Better Late Than Never.

We’re nearly four months late to the Fei|Rari acquisition recap party. We’ve been a bit busy the last few months between leaving our investment banking jobs and beginning our new lives as M&A freelancers. Crypto will very much be a part of this new venture, and maybe soon the entirety of it. In terms of content, a retrospective look at the first DAO-to-DAO acquisition of this magnitude felt as good a place to start as any.

Before we get started, congratulations to everyone involved in the deal. To state the obvious, this was a landmark transaction and an incredibly bold step for each of these teams to take. Doing something that has no real precedent, playbook, or roadmap takes an impressive amount of fortitude.

“Fairness”

“Fairness” is a tough concept and, at the risk of stating the painfully obvious, it can be incredibly subjective. As bankers in traditional public M&A, once the negotiating was done and a proposed price was settled on we would always render a fairness opinion to the management team and board we were advising. Back when Sam and I worked at Qatalyst (a technology-focused M&A investment bank), we helped put together and present a number of these opinions (a few examples: Slack, LinkedIn, Five9, Cornerstone, Quantenna). We strongly believe that a similar concept should exist in crypto, particularly in deals where token-holders need to vote to approve the transaction.

We aren’t here to summarize everything that happened or talk about the strategic merits of the combination. That’s already been done really well by a few folks, most notably by Luca Prosperi (here) and by the BowTiedBull team (here). Our main focus will be on the bottom line: how did this deal value each token, what was the implied exchange ratio, and was it, in a word, fair?

(Spoiler: probably not)

Let’s get into it.

The Deal

The transaction came together quickly, especially relative to a traditional public deal. It was originally proposed by the founders to each other’s community (i.e., the “Term Sheet”) on November 16th, a formal written proposal (i.e., the “Merger Agreement”) was put forth on November 29th, and after a bit more back and forth and a few amendments, the acquisition was confirmed on December 20th.

A quick summary of the final deal structure and terms [1] below:

• Exchange Ratio [2]
  • The exchange ratio was pegged based on the respective 7-day TWAPs [3] of each token starting at ~6:30pm on November 20th and ending November 27th around the same time
    • RGT 7-Day TWAP: $29.07 USD
    • TRIBE 7-Day TWAP: $1.09 USD
    • Exchange Ratio: 26.7x
  • This exchange implied a ~70/30 pro forma ownership split in favor of TRIBE holders (Note: many are calling this a “merger” – it wasn’t. This was an acquisition)
• Pro Forma Governance | Operations
  • Fei took control of the combined treasuries and future assets (obviously, they are now 70% owners of it)
  • Fei took on the liability of Rari’s hack debt (again – obviously. See above.)
  • Both communities were to continue to operate separately but be governed collectively under the TRIBE token
• RageQuit Mechanism
  • TRIBE holders received the right to “cash out” by exchanging any TRIBE held prior to the deal announcement at an estimated “Intrinsic Value” [4] of TRIBE that ended up being calculated at ~$1.08 of FEI (Fei’s stablecoin, read: USD)

Was it Fair?

No. Not really.

A couple quick disclaimers to that statement before we jump into why…

First: it is important to remember that a whole hell of a lot more goes into M&A than deal structure and valuation. Capitalization matters, macro trends in your space matter, relationships matter, synergies matter, investor and team sentiment matters – there are a lot of ways to get negotiating leverage in a deal that have nothing to do with “fairness.” TRIBE holders controlled a much larger treasury than RGT holders at the time of the transaction and RGT holders also had an $11.5M debt hanging over their heads. The market knew about all of this (and thus one can assume these facts are baked into the respective trading prices) but that doesn’t make them insignificant.

Second: we do not have any proprietary financial information on either DAO. There is a ton more to look at once you get under the hood of any entity, particularly as it regards intrinsic valuation calculations. This is all simply based on what was publicly available and mostly focuses on the calculation of the exchange ratio (i.e., the transaction price).

Let’s get started.

VWAP vs. TWAP [5]

Traditional deals often use VWAP prices when setting exchange ratios and/or thinking about deal premia. The primary purpose in using an average price versus a spot price when calculating an exchange ratio is to avoid one party getting overly penalized for sporadic price movement just prior to whenever the respective prices for said exchange ratio are captured. This wisdom holds even more weight when considering lightly-traded DeFi assets often prone to wild swings. The parties clearly understood this fact (at least on some level) and used a 7-day TWAP as the basis for their exchange ratio, presumably to account for it.

To be clear, TWAP and VWAP are both price averages calculated over a period of time, one is “time-weighted” (equally weighted dependent on whatever time intervals you care to use [6]) and one is “volume-weighted” (weighted based on the volume of tokens sold at each price [7]). That said, depending on the circumstance, these two methodologies can have vastly different outcomes.

