bitbanter

Having trouble listening? Download the audio here.

A Liberal Reexamination of Coinbase’s Master Plan

Nearly six-and-a-half years ago, Brian Armstrong wrote The Coinbase Secret Master Plan. Despite very-public “Secret Master Plans” being the cutesy cargo-cultish Silicon Valley thing to do in the 2010s…Brian did an admirable job calling Coinbase’s shots. Certainly better than AirBnB did; nowhere in their master plan did they say “build a dominant rental marketplace out of dark patterns so twisted the UX was rejected by TicketMaster for being too anti-customer.”

But with the benefit of time — and the announcement of BASE — we can, perhaps, reinterpret the not-so-secret master plan and peel back a few layers of carefully constructed copy. So let’s dive in.

Phase 0: Throw Out The Phases

In the original master plan, there were four phases of adoption, likening cryptocurrency’s path to the gradual-then-sudden spread of the internet. This framing is mostly nonsense, so we’re going to throw it all out.

Yes, Coinbase has many many millions of users, and yes, they have developed or supported various things mentioned in 2016’s master plan — that’s why I said Brian did an admirable job calling his shots. But the plan itself lacks a certain texture, a sort-of je ne sais quoi, that particular phenomenon ancient philosophers called honesty.

Let’s add some, shall we?

Phase 1, 2012-2016: Be the Least Sketchy Place To Convert Dollars to Bitcoin (then Ether)

In the beginning, there was Bitcoin. And, perhaps if not Credibly Neutral, it was at least Good. But even three years into its existence, acquiring bitcoin through the purported “One-CPU-One-Vote” PoW mechanism proved out of reach for those of us who couldn’t afford industrial-scale warehouses of “One-CPU-One-Vote” machines. Bitcoin exchanges naturally emerged as a solution.

The trouble is that these exchanges demanded an extraordinary amount of trust, which ran counter to the whole premise of the damn protocol. And most of these places couldn’t even administer their databases properly, or were connected to the mob, or both. This whole state of affairs was decidedly not great for bitcoin adoption.

And along came Coinbase. A YC-incubated Silicon Valley startup with peerless credentials, back when that meant something. Was it just a Bootstrap front-end backed by “web-scale” MongoDB pointing to a hot wallet that Fred and Brian shared? Maybe! But they had trust, and that — again, very much ironically — mattered most.

Look I’m going to level with you: this was actually Coinbase at its best, and I miss this era. Their stated goal — pre-hurr-durr-Elon-esque-Secret-Master-Plan — was to be the best bridge to bitcoin, while reducing fees for users by scaling their business. In modern, catchier parlance: Coinbase was aiming to be a big, dumb, user-friendly pipe from tradfi into crypto. And you know what? I loved them for it.

Everything changed when the Chikun Nation attacked.

Phase 2, 2017-2022: SatoshiLite Dice 2.0, But With KYC™!

I have nothing personally against Charlie Lee. Did he create a useless bitcoin fork kept alive by memes and a desperate will to remain relevant? Sure, but who among us hasn’t? Did he cringe-co-host a podcast with a bunch of lucky LARPers? Yes but I write articles like this, so who am I to judge?

These things are all forgivable. What still grinds my gears is that he leveraged his position at Coinbase to list Litecoin, peaced out, then proceeded to sell his own stash at Litecoin’s absolute all-time-high onto Coinbase’s unsuspecting retail rubes.

And not only was Coinbase totally indifferent to this shtick, I suspect that internally they were quietly pleased. Litecoin was the third cryptocurrency they added — after bitcoin and ether — and it was here that their business fundamentally shifted.

The fees to help users acquire digital gold were peanuts compared to the fees for encouraging users to gamble on digital anything. They added innumerable nonsense coins and started segmenting their pricing to encourage trading; raising fees on simple purchases while lowering prices on the “pro” trading interface. That this still happens is a staggering failure, and one that should shame anyone at Coinbase tasked with advocating for their users. (if anyone like that actually exists at any large technology company)

All of this was happening while they were pursuing the “regulatory mullet” strategy; business up top but party in the back. On the “business” side they tightened their KYC procedures while acquiring an incredibly sketchy and morally dubious blockchain analytics startup — a moment which compelled me and others to delete our Coinbase accounts. But in the back? It was party time. They added tokens with careless abandon, collecting fees every which way.

Not to mention that they IPO’d during this period, which was doubly ironic because:

a) If they believed in the promise of public blockchains as the future of finance, why would they list their equity on a traditional financial market? This IPO could have been a smart contract! b) They did a “direct sale” from insiders, which they claimed allowed them to avoid Wall Street fees (true)…but it also meant they were effectively dumping their insider equity on retail, which uh, wasn’t too dissimilar from what Coinbase’s token listings enabled not-actually-decentralized projects to do. Points for consistency I suppose.

