John Hussman's Latest Comments on the Bubble

In his newest weekly column, John Hussman talks about a feature of the current echo bubble era that we believe will turn out to be an extremely important one. Readers of this site know of course that we have frequently sung from the same hymn sheet, but it is a topic the significance of which cannot be stressed enough.  After briefly recapping the history of the housing bubble and the ensuing credit crisis, Hussman writes:

“Now, as we observed in periods like 1973-74, 1987, and 2000-2002, severe equity market losses do not necessarily produce credit crises in themselves. The holder of the security takes the loss, and that’s about it. There may be some economic effects from reduced spending and investment, but there is no need for systemic consequences. In contrast, the 2007-2009 episode turned into a profound credit crisis because the owners of the vulnerable securities – banks and Wall Street institutions – had highly leveraged exposure to them, so losing even a moderate percentage of their total assets was enough to wipe out their capital and make those institutions insolvent or nearly-so.

At present, the major risk to economic stability is not that the stock market is strenuously overvalued, but that so much low-quality debt has been issued, and so many of the assets that support that debt are based on either equities, or corporate profits that rely on record profit margins to be sustained permanently. In short, equity losses are just losses, even if prices fall in half. But credit strains can produce a chain of bankruptcies when the holders are each highly leveraged. That risk has not been removed from the economy by recent Fed policies. If anything, it is being amplified by the day as the volume of low quality credit issuance has again spun out of control.

Leaving aside for a moment that securities prices are currently (and almost always) distorted by monetary policy, we can state the following: Insofar as the stock market is a mirror of the economy, it reflects only profits and losses that have already arisen in the course of activities in the real economy.

Even in an economy hampered by monetary pumping, the relative prices of different stocks essentially tell us which economic activities have found favor in the marketplace. What the stock market then does is to distribute the resultant profits and losses among investors according to the foresight they have displayed in picking stocks.

There is of course also a discount/expectations component embedded in the prices of stocks – but this forward looking premium also reflects an appraisal of past decisions and the effect they are held to have on future returns. To illustrate this with an example: if a company announces that it has begun to manufacture a new product (say, the iPad, for instance), then this decision is already in the past, but market participants now may place a premium on the company's stock because they estimate the decision to have a positive effect on the company's future profitability.

In any case, the point we are driving at is only this: insofar as stock prices reflect the investment decisions taken in the real economy, the profits and losses of investors in the stock market do not affect the world at large. Just as John Hussman writes above, they only affect investors in the market.  Mr. Hussman also had a few choice comments on the thinking that informs today's stock market investors that strike us as quite pertinent with respect to the current situation:

“My sense is that investors have indeed abandoned basic arithmetic here, and are instead engaging in a sort of loose thinking called “hyperbolic discounting” – the willingness to impatiently accept very small payoffs today in preference to larger rewards that could otherwise be obtained by being patient. While a number of studies have demonstrated that hyperbolic discounting is often a good description of how human beings behave in many situations, it invariably results in terrible investment decisions, particularly for long-term investors. As one economist put it, “they make choices today that their future self would prefer not to have made.” In effect, zero interest rates have made investors willing to accept any risk, no matter how extreme, in order to avoid the discomfort of getting nothing in the moment.

[...] many investors realize that the most reliable valuation measures have never been higher except in the advance to the 2000 peak (and for some measures the 1929 and 2007 peaks), but they have started to treat these prior pre-crash peaks as objectives to be attained.

While recent years have diminished our belief that severely overvalued, overbought, overbullish syndromes are sufficient to derail further speculation, it’s worth observing that present valuations are much closer to those prior peaks than is widely assumed.

This is indeed the kind of thinking we come across almost daily in article after article discussing the stock market and its valuation. A great many people seem to assume that the only relevant yardstick in determining whether it is a bubble is the manic peak of valuations attained in 2000. Not only that, it is indeed implied  (just as Mr. Hussman points out) that this manic peak should therefore be seen as an objective to be attained. Right, and once we have attained it, every single one of these wise guys will sell at the top tick and no-one will lose a cent.

A quick illustration of what we think of this plan

The Achilles Heel

In light of the above deliberations about the effect of losses in the stock market, the Achilles heel of the bubble is therefore something else, and that is the extent to which unsound credit has poisoned the system. Mr. Hussman notes that in 2008, the banks themselves had so much leveraged exposure to the product they were creating (i.e., mortgage securities backed by ever dodgier loans, which enjoyed high ratings by the expedient of housing them in structured credit  vehicles), that even a fairly small decline in their value would have sufficed to wipe out their hopelessly inadequate capital base.

