Early TUSD & HUSD Real Uses

Here we are going to dig in to some early — 2018 and 2019 — flows of TUSD and HUSD. The goal here is to show, using public data, three things:

1. They were used used largely by the same parties
2. A lot of this use was USD for Chinese entities
3. A long slow organic process of “choking off” access for years

These are not Earth-shattering revelations — but we are proving, rather than asserting, them. We firmly establish that, for example, the TrueUSD was focused on offshore USD flows long before this acquistion. We also present convincing evidence the same USD were pulled from early TUSD to fund the initial mints of HUSD. This, along with other evidence, suggests folks related to Huobi were active users of TUSD.

This is all particularly amusing and circular given that Huobi is now owned by Justin Sun, long one of the largest users of TUSD and sometimes conjectured to be the owner of that coin’s issuer. Quite a large slice of the ecosystem looks to have been the same folks. Or at least the same group servicing end clients.

This also gives us a new perspective on recent efforts to “choke off” crypto. Much of what is happening today looks like the logical next step in a multi-year organic process rather than anything new.

Early TUSD

As discussed elsewhere TUSD changed contracts at the end of 2018/beginning at 2019. The blog post announcing this states that:

The old address continues to work, it just forwards all calls such as transfer() to the new address.

What they mean is that when you try to transfer old tokens it will burn them and re-mint new ones. This gives an interesting dynamic for the market cap of the old and new tokens. We are used to seeing minting and burning out of the 0x0 address. But that is not quite what happened here. Here is a chart of the old and new TUSD market caps on Ethereum measured against the 0x0 address:

We see a negative market cap for the new token — again, with respect to the null address — for quite a while after it launches. This tells us that for much of 2019 and early 2020 money that flowed in to TUSD during 2018 flowed out in this period in a greater quantity than the number of new USD coming in.

To foreshadow a little bit here is the market cap of TUSD (new+old) and HUSD during the first 12 months HUSD existed:

Now lets look at cumulative minting of HUSD vs burning of TUSD during this period:

For 2019, at least, there is a pretty good correspondence. This indicates these might be the same USD moving from place to place. We speculated earlier that minor stablecoins were a way to onboard USD which ultimately end up in USDT. Those numbers match remarkably well. Here we are showing there is a chance something like $250 million went from TUSD into HUSD before that process began. This chart comes from that post and compares USDT market cap to the burning of a collection of minor stables:

Notice the process begins to go parabolic around the same time the HUSD minting above kicks off. It looks very much like, during the period where USDT was not growing, these other stables were used on their own. And something happened in early 2020 such that folks specifically wanted USDT but continued to use their existing on-ramps.

Early TUSD Users

If we extend our earlier analysis to cover TUSD use through 2018 and 2019 we find the following largest minters:

and burners:

Recall those many-0s addresses are per-customer redemption addresses for TUSD. These are, um, not so hard to figure out. Go look here for example. Or here. We want to emphasize there is nothing wrong here. This is just data.

Early HUSD

Here is early HUSD minting:

It starts with $50 million in minting on 19 July 2019 to 0x0e7770b63ca9782b9883455f45282ab2a40df347, a Huobi address. Fine. That is wholly unsurprising. Now compare that to the TUSD burning only for the 525 and 540 redemption addresses:

Look at the dates. First there is a sequence of burns with a large one on the 18th! Second these burns trail off alongside the flat portion of the minting chart through August. And third, look at the 525 address:

These are circular flows through the 0x270cd address. This is some party with real USD bank accounts that just deposits and withdraws. The 540 address comes off of Huobi:

The quantities and timings match perfectly. USD were on Huobi in TUSD form and got pulled to fund the initial mints of HUSD. It’s the same users. Whoever 0x270cd is they are treating TUSD as a piggy bank.

The simplest explanation for these coincidences, alongside the market cap correspondence above, is that USD were moved from TUSD into HUSD. And then to USDT.

The Users

First a few public announcements:

1. This gives TrueUSD’s original banking & trust partners
2. This shows TrueUSD had a relationship with the short-lived Huobi US operation.

Again there is nothing wrong with this. TUSD and HUSD (recall HUSD at this time was administered by Paxos) had proper KYC processes. So now let us look at TrueUSD’s original trust partner in Nevada. Here is their website from the relevant time period:

And here is a snapshot from after this activity stopped:

Notice the third paragraph is missing. Why? Did their posture change? At the same time Huobi was running their own trust in Nevada:

And the company that took over when Huobi and Paxos parted ways also had a Nevada operation with a common director:

And this is equally true for the users. Amber Group is mentioned above and has this entity in Nevada where the directors also match the crypto company:

There are many many others too. Again this is likely all legal. Or at least could pass — and did pass — KYC at one time. But it is no longer active. Interpret that how you will.

Oh and you may be wondering where the freshly minted HUSD went? Huobi. The vast majority (well over $250 million) went to Huobi with Alameda receiving about $50 million and then shrapnel. This is both unsurprising and boring. Remember this is 2019 and all of these HUSD were later burned and the USD look to have ended up inside USDT.

So?

Here is the sequence of events involving one group of folks and essentially one (growing) pile of USD.

1. Start with TUSD, often on-ramping from what look like burner accounts
2. Move to a Paxos-administered HUSD (data above)
3. Feed USD into USDT
4. HUSD breaks away from Paxos, eventually closes
5. TUSD bought by “Asian conglomerate”
6. BUSD has problems, is shut down
7. TUSD moves offshore, starts growing again

This is the offshoring of the system. And it is harder to undertake these kinds of activities in New York than Nevada, or for billions vs millions. Over time, organically, the system was pushed through these steps.

The easiest read is that there is no Operation Choke Point 2.0 and this game of cat and mouse has been going on for years. Simply put: gradually the onshore service providers loose interest in undertaking this business and/or the offshore alternatives got easier. So it is probably a mix of why ever these folks left Nevada as described above and the Deltec/Capital Union Bahamaian ecosystem getting easier to use.

Recall much of US financial regulation is about making the bank/trust/broker/etc responsible for KYC/AML and then punishing them if they make a mistake. As a result service providers review accounts and make decisions to off-board, or discontinue, or reduce all the time. This is how it works. And there is an awful lot of paperwork sitting around behind all of this stuff just waiting to be explored.

A Correction

We previously Tweeted

That Sifu/QuadrigaCX-related address was the 0x008024 listed above. We now have better information and list that as Cumberland OTC. There is some historical interaction with Quadriga addresses and wallet tagging is not perfect. Cumberland should be able to set up USD bank accounts and perform proper KYC. Again there is no implication these activities themselves are somehow out of bounds. Doubly so as they phased out as things pushed further offshore.