To be forthright, this isn’t solely Charlie Lee’s fault, as much as I enjoying laying blame on him. Coinbase’s “pull the regulatory drawbridge up while profiting from degeneracy” strategy was practically encoded in its DNA — a consequence of its corporate structure and the fundraising it pursued. One could argue that — particularly with the rise of tokens and DeFi on Ethereum, and Binance’s dominance outside the US — they had a fiduciary obligation to their shareholders to build the greatest shitcoin casino on Earth.

That doesn’t make it any less sad.

Phase 3, 2023-????: Become Citadel for DeFi, With KYC That Haunts The Chain Forever

There’s good news and there’s bad news.

The good news is the digital shitcoin casino revenue dried up in 2022, and Coinbase is pursuing other business lines.

The bad news is that Coinbase’s pursuit of other business might forever taint what we call “DeFi” today and enshrine a financial dystopia that is measurably worse than the current system.

If you’ve suffered this long into this post, you are finally being rewarded with its connection to the headline (congrats). In case you haven’t heard, Coinbase is building an L2 on Ethereum called Base. Buried in the mess of PR and marketing emojis is the most important part of the news: that Coinbase will be running the sole sequencer for Base, with nebulous plans to “decentralize it in the future.”

Oh joy. The future of finance is now “draw the rest of the fucking owl” of finance. Nothing could possibly go wrong here.

So what does the rest of the fucking owl look like here? Well me, I’m no highfalutin tech-executive-cum-prophet, I’m just a lowly crypto advocate, so you know, take this all with a big grain of salt.

But it seems to my untrained eye that the sequencer will never be “decentralized,” and if Ethereum and Base were both to become the bedrock of finance, Coinbase would never loosen its grip on the preeminent gateway to “DeFi.” This is not the place for an L2/rollup primer (and even if it were I would definitely not be the one to give it) but controlling the sequencer means Coinbase controls transaction ordering, putting them in a unique position to both privilege transactions from their centralized service (or ones that pass future KYC-checks that associate all your activity on-chain to your Coinbase account) and profit from ordering a la MEV.

Brian Armstrong was on Odd Lots yesterday (which may have in fact prompted much of this post) and it’s a good listen — both for the questions he answers and the ones he doesn’t. But there was a moment where he claimed that Coinbase had higher fees because others hid theirs with payment-for-order-flow…well knowing what we know about Base, it just didn’t sit right.

You know why Base exists? Because DeFi is eating into centralized exchange revenue (good!), and many DeFi contracts can do things centralized exchanges can’t (also good!). And with completely transparent blockchains (bad!), a whole lot of theoretical centralized exchange revenue is instead lining the pockets of MEV searchers and validators (unclear if good or bad!). Wouldn’t it be nice for Coinbase’s shareholders if they could recapture some of that revenue on a higher layer? Wouldn’t it be great if they could take their huge regulatory advantage — their KYC’d customer base and bridge into the traditional financial world — and enshrine it on a layer they control?

Sure, it would reinforce the incentive for the transaction graph being completely visible on Ethereum, moving us farther and farther away from cryptographically enforced privacy on-chain, and potentially open up your entire “future of financial history” to adversaries with one bad leak of Coinbase’s KYC database. And yeah, it would transform the vast majority of their business into Citadel 2.0 Plus NFTs.

But think of the value that could accrue to Coinbase shareholders!

Phase 4: Can we avoid Phase 3?

Look, this has all gone on a bit too long, when this was meant to be a cheeky shitpost about Coinbase’s arc as a business, and what compels them to make an L2.

And I feel a little bad, because Coinbase has done some great things for our industry, and I happen to have quite a few friends who work there (or have worked there) that have their hearts in the right place. I know there are good people there doing good work, and I am grateful for them.

But man these incentives are absolutely fucked. I am not sure any number of good people can change the systems at play there. It’s up to the rest of us not working at Coinbase to consider the path we want for finance (and crypto’s) future, and to create compelling enough systems to bend the incentives at play at Coinbase and others back to benefiting their users.

There’s this weird, recent tendency for crypto folks to not challenge the dominant players in this industry, which befuddles me; the whole reason so many of us got interested in this movement was to take on the JP Morgans of the world, not stand idly by as new ones are minted and say “oh well, good game, no rematch.”

I am obviously biased (care of Penumbra) but even still, there are myriad approaches here that could transform the nature of centralized exchanges, and bring about a world where fiat bridges are the dumb pipes many of us dreamed they could be ten years ago — while we mediate all of our payments and transactions on publicly accessbile, permissionless protocols, without them violating our individual right to privacy or front-running our trades. We just have to build it…and uh, not on Base.