At the risk of sounding dogmatic, VWAP is simply the proper way to calculate a fair market price in the context of M&A. Even ignoring the “TradFi” precedent, determining fair value by focusing on the volume of tokens traded at each price rather than the time spent at each price is a virtual no-brainer. Again, this is particularly important when looking at a lightly-traded asset – a descriptor that could be applied to nearly every single DAO token in existence right now. One trade of 10M tokens for $20 in a single instance should be weighted more heavily than 100 trades of 1 token for $15 spread out over the course of five minutes. That’s an extreme example, but you get the point.

Looking at the prices and volumes in the period of time used, the TWAP calculation resulted in an exchange ratio that cost RGT holders nearly 15% of value as compared to a VWAP calculation. Not insignificant.

“Unaffected” Date

A number of folks have pointed out the fact that RGT holders got unfairly penalized in this transaction because the exchange ratio looked at a period ending 11/27 when the potential deal was publicly brought to the communities on 11/16. There was actually an amendment proposed to change the exchange ratio calculation to reflect this fact (unsurprisingly, TRIBE holders shot this amendment down given the 26.7x ratio had already been voted on and agreed to). We whole-heartedly agree with this sentiment, the exchange ratio should never have been measured post-announcement. In reality though, the negative effect on the price for RGT holders may have been even worse than people think.

In many public mergers or acquisitions the “announce” date does not end up being equivalent to the “fair market” or “unaffected” date. The term “unaffected” is used because it, quite simply, describes the prices at which shares (or in this case, tokens) trade without any effect of a potential merger. It is important (and standard in traditional deal-making) to begin price negotiations with the baseline being the standalone fair market value, not a value that has been influenced by market sentiment about the deal. Public deals often leak, it’s just the nature of the business. In fact, three of the five deals I referenced earlier that Sam and I worked on dealt with leaks of some sort and we had to base our valuation analysis off of unaffected rather than simply pre-announce prices. There is also plenty of precedent of buyers/sellers agreeing to an unaffected date even without an official leak due simply to unusual trading and/or disclosures (this exact situation happened in one of my deals just last year).

In crypto there are no trading restrictions on literally anyone, ever. This matters because that means that there doesn’t need to be a “leak” to affect trading price and volumes, the moment a single individual with knowledge of the deal makes a trade to buy or sell either token involved - the price is affected. Let’s take a look at TRIBE’s trading volume [8] in the days leading up to the announcement…

That spike in volume (and subsequent price runup) screams “THIS PRICE IS AFFECTED” – could be an anomaly though, right? Let’s take a look at RGT’s trading volumes over the same period.

Both DAO governance token volumes spiked over the same two-day period less than two weeks ahead of the original deal proposal. From 11/4 until 11/15 (the day before the initial public proposal) TRIBE ran 31% and RGT fell 8%. For context, Ethereum was up 4% over that same period.

Let’s be very clear - this could legitimately be an innocent coincidence. The Rari spike is almost certainly due to the fact that there was an RGT Binance listing announcement on that date, though that doesn’t exactly explain the immediate selloff thereafter or the fact that the acquisition news barely registered as a blip on the radar from a trading perspective. Fei announced $160M of TRIBE buybacks on 11/5 as well, which certainly pumped the price of TRIBE in the days that followed.

Whether or not this would have actually changed the proper “unaffected” date in practice would obviously have been a matter of discussion and negotiation. Given the ramifications, it should have been discussed and folks (specifically core team members and investors) with pre-announce knowledge of the deal should have been formally asked to disclose to the communities whether or not they had bought or sold either governance token after having knowledge of the deal. The moment anyone with knowledge of the deal made a trade, the clock on market price calculations should have stopped. End of discussion.

We’ll spare you the math but if you run the 7-day VWAP using 11/15 as your unaffected date (which should have been done at bare minimum) RGT holders gave away more than 25% in value relative to the exchange ratio the deal was struck at… but if you run it with 11/4 as your unaffected date suddenly RGT holders gave away more than 70% of fair market value and an additional ~17% ownership in the combined company. Yikes.

The RageQuit

We won’t get deep into everything here but suffice to say that incentivizing a cash out from existing token holders at a price greater than market is a guaranteed way to encourage less-than-ideal trading behavior (i.e., RageQuitting above TRIBE market value and buying back in at a discount price to artificially increase ownership [9]). Even if token holders were perfectly altruistic (lol) and only those that actually wanted out utilized the mechanism, you are setting yourself up to jettison the value of go-forward token holders in favor of departing token holders. This idea is somehow made even worse when said cash out is funded by your stablecoin that depends on supply control to keep itself pegged to USD. You can find me squarely in the “would not recommend” camp on this one.

