Your ICO Should Accept Tokens
This essay went out to tokenreport.com subscribers earlier this week.
Vitalik Buterin, co-founder of the Ethereum virtual currency, posted a lengthy article last Friday, addressing some of the perceived problems with token sales. He addressed the astronomical valuations, perceptions of founder greed and the inaccessibility of token sales that end in a matter of minutes or seconds. You should read it, if you haven’t already.*
As part of the solution, Vitalik addressed a question every token issuer faces: which currencies to accept. He admonished the developers of Ethereum-based projects not to accept other tokens based on the ERC20 standard, in exchange for their new token, suggesting instead they should only accept ether itself. He’s wrong. ERC20 developers should probably accept one anothers’ tokens. Here’s why.
The most salient point in Vitalik’s article is that exchange rates of ERC20 tokens fluctuate, creating pricing problems in any ICO that lasts more than a couple days.
Smaller token sales that take a while to close can run into real price movement problems. Bitcoin prices can appreciate rapidly. If a token price is pegged to Bitcoin at the beginning of a sale that lasts a month, later tokens can become much more expensive, chilling buyer interest.
For fast-moving token sales, this is less of a problem: Prices can be fixed shortly before launch of the token creation. Nevertheless, Vitalik prefers that people on the Ethereum blockchain accept only ether for their token creation event.
The roots of Ethereum are in providing an underlying support layer for many novel digital currencies. I’ve long believed that this “Alibaba” style approach wins the market by avoiding rent-seeking behavior. Essays like Vitalik’s worry me that this Ethereum value is changing, which would be disastrous for Ethereum.
The holder of an ERC20 token does not typically think of that token as anything other than an asset that can be divided up and used for a variety of things on the blockchain. It’s right for these token holders to be able to compare the value of the tokens they hold against the value of new tokens that are emerging and trade freely based on their conclusions.
In such a trade, converting tokens into Ethereum first is expensive; for some of the tokens listed today at Bittrex or Poloniex, a trader could pay as much as 3 to 4 percent in slippage from the quoted price and fees in order to make the trade for ether. And of course, each of those trades pushes up the price of ether.
There isn’t any utility given to the ERC20 holders in exchange for this fee. It’s 100% rent seeking. I am unclear why a holder would want to do this, and in practice so far, they don’t: They like and take advantage of token sale mechanics that allow them to directly contribute.
Let’s Not Mansplain
On the basis of the fee argument alone, there’s no reason to push hard to keep ERC20 contributions from happening. Beyond that, there’s a sort of arrogance at telling two sides of a market how they should want to interact.
Vitalik calls out TokenCard (TKN) and WeTrust (TRST) as “uncontroversial” counterarguments to accepting ERC20 tokens. In fact, TokenCard had a great reason to take those tokens: They wanted to build up enough balance to be able to market-make trades while fulfilling debit card swipes denominated in ERC20. (Disclosure: TokenCard is a client of New Alchemy, where Peter Vessenes is managing director.)
Vitalik is arguing that TokenCard should have had to get in and out of ETH in order to do this business, but TokenCard launch participants disagreed. Millions of dollars were contributed in ERC20.
ERC20 token holders asked the question, “How much of my portfolio should be rebalanced into these new tokens?” I.e., “Would I sell some of the tokens I hold for the new token, and at what price?”
A token issuer can view another project’s token in the same way. In TokenCard’s case, a few tokens, in particular MakerDao’s MKR, were highly desired by the developers. TokenCard offered a 5 percent premium to the MKR market price at the time of sale, but almost none were contributed.
MKR holders said clearly, “We don’t like the exchange rate between MKR and TKN,” despite the 5 percent bonus. In hindsight, it seems TokenCard should have priced their MKR contributions at an exchange rate high enough to make them indifferent to receiving the MKR. It likely would have been much higher than the quoted price.
Vitalik notes that a token can get pumped during a token sale. He provides Maidsafe (MAID) as an example. Maidsafe issuers accepted Bitcoin (BTC) and Mastercoin (MSC), an altcoin based on the Bitcoin blockchain. The issuance set up a spike in Mastercoin’s price and trading volume. This can happen. A high price will bring out sellers. If the offered price is inflated, it’s natural to see a market correction.
However, the higher price may be based on sound business reasons, in which case it’s quite possible that the market on the “dump” side is ignoring some information that perhaps should affect pricing later. Smart holders take the “correction” as the time to buy. Mastercoin’s spike was nothing more than a case of bad pricing.
It would be great if a smart contract could value a contribution of any sort in real time. Right now the Ethereum Blockchain isn’t fast or cheap enough to do this.
In the interim, saying, “These price problems just go away if only you denominate everything in ether,” is self serving. It applies the same ‘pump’ to the ETH price at the expense of ERC20 holders.
Even with real-time quoting, there would be work to do: Liquidity can be very thin for some desirable tokens. We have technology that will credit token contributions at real-time prices, but it is still very new. It’s not clear what the best practices even should be: Spot pricing? Volume weighted average price (VWAP)? Over what period?
It’s likely this part of the market will continue to innovate into a few stable approaches over the next eight months. Until then, look for some interesting ideas as token issuers try and put their best foot forward with the market while raising capital.
Conclusion: The Long Run
In the long run, prices of tokens will not be driven by so-called “pump and dumps.” Instead, prices will approximate the underlying value. This value will include the imputed value of any ability they have to engage in token sales directly.
In the interim, allowing market participants to put their own price on which tokens to acquire and how to acquire them is sane. The alternative is not.
*This article was updated with a working link; it was originally published with a dead link.