Introducing the TAU Protocol

[Note: the short TAU paper is available here]

Outside of my day job (at Clearmatics) I have spent the past couple of months designing a new synthetic asset protocol that uses a rebase technique to stabilize an asset at a target level. I call it the (τ) TAU Protocol.

Frequent readers have probably seen my writings on stablecoins, mining, and DeFi- related topics. For TAU, I started the process in reverse: I knew what I wanted to stabilize but wasn’t quite sure how to get there. What do I mean?

The landscape of synthetic asset-focused projects is something I have discussed multiple times. My most recent pinned tweet was an entire paper on a specific type of stablecoin that relies on exogenous banks to provide utility.

As a thought experiment, what if instead of trying to stabilize around $1 USD, the protocol tries to stabilize an existing cryptocurrency but do so on a separate blockchain… and do so with as little manual intervention as possible. In steps rebasing.

Note: in 2014 several proposals were published on the idea of stabilizing bitcoins price with respect to the U.S. dollar:

My boss (Robert Sams) wrote a paper around the same time called “Seigniorage Shares,” which outlined a way to stabilize a PoW coin using endogenous information. The idea would later spawn a couple dozen (mostly broken) efforts now live on a couple of public chains.

One notable exception to these failures is FRAX (thus far), which uses a reserve fund partially collateralized in USDC. This is an interesting workaround.

One of the problems with rebase protocols is that once the peg declines from the 1.0 target level it can be hard to credibly move it back up: if it goes above the target the process is a bit easier, solved via inflation.

So to recap: the goal is to synthesize an asset (tAsset) and maintain its target value relative to its facsimile on a different blockchain… and to have a credible way of supporting the rebasing process. How would you go about stabilizing a tAsset in practice?

One way is to follow the model of FRAX or other (partially) collateralized stablecoins: with a fund. But setting up a fund of coins that reside on one blockchain to be used on another is hard. For instance, Bitcoin only resides on the Bitcoin network, right?

Actually over the past couple of years there have been ways to tokenize or “wrap” assets from one chain and shuttle them over to another. However, these often involve new trust models (and attack surfaces).

To-date about 1% of all mined bitcoins have been tokenized or wrapped and “transported” over to Ethereum. But to interact with WBTC on Ethereum mainnet can be expensive at times. Back in January I thought: has anyone tokenized hashrate itself onto another chain? Yes it has.

After some googling I came across a paper from Alex Zhao at BTCST. They had figured out how to tokenize hashrate onto a fairly inexpensive EVM chain (BSC). I reached out to explain what I had in mind for a new Protocol idea. And they decided to try and implement it.

Another quick reminder: a Protocol is separate from an implementation. For instance, in Ethereum the “Yellow paper” provides a neutral Protocol specification from which independent parties can build client implementations of. For TAU, I am striving to reuse a similar model.

Today, the initial Protocol idea (and paper) was announced for TAU. Next week the BTCST team will implement it live. Note: while I am currently assisting them as a protocol advisor I want to make it clear that this isn’t my day job and that others can build implementations.

Lastly, I have some additional ideas for how to expand and enhance the Protocol in the future and am keen to see what kind of feedback and modifications the larger cryptocurrency community may have, especially if it includes ways to minimize manual inputs.

Insurance versus “insurance”

Was recently talking to a close friend who has been working on an insurance-focused technology company the past couple of years.

I gave him this list of projects and asked him how he would categorize them:

  1. Nexus Mutual
  2. Cover Protocol
  3. Tidal Finance
  4. Etherisc
  5. Armor Finance
  6. Nayms
  7. Unslashed Finance
  8. Protekt Protocol
  9. Risk Harbor
  10. Soteria Finance

At first glance he thought there were roughly two buckets: protection against loss, theft, and smart contract failure versus DeFi insurance platforms and parametric risks. But then Nayms is a platform and marketplace so what are other nuances?

According to him, a lot of “insurance” above is really just a derivative product so in practice most of these are basically just prediction or options markets: you are betting a contract will not fail and hoping the pseudonymous claims committee rules in your favor.

A counter-argument is that all insurance is like that conceptually. But in reality, insurers try to underwrite the risk which then leads to a pricing exercise, but the prices are grounded first and foremost in risk and then market forces adjust pricing. As opposed to a pricing exercise which seems a bit divorced from the risk but mainly driven by the price action of a coin.

Other categories

  • Parametric risks (such as Etherisc)
  • Centralized Insurance (only a handful of stealth providers)
  • DeFi Insurance (theft, loss, smart contract failure)
  • DeFi “Insurance” (similar to an option: if this thing doesn’t work I pay you monies, but it’s not designed as an insurance product…a subtle but important difference)

What are some other projects and categories to add in the future?

For more context, I highly recommend this thread on “DeFi insurance” from Lucius Fang (who is a trained actuary).

Lastly, I reached out to Stephen Palley, a cryptocurrency-nerd / attorney who specializes in suing insurance companies. According to him:

So the big issue for me — and I am planning on doing a long form piece on this — is that people who sell insurance are subject to a maze of state level regulatory and licensing requirements that they so far have seemed happy to ignore. If/when/as this gets bigger, people will go to prison.

It’s a huge opportunity for people who want to do things the right way. But as is typical, you have a lot of people who are jumping in who don’t actually know eff all about the foundations they are building on.

Categorizing the derivatives and perpetuals landscape

Over the past couple of years there has been a lot of activity not just in DeFi but in the evolution of on-and-off chain platforms for trading derivatives and perpetual contracts.

Below is a non-exhaustive table that attempts to segment and differentiate who some of the major players known. It is a work in progress and likely is missing some parts. For instance, of those listed: Synthetix and uSTONKS are the only ones that track indices and Mirror (built on Terra) attempts to track real world assets. A notable company excluded from the list: FTX (a large CEX) trades tokenizes equities and indices.

ProductOracleExchangeContract
Own Layer I
Perpetual Protocol (xDai Chain)ChainLinkDEXPerpetuals
TerraBAND/OwnDEXSynthetics
Ethereum Layer I / II
SynthetixChainLinkDEXSynthetics
MCDEXChainLinkDEXPerpetuals
dYdXChainLinkDEXPerpetuals
FutureSwapChainLinkDEXPerpetuals
UMA (uSTONKS)Self-referDEXTWAP
Centralised
BITFOREXN/ACEXPerpetuals
HBDMN/ACEXPerpetuals
BinanceN/ACEXPerpetuals
BBX (sub of OKEx)N/ACEXPerpetuals
Huobi FuturesN/ACEXPerpetuals

What is missing, what should be added, what nuances should be made?

Recent activities

Below are list of interviews, presentations, panels and other public facing engagements I have been involved with the past couple of years.  Taking care of a newborn (now toddler) during a pandemic has dampened some of the external engagement relative to prior years.

Quotes and interviews

Events, presentations, and interviews

Cited