banking sector’s prospects – click to enlarge.

Contrary to widespread expectations, market interest rates on treasury debt have actually declined since the Fed’s rate hike, and the yield curve’s trend toward flattening (which has resumed in the summer of 2015) has yet to significantly reverse. In short, investors may inter alia well be fretting about the fact that an expected improvement in interest margins has as of yet failed to occur (in fact, bank stocks peaked concurrently with the yield curve reaching its widest point of the year in 2015).

10 minus 2 year treasury yield: while the curve widened from February to July 2015, bank stocks rallied and exhibited relative strength. They peaked right around the time when the yield curve reached its widest point of the year – click to enlarge.

However, there is more to it than that. In mid December, on the heels of a secondary peak in the BKX after which the the bulk of the recent leg of the decline occurred, there were two days that saw unusually large trading volume in the stock of a regional bank in Oklahoma, BOK Financial (BOKF). We suspect that well-informed investors decided to get out while the getting was still reasonably good. As reported by Zerohedge, the bank announced in January that it had to book an unexpectedly large credit loss from loans to just a single energy producer. Although the amount was relatively small from a big picture perspective, the decline in bank stocks promptly accelerated.

Unusual trading activity in BOKF shortly after the resumption of the decline in bank stocks, followed by the announcement of energy-related loan losses – click to enlarge.

Since then several more banks, including the biggest ones, have stated that they will have to increase loan loss provisions for energy-related loans. ZH reported over the weekend that the Dallas Fed has reportedly advised banks not to “mark energy-related loan losses to market” and to try their best to work out problem loans in the sector without forcing borrowers into bankruptcy (i.e., they are to employ extend and pretend tactics, presumably in the hope that oil prices will recover sooner rather than later). As Mish points out though, given that mark-to-market accounting has essentially been repealed in early 2009 already, the former advice seems superfluous.

We also cannot judge off the cuff whether it is is possible that the potential losses might already be so large as to imperil bank capital adequacy, but this strikes us as unlikely for the time being, at least for the largest banks. It could well be true for a number of small to mid-sized banks though. That banks may have been urged to go easy on struggling energy producers certainly does sound credible.

In principle the banking system should be more insulated against the kind of problems it experienced in 2008, as massive excess reserves have reduced the dependence of banks on interbank markets and have greatly lowered the probability that big withdrawals by depositors will cause trouble (as compared to the pre-crisis era, a far larger proportion of extant deposit money consists of covered money substitutes these days).

Readers may want to review a few of our previous articles on corporate debt. Most recently we have focused on the fact that it would be a mistake to blithely assume that credit stress in the commodities sector will remain “contained” (see “Junk Bonds Under Pressure” from mid November and “Getting Run Over on 3rd Avenue” from mid December). However, more important may be the background information we discussed in a 2014 article entitled “A Dangerous Boom in Unsound Corporate Debt” (with hindsight it has turned out the the junk bond market had peaked out just a few days earlier).

The article sheds some light on the huge expansion in low grade corporate debt up to that point in time. More and more credit has been granted to increasingly less creditworthy borrowers, combined with covenants offering less and less protection to creditors. This suggests that bond investors rather than banks will actually suffer the greatest damage – not least as new regulations have caused banks to vastly reduce their proprietary bond trading activities. As a result there are no longer any big market makers that can ensure a certain degree of market liquidity, which is obviously a big problem for investors who are forced to sell.

With respect to banks, one must keep in mind though that they still have a lot of indirect exposure junk bonds as well, through their funding of leveraged bond investors, who have inter alia been buyers of assorted structured products (for details on the resurgence in such products and the leverage involved, see “Embracing Leverage Again”). The main purpose of such structures is to let low-rated debt masquerade as something better than it is, by means of creating tranches of different seniority based on assumptions of the overall size of likely credit losses in the event of a downturn. Similar assumptions went spectacularly wrong in mortgage backed CMOs in 2008. Leverage used in bond investments is often especially high and has a tendency to increase as yields go lower, because it becomes increasingly difficult to obtain decent returns without it. This means that the greatest amount of risk is likely to have been amassed at precisely the worst possible point in time.

