Game Theory in Token Valuation and ICOs: Overcoming Adverse Selection

by Brian Tinsman

Traditional Investment Valuation vs Token Valuation

If you’re raising money with an ICO, one of the big challenges is figuring out what price to charge for your token. If you’re an investor, your decision to buy will hinge on whether you think a token’s price is higher or lower than it should be.

Equities and other traditional investments have well-established valuation methods, but blockchain token valuation methods are not yet developed. Token valuation is a rapidly growing and multi-faceted field, but the only consensus so far is that it requires some unknown combination of traditional asset valuation methods and game theory. But wait, it gets worse. Where traditional securities generally have the same kind of purpose and structure, each token may have a unique purpose, making it impossible to define one method of analysis that works for every token. We may need to develop many methods of token valuation and figure out the right combination of methods to value each unique token feature set.

But for now, let’s focus on the game theory aspect. As you may know, game theory is a branch of economics that studies how incentives motivate people to make decisions within specific rule systems. Game theory is a critical part of the design of blockchain and other DLT networks. For example, all blockchain systems are designed to provide incentives to reward those who support the network (honest miners) and punish those who try to cheat.

Game theory-based design has a huge impact on many other parts of decentralized networks as well. In this article I’m going to look at an aspect of game theory called adverse selection and how it influences token valuation.

One Problem with Token Valuation: Adverse Selection

Adverse selection arises in a marketplace when either the buyer or seller knows something the other one doesn’t. This asymmetric information degrades the entire marketplace and leads to low quality goods, low profits, and unhappy buyers or sellers.

One familiar example is when a buyer is shopping at a used car lot. The seller knows which cars are junk in need of expensive repairs, but the buyer doesn’t. In this case the seller has a financial incentive to unload the junk cars for the full price of a good car. If successful, this would bring the seller maximum profit.

The buyer is aware that there’s a risk of buying a junker, and is therefore not willing to pay the full price of a good car. Instead, his buying price will be low - probably in between the fair price of a junker and the fair price of a good car.

This means if the seller actually has a good car, the buyer still won’t be willing to pay the fair price for it, so it will sit unsold until it’s taken off the market. Eventually the only cars offered will be junkers, since those are the only way the seller can make a sale at or above a fair price.

Multiple problems are caused by asymmetric information! Even though good cars are available, they aren’t being sold. Buyers are nervous, as they should be, since sellers are rewarded for dishonesty and can only survive by tricking them into buying junk.

In the token marketplace, a similar situation arises between a token issuer and most buyers.

The token issuer typically has:

  • Inside information on the organization

  • Insight on whether the software will deliver on its lofty promises

  • An evaluation of the competence of leadership

  • More expertise in the industry and knowledge of competitors

With all of this knowledge, the token issuer knows whether the token is good or junk to some extent. Just like the car seller, the issuer has an incentive to hype up the token and sell it at a high price.

The marketplace can follow almost exactly the same path toward rewarding sellers for dishonesty and creating a feedback loop resulting in low-priced, low-quality tokens. Everyone is worse off than they would be if fair prices for both high and low quality tokens were ensured.

A Solution: Game Theory in Token Valuation

How does game theory solve these problems? The best solutions introduce new rules that help align the seller’s incentives with the buyer’s incentives. Here are three examples of new rules:

1. Guarantees

The car seller can offer a warranty, lease, or free returns within 90 days. The seller is happy to sell a good car with this new rule and unhappy to sell a junker with this rule. The buyer is more willing to pay full price for any car under this rule since it protects him from expensive repairs.

In the token marketplace the seller can offer guarantees of token burns and sales restrictions or lockups. For example, Ripple announced that the tokens in their company reserves would be locked and not able to enter circulation according to a multi-year release schedule.

A seller can also offer a guaranteed future service at a known valuation. For example, if a buyer buys the token today for $0.25, he will be able to redeem it for $1.00 worth of banking services in one year. If he doesn’t want those banking services he can sell it to someone who does at a slight discount, effectively pegging the token to a known value.

2. Reputation-Building

The car seller can sell good cars for less than fair value in order to build up a reputation for good quality. After some time, buyers will accept a new rule - that the seller’s reputation acts as collateral against the risk of selling junkers. After all, the seller gave up (invested) a lot of potential profits to build that reputation.

Reputation-building only works when buyers and sellers are in ongoing contact with each other. Tourist traps have an incentive to offer expensive low-quality goods since they get a completely different group of customers every weekend.

Token offerings can attempt to leverage the reputations of prominent advisors or industry leaders. Sellers also try to form alliances with big-name companies in order to piggyback on their reputations. However, if the endorser is extremely rich or famous, it can actually make the endorsement less valuable since that person may not care about a dent in their reputation. On the other hand, if buyers feel that the advisor could be ruined professionally by endorsing a junk token, they will place more value on that endorsement.

3. Eliminating Asymmetric Information

The car buyer can take the car to the shop and get a third party inspection to verify the car’s quality. This costs time and money, so only highly motivated buyers will do it.

In token markets this is equivalent to research. It’s probably not possible for a buyer to get the same inside knowledge as the token issuer, but a little knowledge can still help. It’s also possible for the buyer to use the services of a research provider, although there are also some adverse selection issues there. Is the researcher recommending the token because it’s good, or because he has an incentive to shill it?

It’s interesting to note that the SEC uses this as a major component of keeping equities markets healthy, and it’s one reason why they have been troubled by ICOs. The point of many securities regulations is to make sure the security issuer is telling the investor everything needed to avoid these asymmetric information problems. In fact, the SEC’s standards of information disclosure are different for different kinds of securities, balancing the need to inform investors with the risk of disclosing too much information to the point of harming the business. Startups and investors sometimes see the SEC as a pain, but it’s helpful to understand why they created all these regulations around disclosures. They are using a direct application of game theory.

“Do your own research.” is an often-spoken but seldom-followed refrain among investors. The token seller can make this process easier by providing lots of transparency into company decision-making, developer interviews, speaking engagements, blog posts, and so forth. This should have a real effect on buyer confidence and therefore the token price.

There is plenty more to discuss about adverse selection and other aspects of game theory in the token marketplace, but I’ll pick it up again in another article.

In the meantime, if you are a token seller these solutions can help push up the value of your offering by giving buyers confidence that they won’t get a junker. If you’re a buyer, you should weed out the tokens that don’t care enough to protect your interests with these solutions.