Economics
Measured history
200 aligned UTC days, 22 Apr 2026 to 7 Nov 2026, 
USDT leg.
| Statistic | Value |
|---|---|
| Mean | −0.20 pp |
| Standard deviation | 1.13 pp |
| Low | −3.09 pp on 1 Jun 2026 |
| High | +2.43 pp on 8 Oct 2026 |
| Average absolute daily move | 0.40 pp |
| 95th percentile daily move | 1.06 pp |
| Largest daily move | 1.61 pp |
| Days beyond ±1 pp | 47% |
| Sign changes | 12 |
What that means on a ±5 pp band
The full band is 10 pp, so a contract's whole collateral is $100 per long and short pair. An average day moves the index 0.40 pp, which is $4 on a $10-per-point contract. A long at 0 locks $50, so an average day moves its value about 8%. A 95th-percentile day moves it about 21%. From the low of −3.09 pp on 1 Jun to the high of +2.43 pp on 8 Oct, a long bought at the low would have gained $55.20 per contract on $19.10 of margin.
This is a currency-grade product. It moves every day and trends over weeks. The 2024-style +10 pp episodes sit outside this sample; launch series would cap them at the band edge, and a later series can carry a wider band (the program allows up to ±15 pp).
Fees
| Item | Value |
|---|---|
| Trading fee | 10 bps of the pair's full collateral, paid by each side |
| Per contract | $0.10 per side |
| Program maximum | 100 bps (MAX_FEE_BPS) |
| Report and dispute bond | $100 each |
Open interest cap
The cap comes from the cost to bend the index (Security): an attacker who could move S by 1 pp gains open interest × 1 pp, and holding 1 pp for 24 hours costs $0.9M to $2.6M. Requiring the cost to be ten times the gain gives $9M to $26M of notional. Launch series take the conservative end; the cap grows linearly with Korean USDT volume.
Who is on each side
- Long the premium: anyone who expects Korean demand for dollars to rise, and anyone short the premium in real life, for example a desk holding won that it must convert later.
- Short the premium: anyone holding crypto priced in won who wants to lock today's premium before cashing out, and Korean desks carrying USDT inventory.
- Sizes of these groups are not measurable from public data and are not claimed.