There is thus a significant danger that the so-called “portfolio channel” effect that central banks have encouraged with their loose monetary policy after the GFC will substantially reverse. This will eventually affect an ever wider circle of borrowers and creditors if forced selling and disappearing liquidity continue to feed on each other in a vicious cycle. In spite of the presumed greater resiliency of the banking system, banks will definitely not be immune to such a development. Their direct exposure to energy loans is only one of several potential problems.

Remarkable Equanimity and Ominous Patterns

We get the feeling from reading articles in the mainstream financial press that there is still a lot of complacency in the markets. For instance, we have come across sotto voce assertions that “this is not 2008” or variations thereof several times already. This is of course true, as every downturn is inherently different (even though Tom McClellan shows that there is an eerie similarity between the 2007/8 and the 2015/16 chart patterns).

We feel however reminded of something Paul Singer of Elliott Management pointed out a while back in this context. He noted that the 2008 experience has probably taught investors to act more quickly once they sense that danger is in the air. In other words, once a decline in risk assets that looks like it is not a run-of-the-mill pullback is underway, it could very easily and quickly accelerate.

As we have recently mentioned, the currently still prevailing complacency is also reflected in “fear measures” such as the VIX. Meanwhile, the deterioration in market internals, the market’s overall technical condition, as well as long term positioning and sentiment indicators continue to look quite dangerous – even if the market is oversold in the short term.

John Murphy of stockcharts has recently pointed out that a well-known bearish technical pattern has become visible in the S&P 500 Index over the past 18 months. Specifically, since October of 2014, the SPX has formed what looks like a slanted head & shoulders pattern. We would add to that observation that the “head” of the pattern looks like a diamond pattern to boot. Both of these patterns are reversal patterns of some statistical significance. Naturally, they can and do occasionally fail (which is bullish if it happens), but their appearance definitely increases the likelihood that a major trend reversal is underway.

Slanted H&S pattern in the SPX. The “neckline” has already been slightly undercut – click to enlarge.

Murphy also reminds us that the “All World ex-US” index is by now clearly in a bear market, after having put in a widely spaced double top:

The FTSE All World ex-US Index has declined below both its October 2014 and August 2015 lows. Recently it has even fallen below the lowest close of 2013, to a level last seen in late October 2012. This was incidentally shortly before the beginning of the Fed’s QE3 program – click to enlarge.

It is a bit incongruent that there is seemingly still so much nonchalance about the market’s prospects while so many obvious warning signs are in evidence. One aspect of the stock market’s recent move strikes us as a bearish confirming indicator, namely the fact that stocks are now finally catching up to the trend in risky corporate debt.

A lag between the trend of the latter and the trend of the former is typically seen at every important market peak. Once stocks do begin to follow junk debt lower, it is basically an “endorsement” that shows that one has to take the trend in bonds seriously. In other words, if it was ever contained, it isn’t any longer.

The S&P 500 compared to junk bond proxy HYG (an ETF holding high yield bonds, not adjusted for dividend income). Junk bonds have been leading the way lower since mid 2014 – click to enlarge.

Conclusion

Based on the fact that the recent decline in junk bonds and stocks is different from the 2007/8 event in terms of the fundamental backdrop, it is widely held that has to be less serious. This assessment may turn out to be wrong. In fact, we believe that the crisis has never really ended – it has merely been masked by papering it over with enormous money supply inflation (US money TMS-2 +116% since 2008) and deficit spending (total federal debt +92% since 2008). Some of the risks have been shifted from the banking sector to the government and new risks have been incurred by other market participants.

Broad money supply growth remains relatively brisk, which normally lowers the prospects of a very large decline in risk assets, but then again, growth in narrow money AMS (resp. TMS-1, close to M1) has slowed considerably more. The slowdown in its growth rate may already suffice to capsize the bubble boat (we plan to soon discuss in these pages in what way these measures differ from an analytical perspective).

It was easier to pinpoint the weaknesses and guess what could happen prior and during the 2008 event, as the bubble was concentrated in a single highly important sector. By contrast, the echo boom has been somewhat more opaque, as bubble activities seem to be spread across numerous sectors – aside from the fact that malinvestment in commodities and especially in the energy sector has obviously gone through the roof in recent years. In this sense, what is happening in energy sector debt strikes us as likely just the tip of the iceberg.

Charts by: StockCharts, St. Louis Federal Reserve Research