Control Premium

This deal deserved a control premium. This is probably the simplest concept in this piece: when you’re buying up governance instruments in an entity (whether it’s represented by stock, tokens, marbles, whatever) the most expensive piece of equity should be the one that tips you over the 50% ownership threshold because, like the name implies, you are now in control. Note that this premium concept isn’t always true when ownership is split roughly 50/50 between entities in what is often called a “Merger of Equals” … but it isn’t even close in this case. Tough to put a price on power but “market” in tech usually starts around 20-30% with some premiums getting a hell of a lot higher than that.

This wouldn’t be fair to layer on top of all the price adjustments we’ve already made to find a “fair” exchange ratio because those already would have tipped the scales a lot closer to 50/50 and into territory where no control premium would normally be required… but something that should likely have been considered given the price that ended up being struck and should be noted on a go-forward basis, nonetheless.

Conclusion

As stated before, these are all exchange ratio calculation critiques taking market prices at face value, we haven’t even touched relative valuation or the wide range of different potential deal structuring options the participants in this deal might have considered. Real valuation work is effectively impossible to do without some sense of financials from each DAO to base it on. There would obviously be a lot more to consider once you get under the hood and a lot of this valuation work may have tipped the scales a bit more back in TRIBE holders’ favor relative to our analysis above. All that said, when simply going through the market-based exchange ratio baseline exercise there appears to be a good amount of value left on the table for RGT holders given how the free-trading, unaffected markets seemed to value each token. At the very least the original negotiating baseline should have started in a completely different place.

With that in mind, here are a few things we would suggest for protocols thinking about M&A moving forward:

• Don’t be afraid to negotiate: There’s a lot of value at stake when looking at a deal like this. You don’t necessarily need to be aggressive or hostile about it, but it’s important to spend real energy to ensure you aren’t leaving anything on the table for your community and team.
• Utilize precedent: Precedent provides super helpful guardrails for these conversations. Even in a space like this without any real precedent there is plenty of precedent and common practices from TradFi that are still very applicable and should be employed to try to come to a fair outcome.
• Engage an advisor: This is a transparently self-serving statement… but having an experienced third-party involved to help with both of the above (in addition to managing the whole process) can be invaluable. This has always been true in traditional business and we believe it will continue to be true in crypto.

We’re going to continue to write M&A- and valuation-focused crypto pieces like this, hopefully with some degree of regularity. Shoot us a DM or @ us and let us know what you think! Feedback, thoughts, and discussion are very much welcome.

@jstastny101 / @sam_bronstein

About the Authors

Jordan Stastny and Sam Bronstein were previously M&A advisors at Qatalyst Partners. While at Qatalyst, they advised on over $150Bn of M&A volume across some of the most significant deals in the technology industry, including the sales of Slack, LinkedIn, Mailchimp, Qualtrics, Glassdoor, and others. Jordan and Sam co-founded MSPC Partners at the beginning of 2022 to advise early-stage technology companies on M&A.

[1] The majority of these terms can be found in the originally voted on merger doc. There were some adjustments after the fact that we’ve captured in our summary, but the structure from this doc remained mostly unchanged.

[2] For the uninitiated: the “exchange ratio” in a deal is the relative number of tokens issued by the acquiring entity per token of the acquired entity. In this case, it is calculated simply by taking the offer price of RGT divided by the pegged price of TRIBE. Because this number determines how many cumulative tokens the acquired entity token holders receive, it ultimately determines the relative ownership in the combined entity. This is the single most important number in any token-for-token deal.

[3] Time-Weighted Average Price. More on this later.

[4] The “Intrinsic Value” of TRIBE was calculated using the TRIBE Protocol Equity (i.e., the estimated net assets of Fei) divided by the TRIBE Circulating Supply (i.e., all outstanding TRIBE tokens, including unvested tokens but excluding TRIBE held in treasury and TRIBE yet to be minted)

[5] Note that our illustrative VWAPs calculated here are done on a daily basis and are fairly archaic. Simple Daily $ Volume ÷ Open/Close Price Average = Token Volume then a simple SumProduct of the Daily Close Prices against that Token Volume. There are better, more precise, ways to do this using individual trades but there is no “Crypto Bloomberg” yet as far as I can tell… If anybody is making such a thing give me a shout, I have thoughts.

[6] TWAP = (P_1 + P_2 + P_n) / n where P = Price and n = number of intervals.

[7] VWAP = (P_1*V_1 + P_2*V_2 + P_x*V_x) / ΣV where P = Price, V = Volume, and x = the number of trades – written more simply: ΣPV / ΣV.

[8] Trading data per CoinMarketCap.

[9] Anyone with a compliance or regulatory background should probably avoid that linked Twitter conversation altogether to avoid being sent into cardiac arrest.