One could therefore argue that the current bubble in non-investment grade corporate debt is different, because new regulations have forced the banks to vastly reduce their proprietary trading activities. They have essentially withdrawn from the function of “market makers” in these securities. However, this argument would gloss over the interdependence of credit markets. It also overlooks the important fact that the manner in which investors view corporate debt today is exactly equivalent to the views that prevailed with respect to mortgage backed securities during the housing bubble. It is inter alia this datum that leads to this high degree of complacency:

Annual default rates of high yield bonds 1980-2013 

This chart is telling investors is that even under very stressful conditions, high yield default rates tended to remain at a “manageable level”. Moreover, they are currently extremely low – and are forecast to fall to even lower levels.

Nevertheless, there are a number of problems. One of them is that as yields keep declining, and with administered interest rates held at zero by central banks, the temptation to employ leverage to juice returns keeps growing. Not only that, the hunt for yield has also led to a revival of structured products that actually have a history of being a) fairly safe and b) providing high recovery rates in the event of defaults of the underlying credits.

It may be precisely this history that makes the especially dangerous: just as investors believed that there “could never be a nationwide decline in house prices” and that defaults on mortgages would always conform to certain historical norms, they now hold analogous beliefs in the context of high yield bonds and structures like CLOs (collateralized debt obligations), which are often backed by leveraged loans.

Given the fact that CLOs – similar to the mortgage backed structured credit products of yore – enjoy very high credit ratings (these are a function of  overcollateralization and the splicing of these securities into tranches of different seniority), they also have very low yields, in spite of the high yields paid by the underlying credits. As a result,investors often leverage their positions up to ten times to achieve decent returns.

We have discussed the comeback of these structured products in a previous article which you might want to review for additional background information: “Embracing Leverage Again – More Credit Insanity”. Who provides the leverage? The banks of course. In other words, even if they do not hold the securities themselves, they remain fully exposed to their risks.

Most leveraged loans are produced in the course of LBOs (leveraged buy-outs) and other M&A activities. They are as a rule safer than junk bonds, as they are usually secured by the borrower's operating assets. Moreover, investors have far better access to information about the borrower and covenants tend to be more protective. Historically, both default rates and the volatility of leveraged loans have been lower than those of HY bonds. However, one must not forget that leveraged loan borrowers are the same type of borrowers that are issuing HY bonds: they are below investment grade, and quite often they have a very limited track record as well (hence they opt for loans that force them to agree to a lot more disclosure and tighter covenants rather then turning to bond issuance).

Obviously though, every credit bubble has features that are quite similar. Specifically, as more and more credit is issued to satisfy soaring demand, the average quality of borrowers must perforce decline. At some point, one simply runs out of relatively sound borrowers and has to move down the quality chain to those one may have shunned earlier in the process. This is precisely what the mortgage credit bubble ultimately tripped over: in the final few years, more and more “NINJA” loans were issued – to people with “no income, no job and no assets”. Such a credit structure can only survive as long as two things are assured: continued asset appreciation and a continuation, or even acceleration, of credit issuance. This combination allows the repeated refinancing of loans to borrowers that are actually not creditworthy.

As corporate borrowers are currently practically pushed to supply the market with more debt means also that the loan covenants which are supposed to be more stringent in the case of leveraged loans will tend to become weaker and weaker the more deals are struck – this is equivalent to what happens in HY bonds, where covenants are likewise offering less and less protection, and the issuance of bonds like 'PIK' (payment in kind) bonds has soared. It is of course understandable that low quality borrowers love to issue bonds that give them the opportunity to “pay” their debts by issuing even more debt.

In light of the above, let us consider two relatively recent charts:

US HY bond and leveraged loan issuance, 2002-2013 

As you can see, the issuance of leveraged loans has recently soared even more than junk bond issuance. While junk bond issuance in both 2012 and 2013 was at new record highs more then 100% above the levels seen in 2006 and 2007, it took leveraged loans a bit longer to set a similar record (i.e., almost doubling the previously seen record highs of 2007). This “catch-up move” was accompanied by soaring issuance of CLOs, which are probably a main source of demand for leveraged loans:

US CLO issuance goes parabolic as well 

To summarize: we have borrowers of little creditworthiness borrowing money at very high yields with ever looser debt covenants. Then these loans are packaged into structured vehicles, which make it possible for non-investment grade debt to receive investment grade ratings (giving the 'packagers' a nice spread), which then are bought by investors with leverage of up to 1:10. The loans to investors are in turn provided by the very same banks that arranged the leveraged loans and CLOs (and often themselves hold various tranches of leveraged loans. These loans are usually syndicated, and consist of several facilities such as amortizing loan portions and bullet maturity term loans, in which the principal is paid at maturity. While banks tend to hold all types of these tranches, it is mainly the bullet maturity term types that are marketed to investors).

