Why couldn’t outside investors arbitrage away the Treasury mispricings during the LTCM crisis?

External question — Quantitative Finance Stack Exchange Author: bravesirrobin Original post: https://quant.stackexchange.com/questions/85848 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. I’m trying to understand the mechanism behind the claim that LTCM’s forced liquidation in 1998 could have caused severe market dislocations. Consider a simplified LTCM convergence trade: Long an off-the-run Treasury. Short a similar on-the-run Treasury. Duration/rate exposure is approximately hedged. The spread between them widens substantially during the crisis. The Fed was afraid of a market catastrophe, and put pressure on banks to save LTCM. My question is: why wouldn’t outside investors arbitrage this away? For example, suppose I already own $1M of Treasuries. If brokers/dealers allow Treasury collateral to support a long/short Treasury position with, say, a few percent margin, I (as a regular citizen) can: Post some of my Treasury holdings as collateral. Buy the cheap Treasury. Short the expensive Treasury. Continue earning approximately the risk-free return on most of my collateral. Earn the convergence spread if the relative pricing normalizes. Transaction costs on institutional Treasury trading seem much too small to explain a 1%+ relative price discrepancy. I understand several possible objections, but I’m not sure which one quantitatively explains the 1998 situation: The on-the-run Treasury may have been hard to borrow. This doesn't make much sense. LTCM sold it to someone to get to the short position. If the gap widens, that someone now has an intensive to sell. The on-the-run Treasuries actually were worth more. Why would this happen? I don't undersand why the liquidity premium will be large than the bid/offer spread , which is a few bps at most. What I’m looking for is a quantitative explanation . I’m not asking why LTCM itself could not survive the mark-to-market loss. I’m asking why new, unleveraged capital could not step in and arbitrage the dislocation once LTCM became a forced seller. The Fed put pressure on banks to save LTCM. Why? Why does it matter if spreads of 2 treasuries increase by a 1%? And how can this happen, when such a spread gives large intensives for everyone to close it. I feel like there is something in the mechanism i don't understand.

External answer — Quantitative Finance Stack Exchange Author: dm63 Original post: https://quant.stackexchange.com/a/85857 License: CC BY-SA 4.0 — https://creativecommons.org/licenses/by-sa/4.0/ Adaptation: HTML converted to plain text; contact email addresses removed. I remember this situation, being a trader at one of the large banks facing LTCM. The trade of most concern, as mentioned by @Dimitri Vulis, was the ‘short swap spread trade’, meaning they were short Treasuries versus receiving fixed/paying Libor on a swap. This spread started moving against them for a variety of macroeconomic reasons including the Russia default which led to a flight to quality into Treasuries. The big problem for the banks is that we were facing LTCM on swaps which were collateralized using Daily Cash flows. If a HF that large defaults, the daily margin payments cease. When that happens the swaps are cancelled and have to be replaced in the market , which would inevitably lead to large losses for the banks. So it was in the banks’ interest to instead design an orderly liquidation procedure. I might add that the regulators supported the idea with the objective of orderly swaps and Treasury markets.

Why couldn’t outside investors arbitrage away the Treasury mispricings during the LTCM crisis?

Why couldn’t outside investors arbitrage away the Treasury mispricings during the LTCM crisis?

Post Reply