###### tags: `Fingerprints DAO` `Governance`
# Formulation of a Buyback System

> *Whenever $PRINTS are mentioned in this document, Voxelglyphs are understood to be included.*

### When should it be in the interest of the DAO to repurchase its own token?

It stands to reason that whenever there are people willing to sell tokens at a price below the market value of the underlying assets (minus its liabilities), divided by the total number of outstanding tokens, the DAO (or any investor) should be willing to purchase them.

Thus, if:

$$Price\ per\ $PRINTS < \frac{Net\ Asset\ Value}{$PRINTS\ Outstanding}$$

then a buyback (or any purchase) is worthwhile.

### Why are tokens, then, consistently traded below NAV?

Given the underlying assets are locked in the DAO's vault, an investor looking to arbitrage the $PRINTS token will find no way to sell the underlying assets to close their trade, unless if taking full control of the DAO, which is impossible with the current available liquidity.

Additionally, there is inherent uncertainty in the estimation of the value of the assets contained in Fingerprints' treasury, making the previously stated calculation riskier. Factoring this risk in means a higher spread is necessary to make the trade worthwhile.

Other details such as settlement time (very important since dealing with volatile assets), transactions costs, and liquidation procedure of the assets — in which the speed of liquidation could impact the prices of treasury-held collections — add further risk to any investor seeking to arbitrage the $PRINTS token.

Thus, these problems need to be addressed:
1. Lack of access to underlying assets
2. Proper estimation of value of underlying assets
3. Settlement time
4. Transaction costs

In the absence of these hurdles, investors can easily arbitrage whatever amount of lower-than-NAV liquidity is ever made available in decentralized exchanges, making the price-to-NAV gap vanish.

This would make $PRINTS closely track the value of the underlying collection.

### Solving identified problems

**1. How to make treasury assets available to tokenholders?**

As is, $PRINTS holders have no direct access to their portion of underlying assets, making direct arbitrage impossible.

This particular hurdle isn't relevant if considered from the point-of-view of the DAO itself, as it can exercise control and liquidate them if needed.

Consistent buying of $PRINTS from the DAO could then provide a way for active traders to close *most* of the price gap to NAV. This solution bypasses the need to make treasury-held assets directly available to tokenholders, which would be operationally challenging and risky.

Actual liquidation of treasury-held assets would only be necessary if the amount to be repurchased nears the available cash held by the DAO, as a certain amount of cash is necessary to sustain daily operations.

However, active arbitrage trading will not happen in the absence of an effective buyback system that addresses problems 2 and 3, and that effectively gives traders a way out of their positions if needed.

**2. How to accurately estimate the value of the treasury?**

Here's the current breakdown of Fingerprints' treasury:

| Asset | Value (Ξ) |
| -------- | -------- |
| NFT Collection | 8,463.93 |
| Cash & Equivalents | 1,174.45 |
| Other | 12.5 |

It currently has no liabilities.

As we can see, the NFT collection represents a majority of the treasury assets, which makes accurate marking of its value critical to an informed buyback decision.

The value in the table above has been estimated through fetching floor prices from each sub-collection and accounting for tier premiums.

Here are some estimates done through different valuation methods:

| Marking Method | Collection Value (Ξ)¹ | NAV (Ξ) | $PRINTS price (Ξ per '000) |
|---------------------------------------|----------------------|---------|-----------------------------|
| Floor prices w/ tier premiums | 8,463.93 | 9,650.88| 1.363793 |
| Floor prices | 6,588.82 | 7,775.77| 1.098816 |
| Midpoints w/ tier premiums | 5,421.61 | 6,608.56| 0.933874 |
| Midpoints | 4,015.55 | 5,202.5 | 0.735180 |
| Highest current bid w/ tier premiums | 1,947.79 | 3,134.74| 0.442979 |
| Highest current bid | 1,424.16 | 2,611.11| 0.368983 |

¹ *as of 2023-09-04*

Assuming assets are to be sold in an orderly and spaced out manner, then higher estimates are likely applicable.

For the sake of a buyback system, however, it is safer to value the NFT collection using a conservative method, should the need for rapid liquidation of collection pieces ever arise.

In which case, the DAO can make a profit by providing below-value liquidity to those who want it fast, and then slowly disposing of underlying assets at higher prices (or holding them until favorable conditions arise).

**3. How to make token buybacks timely enough?**

As previously discussed, there is a need to mitigate warehousing risk by providing a fast settlement option for those willing to arbitrage $PRINTS tokens.

The ideal scenario is an ongoing buyback purchase system where investors can exchange their $PRINTS for the equivalent amount of ETH at NAV valuations, thus eliminating settlement lead time risk.

It presents the upside of being easy to develop and deploy, however, there will be a large risk of manipulation if a live feed of oracle-based collection estimates are to be used.

Checks and balances must be put in place to mitigate these risks, such as hard-coded NAV prices that are manually updated on a weekly basis with human supervision, or a moving average-based NAV estimate.

Additionally, limiting the total amount of buyback proceeds available in a specific period is an important way of capping losses should anything go wrong with the buyback system.

**4. How to actively fund buybacks, if needed?**

The DAO requires a minimum amount of cash to fund its basic operations. This monthly burn currently sits at **24Ξ**, assuming no revenue or yield is generated.

Choosing to keep at least **6—12 months** worth of expenses in liquid assets means the liquidation of collection pieces should begin once the available cash reaches **144—288Ξ**, thus making about **950Ξ** currently available to fund the buyback system.

If the buyback is to operate at the lowest marking method, that means ~**2.57m $PRINTS** could be repurchased before selling of treasury pieces is necessary.

This would be enough repurchase all Voxelglyphs and all $PRINTS from holders with below-50k balances, thus making the need for selling treasury pieces an unlikely scenario in the short-run, specially when buying back at such a low NAV estimate.

This could also be made less likely by limiting the total amount of buyback proceeds available in a given time-window, which would make the prospect of having to sell pieces from the collection obvious and allow for plenty of time to attend to it.

**5. How to reduce transaction costs?**

A DAO-deployed buyback smart contract with a custom UI is likely the most cost-effective way to implement the buyback system, since it can avoid DEX and marketplace fees, and will likely incur lower gas costs due to reduced transaction complexity.

It will also allow for a simpler operational overhead and the setting of critical parameters previously discussed in this document.