And all of this happens with administered central bank interest rates at zero, yields on bonds of all credit quality classes at or near record lows, a vast consensus that “inflation” (rising consumer prices) cannot possibly happen, a marked deterioration of investor protection in the event of defaults, while expected 12 month default rates are among the lowest in history.

What could possibly go wrong?

High yield bond issuance in Europe is soaring as well, even as Europe's economies continue to flounder 

The Expected Default Rates Trap

Lastly, we want to briefly comment on the one data point that probably contributes more than any other to investor complacency: the expected 12 month default rate. All credit agencies put this expected rate at a new low for the move in 2014, not unreasonably arguing that because companies have issued so much debt, they are currently flush with cash and can postpone their refinancing plans if necessary.

It is also forecast that default rates will increase only slightly in 2015-2016. Naturally, this forecast must be based on an underlying assumption about future economic developments. Evidently, no-one believes that a recession or a major bust might interfere with the current happy state of affairs.

Interestingly, while default rates are near record lows, recovery rates are recently  declining sharply– a sign that the quality of borrowers that are currently keeling over is very bad indeed 

Very low default rates in HY bonds, but recovery percentages have plummeted from 59.4% in 2011 to 47.7% recently

Why are default rates so low when both economic growth and corporate profit growth are really quite underwhelming? Just look at the data of US real GDP growth and corporate profit growth compared to HY default rates below – and this chart doesn't even include the plunge in Q1 2014 GDP yet (default rates have not worsened in its wake). Note also that economy-wide, corporate profit growth has actually turned negative in Q1 2014, but even that has so far had no negative effect on HY default rates:

HY defaults, GDP growth and corporate profit growth 

The reason becomes clear when considering the use of HY bond issuance proceeds:

Globally, between 53% and 64% if HY bond proceeds have been used for refinancing between 2009 and 2013 – i.e., replacing maturing debt with new debt. The proportion employed for M&A activities is still far lower than in 2004-2007, but a lot of M&A funding probably involves leveraged loans at the moment

Only in the 2001-2003 post technology bubble recession was a higher proportion of HY debt employed for refinancing purposes. Obviously, with demand for HY debt extremely high, even bad credits have little problem refinancing their debt. In other words, we are observing a feedback loop here: on the one hand, investors are emboldened by low default rates to keep refinancing HY borrowers, on the other hand it is precisely because they keep refinancing them that default rates remain so low.

As long as enough new money is provided the show can go on, but obviously, this is highly dependent on investor confidence. The question is therefore, when is investor confidence at its most vulnerable? Ironically, this is precisely when it is at its highest. When spreads and volatility are extremely low, they indicate a high point in confidence. This implies however that confidence can no longer improve further. Once things are “as good as they ever get”, there is just one way left for them to go: they can only get worse.

As an aside to this, whenever a CEO of a company in a cyclical business mentions that business is better than it has ever been, one should immediately sell the shares for the very same reason.

One last chart we want to show is the ratings mix of new junk bond issuance over the years. This chart indicates that the last time when the ratings mix was fairly strong was in 2002-2003. Ever since, junk bond issuance has tended to become junkier, although the 2013 vintage showed a little bit of improvement in that respect:

Rating mix of new HY issuance

The rating designations above are those employed by Fitch. Here is a brief explanation as to what these ratings mean:

BB: Speculative.

‘BB’ ratings indicate an elevated vulnerability to default risk, particularly in the event of adverse changes in business or  economic conditions over time; however, business or financial flexibility exists which supports the servicing of financial  commitments.

B: Highly speculative.

‘B’ ratings indicate that material default risk is present, but a limited margin of safety remains. Financial commitments  are currently being met; however, capacity for continued payment is vulnerable to deterioration in the business and economic environment.

CCC: Substantial credit risk. Default is a real possibility.

CC: Very high levels of credit risk. Default of some kind appears probable.

C: Exceptionally high levels of credit risk. Default is imminent or inevitable, or the issuer is in standstill.

Conclusion:

The area of the market that is most likely to create problems for the echo bubble is corporate debt. The proportion of high yield debt issuance to total corporate debt issuance has soared to new record highs this year. Given that capital spending remains fairly anemic in a historical context, we can conclude that the reason for this soaring junk debt issuance is mainly strong investor demand – an unsettling parallel to what occurred in 2005-2007 in mortgage backed debt. Moreover, we can conclude that a great many of the debt proceeds, aside from refinancing maturing debt, are finding their way into various forms of financial engineering.

This is confirmed by soaring stock buybacks over the past few years. Buying back stocks at extremely high valuations by issuing debt strikes us as a very dangerous exercise – incidentally, this also happened in 2005-2007. Whether banks are really less exposed to the underlying risks as a result of stricter regulations may be debatable, but one suspects that their level of exposure is much higher than is generally assumed. The interconnectedness of credit markets has not magically disappeared, and the fractionally reserved banking system remains central to these markets.