A reserve held in ETH loses a third of its value inside one quarterly reporting period
Should the treasury reduce its ETH reserve before the quarter closes, or keep the policy allocation and accept another quarter of mark-to-market losses?
The market facts are real; the decision maker is illustrative.
The week and the position
In the week of 3 September 2018, ETH fell 33.74%, from 295.51 to 195.81, including a 19.86% fall on 5 September, its worst day of the week. The return sits at the 0.0 percentile of 29 weeks of history, and ETH is 83.11% below its 52-week high. Volume ran 1.28 times the prior 12-week average, and the close landed in the lower third of the week range. ETH also lost 22.6% against BTC, underperforming by 19.34 percentage points. An illustrative treasury holds ETH as a reserve and reports quarterly, with three of the prior four weeks already down.
Judgment
A quarterly reporter cannot wait for a policy review that falls outside this drawdown. The week's loss is an extreme tail event for this asset's 29 weeks of history, and the 0.84 hourly correlation with BTC means the fall was partly market-wide, yet ETH still lost 19.34 percentage points more than BTC, so a large part is asset-specific. Reduce the ETH reserve to a size whose further decline the treasury can absorb within one reporting period, and shift the remainder to the BTC portion of the reserve. Accept selling near the low; the accepted loss is smaller than the uncertainty of holding through another week at 125.3% realised volatility.
What would change it
If ETH's weekly return stabilises and its spread to BTC narrows for several consecutive weeks, the reserve can be rebuilt to its prior policy size at the next quarterly close. If the treasury's own spending needs require no ETH, the reduction can be complete.
What sets this case apart
This is a reserve-allocation decision driven by a reporting calendar colliding with an extreme-tail week, not a trade on price direction.