HYG notably outperformed this afternoon as it disconnected dramatically from the rest of the risk and credit complex.

We often discuss how credit markets have provided useful insights (and potential pre-emptive indications) with regard to risk appetite and whether ES should rip and today’s incredible rally in HYG (the high yield credit bond ETF) is one to be aware (beware) of. The rumble of liquidity-driven hedging being unwound was very loud indeed and as spreads reach significant levels on a medium-term basis and HYG recovers its major drop in price, we wonder if the market is now less prepared to handle any downside shock. HY (and IG) bonds have been rising in terms of advance-decline (sell-side to buy-side) and are above long-run average, HY relative-value to equities has compressed at the index-comparison level, but HY cash bond liquidity remains light at best and indices have done little except catch up to intrinsics as technicals (flows) leave demand for primary and secondary HY less than stellar despite what you see in HYG or JNK. It may be fun to chase momentum (especially when they offer a decent carry) but we suspect, given below-the-index-level analysis, that risk appetite is just not there and as usual the removal of hedges causing the most pain to the most traders.

Evidence from today’s HYG movement relative to HY credit says this was not a concerted risk-on move, but much more likely a mass hedge unwind in what was the HY market’s cheapest and most liquid hedge (in anything other than huge size).

Comparing HYG price to its Shares Oustanding offers insights into the under-the-surface arbitrages and unit creation.

When a manager of a large portfolio of high-yielding, carry-generating, upward-curve sliding flotsam and jetsam (HY bonds) fears his entire month’s carry is about to up in smoke in the next 15mins, he will reach for a bag of his favorite hedging instrument. The favoritism will be biased to how-well-it-worked-last-time, how-cheap-it-is, and how-liquid-it-is with the latter becoming a higher priority as the sell-off accelerates.

We noted the large slides around mid June 2011, early August 2011, and early October 2011 all of which had similar major underperformances of their peers (beta-adjusted), similar magnitudes but the last two major downdrafts saw a seemingly paradoxical rise in shares outstanding. The orange rectangle shows a rather significant drop in price and yet Shares Outstanding continue to rise. Fund flows have been volatile but more positive than negative which obviously drove some demand for creation units but the absolute inversion of the shares outstanding/redemption relative to the share price movement makes little sense…unless funds were able to approach the ETF managers and have them force-buy in the illiquid cash HY market (or direct from the approacher themselves) as manufactured Creation Units. note: Another example of why ETFs are more complex than most want to know.

Our SPY Arbitrage model indicates just how much outperformance HYG created today over SPY when adjusted for empirical beta.

HY bonds have indeed been bought, though its tough to find a client, friend, colleague who has been aggressive buyers at the single-name level. HY Advance-Decline just shifted above its medium-term average perhaps suggesting cyclically that buy-side buying is peaking. Risk appetite definitley remains low (which sounds odd on a day such as this) but concessions in HY new issues are still incredibly high – SONOCO 10Y recently same at over $5 cheap to market fair-value in order to encourage buyers/lenders…so why do we see mad scramble buy-buy-buy days in HYG like today when risk appetite is low…the answer is hedge/short-covering (among other things agreed).

That is how HYG managed to outperform SPY by an immense 350bps today (apples-to-apples beta/vol adjusted) see chart below.

The performance of the black line is adjusted based on HYG and its empirical performance characteristics with several credit related assets and in sync with SPY. Typically they will be tick-for-tick – today is massively unusual.

It appears quite clear that the huge HYG-trickle-down-hedges that were put on over the last few weeks/months were rapidly unwound today. How do those hedges play out? (in brief)…

    Step 1a, professionals use HYG (as the most liquid at the time) to hedge their long HY bond positions (note that with rates so low and spreads so wide that IR duration hedging is a hugely secondary consideration). They will short-sell the ETF (at a borrow AND carry cost). This will cause a dislocation between the value of the ETF and its NAV and then that will trickle down to the underlying components making up the NAV.

This chart shows the considerable underperformance of the HYG ETF compared to the performance of the HY credit index in late Sept/early Oct. Liquidity was clearly preferred and HYG was offering more – no matter what the price – to a point.

Step 1b, Professionals use HY (the liquid CDS index) to hedge but at times it can be very expensive, illiquid, and/or technically complex. If there is too much demand for the index protection (hedging/shorts) then there is no direct reason why demand for underlying names will also rise (not like in stocks where real-time tick-for-tick arbitrages keep indices and underlyings tightly in sync).

So we see the value of the index fall (in the case of hedging a long portfolio with an index short) much more quickly than the value implied by the underlyings. The index becomes more and more cheap relative to its underlying ‘value’ and at some point buyers will return as the index has become just too cheap, or hedgers will hedge less expensively.

The most recent on-the-run index is in green and has only traded for a few weeks but note how the index price (dotted green) has remains below the intrinsics (solid green) for most of the upswing and for the previous index (in purple) the same starting back in late August when the index was clearly offered as its demand as a macro overlay/hedge remained high.

Step 1c, professionals will try and buy individual  (CDS) protection on those names they fear the most and ride the wave. Given the potential richness of the underlying single-names when a market switches to index-based hedging, many may choose to shift specific risks at the single-name level as a situation develops. As the technical flows from demand for single-name protection hit they will push the cost of protection up notably above the credit risk premium of the underlying bonds (suggesting basis trade or to sell yr relatively over-priced bonds).

Step 1d, professionals will sell their longs (sell bonds). We haven’t seen dramatic HY bond selling days recently – we suspect it is largely due to the Nash Equilibrium that all traders find themselves when faced with an illiquid market and a mission to cover. Everyone knows that they can’t offer enough of their inventory without causing a broader sell-off which will drag down asset values across the rest of the book. This step does tend to be the last and most desperate (or the first and most simple if you are really good at timing).

Step 2 – Wait. Hold your breath as all the basis (drivers of differences between hedge MtMs and portfolio MtMs) we discussed above (and likely others) provide daily risk flares across your book.

Step 2a, Profit/Loss. With any luck, the hedge may have a higher delta/beta (or even gamma) than the underlying portfolio and make you money even as you are hedged (or in a worse case the cost of protection outweighs the carry of the book).

Step 3 – Cover. At some point a manager will become comfortable enough with the outlook to remove hedges and revert back to his long duration strategy. The other side is of course that at some point the use of a hedge becomes irrelevant as a manager becomes convinced given perhaps binary outlooks and scenario analysis that is incapable of providing any useful mean for valuation. In the latter case, the hedge will be lifted and so will the longs.

Done well, this can be performed in a step-wise fashion – unwinding short index positions into an up-trend will exaggerate the movement and encourage more buyers of HY in general as the momentum jockeys jump on. Then you simply turn around and sell your single-name positions into the buying herd (as in today).

So the bottom line from all of this: HYG outperformed SPY dramatically (suggesting forced covering or unwinds as opposed to rotation), HYG outperformed HY dramatically (suggesting a more flow focused hedge unwind), HY outperformed intrinsics dramatically (HY -35bps vs intrinsics -5bps), HY term structure flattened with the short-end actually broadly wider today!!, and while TRACE shows notable net-buying today it was very much butterfly-based as short- and long-end bonds were net-sold as 3-12Y were most net-bought.

 

Much of today’s action in the high yield credit market seemed as much about catch up to each segment’s relative-value as any real aggressive buying as hedges were clearly unwound. We would warn traders who use index aggregates to judge relative asset allocations between credit and equity to be very careful with this shift as bottom-up, it does not exist yet.

UPDATE: HYG’s premium to Net Asset Value (NAV) is its highest since May09!!

HYG's premium to NAV is its highest since MAY09.

Leave a reply