Digital Asset Valuation: Lessons from FTX

What can we learn about digital asset valuation from the collapse of the once-dominant cryptocurrency exchange?

In one of the highest-profile digital asset litigations to date, the cryptocurrency exchange FTX filed for bankruptcy after a liquidity crisis caused the exchange to cease operations. The basis for customer claims against FTX in the bankruptcy proceedings comprised over 1,200 different digital assets, including cryptocurrency, digital tokens, and derivatives such as futures, leveraged tokens, and tokenized stock. The bankruptcy, which involved the liquidation of FTX's holdings, illustrates the unique challenges of valuing a wide range of digital assets against the backdrop of significant liquidity and asset marketability constraints.

In this feature, we describe some of the considerations that informed the digital asset valuation in FTX and provide explanations of key concepts.

Valuation of Illiquid Assets

The FTX case was unique in that the asset valuations had to account for the effect of the complete liquidation of FTX's holdings. Given the magnitude of FTX's holdings, even an orderly liquidation was expected to have substantial valuation implications for assets with few buyers and/or low daily trading volume. For these illiquid digital assets, complete liquidation of FTX's holdings would likely have impacted the market price, and therefore valuation required the calculation of "asset liquidation discounts," or the discounts at which FTX would have been able to sell their holdings in an orderly liquidation that started at the time of the bankruptcy filing.

The magnitude of an asset liquidation discount will vary across classes of digital assets. In the FTX matter, there were several digital tokens for which the underlying markets could not realistically absorb FTX’s positions, given the size of the positions relative to the tokens’ daily trading volume. In one example, liquidation of the token by FTX, assuming an orderly sale of 5% of the daily trading volume per day, would take more than 1,000 years. In that case, the valuation discount reflected the fact that, even following a slow liquidation strategy, the liquidated value of the token was very small relative to the original holding.

Non-Marketable Assets

In the FTX bankruptcy, the face value of customer claims associated with non-marketable assets was approximately $1 billion. When valuing these customer claims, it was necessary to apply a valuation discount for lack of marketability (DLOM). Non-marketable digital assets are those that are subject to a vesting or lock-up period, during which customers cannot sell the asset. DLOM accounts for the opportunity cost and additional risk associated with holding a locked asset: A holder can’t take advantage of price upswings and has increased exposure to price crashes.

For non-marketable digital assets, the size of a discount for lack of marketability increases with the length of the vesting period and the volatility of the underlying asset's value. The high volatility of crypto assets and the long lock-up periods meant that discounts for lack of marketability associated with at-issue assets in the FTX bankruptcy were substantial.

Discounts for lack of marketability tend to be model-dependent, particularly at longer time horizons. For example, two of the commonly used DLOM models, Chaffe and Finnerty, perform poorly at long time horizons, generating results that are artificially constrained or that run contrary to economic theory. The DLOMs produced by the Finnerty model arbitrarily cannot go above a discount of approximately 32 percent, no matter the length of the non-marketability period or the characteristics of the asset at hand. Perhaps even more problematically, at long time horizons, the Chaffe model improperly implies that DLOMs decrease as the length of the non-marketability period increases. To improve estimation accuracy, DLOMs in the FTX bankruptcy relied on Monte Carlo simulations based on the modeling setup of Finnerty (2012), allowing for the direct estimation of the discount and avoiding the constraints imposed by the closed form solution.

Crypto Derivatives

Some of the creditor claims in the FTX bankruptcy were for crypto derivatives. Unlike tokens traded across a wide range of venues, crypto derivatives are specific to the exchange they are traded on.

As a result, the appropriate valuation methodology for the FTX crypto derivatives depended on the specifics of the FTX exchange, including its method for clearing trades and the nature of the contracts. Among the most actively traded digital assets on the FTX exchange were perpetual futures.

An asset liquidation discount was not applied to FTX perpetual crypto futures. Because of the high frequency with which the exchange updated customers’ margin account P&Ls, it was reasonable to expect that customer P&Ls closely reflected futures prices at any snapshot in time. Accordingly, the value of customer claims associated with crypto perpetual futures was the value of their margin account at the time of the bankruptcy filing.

Takeaways

The FTX bankruptcy case illustrates that valuation requires a thorough understanding of the characteristics of the assets and the markets they trade in. In the FTX matter, those characteristics included liquidity and lack of marketability issues, along with the specifics of the settlement process for the derivative instruments. ■