While I was enjoying five days of fresh powder skiing in Jackson Hole, WY, the equity markets pretty much completed their round trip from euphoria, to fear of an impending crash, and back to euphoria. Once again the corporate bond market and related derivatives were the tells to follow. They weakened a little ahead of the drop in stocks, they steadied as stocks kept dropping, and ultimately they recovered back to their recent highs and violently dragged stocks back up with them.
In the first chart below I flipped the scale of the S&P 500
(INDEXSP:.INX) upside down to show how tightly correlated it traded to changes in my index of Credit Default Swaps of large US financial institutions, and to the more broadly watched Markit HY CDS Index (IBOXHYSE). What doesn't jump out on the chart, but I chronicled on the Buzz & Banter
and on Twitter, is that the financials' CDS index was the tail wagging the dog, leading equities by 24-48 hours throughout the recent swings.
In the second chart, you can see further divergence between the fear that seized the equity markets and the nonchalant behavior of junk bonds, which continued to show positive returns for the year even as stocks dropped 5% in a short and scary period. The icing on the cake is that large corporate-bond types continue buying with reckless abandon ($149 billion in January and $50 billion MTD), seemingly limited only by a lack of supply.
However, rather than trying to merely explain what has already happened, the gist of this article is to attempt to determine if the relationship between corporate bonds/derivatives is a matter of coincidence, correlation, or causality, and how the answer will play into future rips and drops. Coincidence would indicate that the two asset classes move in concert but seemingly out of happenstance, and presumably this should last for relatively short periods of time. Correlation would argue that, though unrelated, the assets show a dependency on each other, and the statistical tendency to move together in the same direction is strong enough that it can be extrapolated into the future. Lastly, causality would introduce a direct nexus between the two.
I'll give the first possibility short shrift: The symbiotic moves of corporate bonds/derivatives on the one hand, and equities on the other, have repeated enough times (see also, Falling Stocks and Corporate Bonds: 5 Charts That Explain the Connection
) to put to rest the notion that they are mere coincidence. What we are left to answer is whether we are looking at correlated events or outright causality. If you are involved in these assets, the answer is not just a matter of wonky curiosity. If it is only correlation, it will last until the correlation weakens enough to no longer be actionable. If it is causality, then the assets will continue to move in unison until one stops fueling the other.
Reduced to more practical trading terms, correlation is tradable until it isn't, and any divergence could be the beginning of the end. Causality, on the other hand, makes a temporary decoupling (like what we saw in the past two weeks) an obvious opportunity.
So which one is it? Depending on the relevant time frame, it may actually be both. Let me explain the longer-term "causality" angle first, because it is by far the more important of the two. For a while now I have embraced and been repeating the mantra defined by Brian Reynolds, former Minyanville contributor and now Chief Market Strategist at Rosenblatt Securities. Brian argues that there is a direct link between i) the ability of companies to sell bonds / the ravenous appetite of bond players to buy them, and ii) stocks rising because companies use the bond proceeds to buy back their stock and crush sellers / short sellers in the process. You can watch Brian expand on this idea here
. The evidence of this link -- and I will let the lawyers out there decide whether it is direct or circumstantial evidence -- is found in the quarterly Federal Reserve Release Z.1 (Table F.213), which shows that since at least 2007 the only net buyers of stocks have been the companies themselves.
I will repeat this because it is critically important: Since at least 2007 (it actually goes back to the late '90s) all but one type of participant in the equity markets have been doing little more than swapping shares among themselves in a zero sum game of buyers and sellers. The exception is corporations, which have been huge net buyers of their own shares. In other words, whenever stock prices either flatten out or start falling because participants fret over iffy fundamentals, macro risks, dislocations in foreign markets, etc., companies flush with cash (be it from cash flow or debt) step in to break the tie and walk/rip their own stock prices higher.
Let me preempt negative commentary and questions regarding the above. Do I think this is a good way for companies to use their cash or balance sheet? With few exceptions, Apple
(NASDAQ:AAPL) being perhaps the most obvious, the answer is no. Can it go on forever? No. Does it violate most tenets of fundamental investing? Almost certainly. Is it stock "manipulation"? No argument from me on that. But do the stocks ultimately go up? Most of the times yes, and when they don't, sellers and short sellers have a much tougher time pushing them down. Call it a rigged game, or any other disparaging adjective, but the bottom line is that if the stocks do go up, and you are short or not involved, you are going to be worse off than if you were long.
The link between corporate bond issuance and rising stock prices eventually will be broken (with very ugly consequences), but only when companies can no longer fuel their buybacks by selling bonds, and, even worse, when the then-overleveraged corporate bond buyers are turned into forced sellers. (See here
for that happy scenario.)
Now, this causal relationship is not something that transpires in hour-by-hour or day-by-day trading, but rather it is a matter of longer-term money flows from buybacks which ultimately overwhelm marginal sellers. The more immediate correlation that I alluded to at the beginning of the article is more difficult to explain, which is why I view it as a form of "dependency" between the two assets rather than causality. The evidence around it is in part anecdotal and in part guesswork, and it goes something like this.
When large macro funds want to take bearish positions, they gravitate toward the instruments that give them the biggest bang for their buck. From the emergence of the crisis and still today, this consists of buying CDSs on various debt; if the CDS trade gains traction, they then press it by shorting equity index futures.
CDSs remain the preferred tool because i) they are highly leveraged; ii) the buyers of CDSs have defined risk; iii) on a valuation basis they tend to be cheaper than equity or index options; and iv) thanks to the mayhem created by CDSs during the last crisis, when CDS spreads start widening (i.e. they move in a bearish direction) they conjure up the nightmares of a few years ago, which often results in even wider spreads, falling stock prices, and an eventual downward spiral.
When a bear raid fails it is usually because of combination of three things: i) a failure of the CDSs' spreads to continue widening as other bearish players refuse to join the raid; ii) equities not going down thus removing one of the legs of the "downward spiral"; and iii) and most important of all, continued vibrancy in the issuance of corporate bonds. The latter is the dagger in any bear raid because macro bears recognize very quickly that if the very securities they are trying to pressure via the CDSs remain in high demand, it is only a matter of time before their derivatives get crushed.
And when a bear raid fails, CDSs present a major problem for bearish speculators: They are very illiquid, and the securities underlying corporate CDSs, i.e. corporate bonds, are almost as illiquid. That means that when the CDSs start moving against the bears, it is very difficult for them to unwind or hedge out their exposure using those very instruments. So the bears turn to the market that can give them the fastest and most leveraged upside, i.e. the very equity index futures that they were trying to crack.
So is this dynamic what has been causing scary air pockets for stocks over the last five-plus years only to be followed by even more vicious rallies? Is it the repenting macro bears who relentlessly ramp up index futures as everyone wonders aloud, "Why are stocks going up"? I can't offer you clear evidence, but I have heard this pattern recounted to me by enough reliable sources to put enough belief into it. And without question, the footprints left behind by CDS indices, corporate bond data, and how stocks react to it, line up almost precisely with the anecdotes.
To sum up, hopefully this gives a little more color to why I constantly harp on credit and credit derivatives. Right, wrong, or indifferent, by all indications they are a significant, if not primary, reason stocks have behaved the way they have over the last five years, and all things staying equal, they will be the drivers of stock prices for quite a while longer.