A 15.98% Sunday inside a 42.81% weekly range forces a margin reset on leveraged ETH
Should the exchange raise initial and maintenance margin on leveraged ETH positions now, or keep current levels and rely on the liquidation engine to clear clients through moves like Sunday's 15.98% day?
The market facts are real; the decision maker is illustrative.
The week and the position
In the week of 8 July 2019, ETH fell 25.88%, from 304.72 to 225.85, with a range of 42.81% and annualised realised volatility of 108.3%. The worst day was Sunday 14 July at -15.98%, including an -8.61% hour at 09:00 UTC that day. The close sat in the lower third of the week range. Volume was 0.72bn USD, 0.92 times its prior 12-week average. ETH underperformed BTC by 15.05 percentage points and the ETH/BTC ratio fell 16.88%, while correlation was close at 0.71. Three of the four prior weeks were up.
Judgment
Set margin from the moves the client must survive intraday, not from the weekly average. Raise initial and maintenance margin on ETH relative to BTC, using the -8.61% hour and the -15.98% day as the sizing basis, and apply the new levels to new and increased positions. Leave existing positions at their original levels unless coverage falls into a tighter band, then require the difference. The below-average weekly volume makes the case for a full retroactive reset weak: thin weeks can overshoot, and forcing existing clients to post margin into a falling book converts a price loss into defaults.
What would change it
If the liquidation engine reports fills close to trigger prices through Sunday's hour, the case for a large increase weakens and levels can be reviewed. If ETH/BTC stabilises and weekly realised volatility returns to the prior four weeks' range for several consecutive weeks, the margin premium can be stepped down.
What sets this case apart
The decision turns on ETH's 15.05-point underperformance against BTC and on Sunday's -8.61% hour arriving in a week whose volume was 0.92 times its 12-week average.