How JPM's "Hedge" Blew Up In One Easy Chart

Tyler Durden's picture

It seems every critical-to-stay-relevant talking head and blogger is trying to make sense of, and gain as much airtime discussing, how JPMorgan's CIO unit could have been so 'stupid'. The answer is - they weren't. As we described first here and here - and has now been accepted by the mainstream media as fact (of course we are flattered by the mimicry) - the reason that the hedge got out of control was the massive amount of delta-hedging that Iksil had to do to manage the position as the Fed and ECB crushed the systemic risk out of the system and blew up the correlation assumptions in his models. This is complex to explain but, by way of example, we show a chart of the implied delta of a proxy for the JPM hedge. The lower the delta, the more and more index protection that needs to be sold to maintain a stable hedge - and as is clear, not only did the delta collapse (almost halving in 4 months) but it reached pre-crisis levels which would have been generally unthinkable in the risk scenarios - given the backdrop of reality. Whether Iksil arrogantly enjoyed ignored the cornering of the IG9 index market and the momentum and P&L he was relishing in is a different matter but to comprehend the forced selling protection pressure he was under, this chart is all you need to understand...

Source: Morgan Stanley

Example - if the tranche notional was $100bn (we suspect it was more given that it was a tail-risk hedge for JPM's aggregate book), then the shift from Q4 2011 to end Q1 2012 would have forced Iksil to sell over $30bn IG9 protection alone (as the delta dropped from over 0.7x to under 0.4x). The trend was so strong that there is little wonder that day after day we saw the IG9 skew (the spread between the technically-rich-and-under-pressure index and the trading-on-fundamentals intrinsics or single-names) widen massively - even in the face of an almost ubiquitous (no it wasn't just Weinstein) knowledge of the discrepancy and attempts to arb the difference.

While greed and avarice were likely strong drivers that blind-folded Iksil and his colleagues to the fact that they were getting too big and that the position was out of control, it remains fact that it was a combination of the Fed/ECB's actions to squelch systemic risk entirely and a total reliance on models that is now bleeding to every other credit index and just as we said - leading to increased losses at JPM.

What is more worrying for them now - is that with systemic risk re-appearing, deltas are rising AND that means that the models will be screaming to unwind the sold protection positions even more aggressively. So as the market rallied they bought more and more and with the market selling off, they are forced to sell more and more - not a fun position to be in - and we suspect unwinding the original tranche deal will be practically impossible, given a lack of liquidity in tranche land in the last week.