Improvements to Buy-and-burn Token Models

  • Commentary
  • September 12, 2020

In a recent article on Bankless, Jon Itzler, a researcher at the venture capital firm Accomplice, discussed how he reasons about token value flows in the context of blockchain protocols. Among other points, Itzler critically evaluates the standard buyback-and-burn model used by protocols. Citing Maker as an example, he argues the protocol has underserved MKR token holders by buying tokens back at historically expensive points, effectively destroying value. While such practices in traditional markets would be considered reckless financial management, they are commonplace and largely accepted in cryptoasset markets.

Admittedly, MKR is a somewhat unique example: MKR can be minted to bailout Maker in the event of system failure, as happened in March. Given that system failures inevitably lead to declines in MKR’s price, regardless of the price at which tokens are bought-and-burnt, any value added by such practices can be written-off by fresh issuance—recapitalization is more expensive than it would be in normal market conditions.

Notwithstanding this, Itzler’s core thesis is well-reasoned and has spawned discussions at Smith + Crown about how buyback-and-burn models might be improved. Financial officers at publicly traded companies can conduct share buybacks based on stock price deviations from what they consider to be intrinsic or fundamental value. Yet, such analogs are more troubled in crypto markets, where valuation is still a contentious issue and most assets have multiple (and often conflicting) value accrual mechanisms. As such, at least for now, protocols should use purely deterministic algorithms for triggering token buybacks. The most simple and least contentious of such algorithms could be tuned to a 50-day moving average, whereby increased deviations below the average would increase the rate at which a protocol’s treasury conducts repurchases. Potential design issues include the added risk of using price oracles (required for such a mechanism) and the question of who should determine and adapt a protocol’s buyback algorithm. Both these issues are, however, merely gradations of existing concerns, rather than anything novel. Finally, it is worth considering what effect this mechanism may have on a token’s price volatility. Assuming a protocol’s treasury was one of the largest purchasers of its tokens, the public knowledge of the algorithm may introduce a price floor, since other participants would be assured of significant buying pressure at deterministic price points.