Three months ago, Referendum #335 funded a controlled test of protocol-owned HDX liquidity on Kraken. The test is complete and the full data review was published and discussed here: 3-Month Review & Continuation Discussion. The feedback favored continuing with a stable-side top-up - this referendum enacts exactly that (Option C from the discussion).
What the test delivered (May 22 - Aug 11, all data in the review):
- HDX/USD spread: from ~1-2% to a 1-3 bps median (0.012% on CoinGecko today)
- Two-sided quotes live 99.7% of hours - 100% over the last 28 days
- ±2% order-book depth roughly doubled
- Treasury inventory fully intact and grown: ≈ +$5k vs. simply holding at the review cutoff, ≈ +$6.5k as of Aug 17 - net of all fees
- No funds lost; two attempts to exploit the operation failed; every incident led to a fix (289 commits)
- Net result for the treasury after the operator fee: slightly profitable - ≈ +$3k as of Aug 17
This referendum transfers:
- 20,000 HOLLAR to the LP operations account - restoring the HDX/stable inventory balance that the HDX rally (+~100% since launch) has shifted to roughly 70/30. Deeper stable-side inventory funds deeper bids during sell-offs (fewer and shallower wicks), reduces the book's directional HDX weight and enables a properly quoted HDX/EUR.
- 7,200 HOLLAR to the operator - operator fee: 1,200/month × 6 months, paid upfront - same rate and account as ref #335.
The call mirrors ref #335: Dispatcher.dispatch_as_treasury → Utility.batch_all of two HOLLAR transfers (wrapped in dispatch_with_extra_gas, as HOLLAR is ERC20-backed).
All inventory (~8.8M HDX, stables, and accrued P&L) remains treasury-owned and recallable by referendum at any time.
Honest expectations, unchanged from the review: protocol-owned market making pays for tight spreads through inventory risk - some quarters may show a negative P&L. The test quarter ended profitable.