Just Say No to (Most of) the Current ICOs

I’m a big believer in blockchains, and the sale of tokens to encourage and enable new protocols. But I’ve never been as dispirited about their prospects as I have been this week after the latest glut of greedy, unethical and/or ill-conceived ICOs that slowed the Ethereum blockchain to a standstill. Let’s diagnose the problems and prescribe the discipline necessary to get us out of the current broken state where scammers can thrive and no protections exist.

I’m hardly the first to point out the danger in how ICOs are being issued. Nick Tomaino sounded alarms way back in September (which we now think of as the good old “sane” days) and gave 10 excellent checkmarks to monitor. This week, Fred Wilson was even more blunt: http://avc.com/2017/06/buyer-beware/

Fred’s points were these: 1) The amounts raised matter, a lot. 2) Understand what the token does and the size of the opportunity. 3) Valuation matters. 4) Avoid scams and things that feel like scams. 5) Look for projects where the technology is well specified and is working. 6) Don’t be greedy.

These are all solid warnings, and deserve further discussion.

As an aside, I come to Ethereum and the blockchain from a Wall Street trading and institutional investing background. To show my bias, here’s an article I wrote in January 2014 talking about 4 hurdles the blockchain needed to clear before Wall St would adopt it; regulatory clarity; credibility from reputable backers on both buy and sell side; “DVP”, or delivery vs. payment (doable via smart contracts); and clear identity (doable via ENS domains). Redditors hated it, but Wall Street has been following that path. It’s been a long time coming, but finally Ethereum and its offshoots, like the Hyperledger project, are now becoming legitimized for their importance in clearing and settling securities. Like before, I see 4 necessary best practices ICOs need to adopt, most of which is hinted at in Fred’s manifesto.

  1. Clearly enumerated fund raise limits. For truly hot offerings, Wall St has adopted a “green shoe” policy, allowing a known amount, generally 10% or less additional issuance, to help ameliorate unfilled demand, but that’s it. When the green shoe runs out, it’s the end of the sale. This week we saw massive shenanigans and extended offerings contra to earlier promises. I consider Bancor’s “hidden cap” highly unethical, and those who got caught up in the hype have only themselves to blame. That goes to Fred’s first and third points. And especially the sixth, regarding the issuer. Time will tell, but I’d short the BNT token in a heartbeat if I could. Karma is a bitch.
  2. Clear allocation processes that do not favor high frequency trading techniques, and which are not time sensitive. Many people blamed Gnosis’s huge valuation on the mechanics of a Dutch auction, but the process is not to be blamed — that’s the exact method used by the Fed to issue Treasury securities and it works exceedingly well. The blame lies probably more with greedy but lazy bidders not noting the terms of the trade. I’d recommend sealed bid Dutch auctions to establish a fair market clearing price, but done over time so that a flood of bids doesn’t clog the blockchain and take down the exchanges as we saw this week. Perhaps a minimum of 1–2 weeks to put in your bids with escrow, with the pre-defined process then verifiably followed with no prejudice as to time. It’s like elections — if there wasn’t absentee or mail-in votes available weeks before, we’d all be standing in line forever at the polls. Time should NEVER be a factor. I say the fairest version is price — If someone wants to pay higher prices, they deserve the tokens…not just the investors with the slickest programming and dirtiest trading tricks. (For more on HFT and other dirty tricks, read Michael Lewis’ Flash Boys. We need to establish standards that keep allocation rules fair.)
  3. Independent Review Udi Wertheimer did a masterful post-mortem on Bancor’s offering which clearly points to duplicitous behavior: saying one thing on your white papers and marketing material, but hiding a different truth in the code. But why did this article have to be a POST-mortem? We can’t legally prohibit someone from selling a token, but we can give a Good Housekeeping-type Seal of Approval on an independent review done by disinterested parties, focusing on code and comparing terms in a standardized fashion with reference to other historic deals. Say what you will about Moody’s or S&P and the 2008 MBS fiasco (believe me, I was deeply involved and plenty of expensive lessons were learned), but regardless of those rating agencies’ flaws, they have saved many, many investors from buying crap. There is a role here for the various foundations here, or some party that is truly disinterested, with a reputation at stake who works for a set fee. How about a team headed by William Mougayar and Emin Gun Sirer, for instance? To get blessed with a seal of approval, an independent review (noting robustness and completeness of code, depth of team, usage by customers, use of proceeds) should be in hand at least 1 week before the multi-day bid submissions start. Cost of the review would be included and paid for out of proceeds from the ICO, just as legal fees are. The anarchist wing will be alarmed, but it’s the price of growing up.
  4. Standard term sheets Really — this is simple. Promulgate simple, standard terms, and show where they deviate from recommended ranges. From the terrific podcast done yesterday with William Mougayar, http://pca.st/V8Ir, it seems like something is in the works, at least for Ethereum. You can spin a white paper or sales memorandum, but you can’t make the numbers lie if you have rock-hard definitions.

The bright side is that from these many recent investor mistakes and shoddy issuer practices will come enough pain to forge better practices ahead. A shoutout to Civic for handling their ICO with class in the face of the blockchain shutdown caused in large part, as far as I can tell, to a poorly designed Status ICO.

But dozens if not hundreds more ICOs are coming, and soon, and not all will be as well designed and open as Civic. As Fred Wilson said, Caveat Emptor. Just say no to any token sales which aren’t EXPLICITLY clear in their offering.