1. What does the smart contract look like. Anything interesting?

2. What kind of custodial access does PayPal have?

3. What does the PayPal wallet look like?

4. What is PayPal’s financial incentive with PYUSD?

Here’s what I found.

1. What does the smart contract look like? Anything interesting?

Here’s the smart contract deployment: https://etherscan.io/address/0x6c3ea9036406852006290770bedfcaba0e23a0e8

And here’s the smart contract code repo: https://github.com/paxosglobal/pyusd-contract

By looking at the transaction history, we can see initial deployer activity nearly 282 days ago (November 7, 2022). PayPal has been working on this since early November 2022 at least, around the same time as the FTX collapse.

Note that PayPal worked with Paxos to issue PYUSD. Paxos has worked with other fintech giants, like Mastercard, to deliver various blockchain services. Read more about the Mastercard build here.

Three things in particular stuck out to me in the PYUSD smart contract:

First, the betaDelegatedTransfer function.

Generally speaking, delegate functions allow users to interact with smart contract functions without needing gas tokens. This is done utilizing meta-transactions, which allows users to sign a message describing an intended transaction, and passing the resulting signature to a third-party (called a relayer) who executes the transaction using their gas tokens.

In this function, a user who wants to transfer PYUSD to another address can do so without holding ETH by delegating the on-chain transaction execution to a relayer who holds ETH and calls the betaDelegatedTransfer function on the user’s behalf.

I’m not sure exactly how PayPal is utilizing this, but providing this option is pretty cool.

Second, the freeze function, which gives PayPal the ability to freeze the PYUSD balance of any Ethereum address.

Third, the wipeFrozenAddress function, which gives PayPal the ability to wipe the PYUSD balance of any Ethereum address.

These two functions help answer my next question.

2. What kind of custodial access does PayPal have?

The two aforementioned functions in the PYUSD smart contract work together to give PayPal full control and custody over the PYUSD balance of any Ethereum address, not just those that operate within the PayPal wallet interface.

Although concerning, this implementation showcases PayPal’s intention of “crypto supercharging” their existing digital fiat currency framework. PayPal’s goal with PYUSD was not to create a decentralized, permissionless token, but rather to merge other benefits of crypto with their existing architecture within the safe zone of US regulation. These benefits include near instant settlement and simplified cross-border transfers, both of which PYUSD enables.

I also found the following in PayPal’s terms:

“PayPal will review Crypto Assets sent to you, including for risk holds, sanctions screening, and other compliance and risk considerations. Crypto Assets will not be credited to you until they pass PayPal’s risk and compliance review.”

PayPal has the right to restrict any funds sent to your PayPal wallet for any reason they want. This extend’s PayPal’s crypto custody rights beyond just PYUSD when operating within their wallet interface.

Furthermore, PayPal boasts fee-less PYUSD transfers to other PayPal users. This is almost certainly enabled by an off-chain ledger as Ethereum transaction fees are wildly expensive. This is not dissimilar to other implementations like Coinbase’s exchange accounts.

These points lead me to my next question.

3. What does the PayPal wallet look like?

First, we know that the PayPal wallet is a custodial Ethereum EOA. Therefore, it is capable of receiving any Ethereum token. However, given the terms I shared above, PayPal has the right to do whatever they’d like with the funds sent to your PayPal wallet. Classic “not your keys, not your crypto”.

Will the PayPal wallet offer swapping, bridging, and other in-wallet tools? Their terms revealed the following about converting to/from PYUSD:

“When you do a conversion: First, your Crypto Asset is sold for U.S. dollars. Second, the amount delivered into your Balance Account will be used to fund the purchase of your selected Crypto Asset”

PayPal makes it clear that they are not facilitating swaps, but rather sells and buys that replicate swaps. They are careful to use “conversion” terminology for this reason. Therefore, these conversions fall under the fee structure, which PayPal has outlined here. Note that these fees only apply when converting from PYUSDC to another cryptocurrency, aka selling PYUSDC and buying a different cryptocurrency.

4. What is PayPal’s financial incentive with PYUSD?

I had trouble finding this information, so my speculation is that PayPal will follow Circle’s business model.

Circle’s CEO discussed their financials for the first half of 2022, outlining the following:

"Circle had $779 million in revenue in the first half of the year, already surpassing the $772 million for all of 2022… generated $219 million in adjusted earnings before interest, taxes, depreciation, and amortization in the first half… The company has over $1 billion in cash on its balance sheet as of June."

Circle’s cryptocurrencies are backed 1:1 in their reserves. They currently have $26.1B of USDC in circulation, and $26.2B in reserves. These reserves are composed of cash at Reserve Banks, as well as the Circle Reserve Fund which is “a SEC-registered money market fund which holds a portfolio of short-dated US Treasuries, overnight US Treasury repurchase agreements, and cash.”

Simply put, when a user buys USDC using dollars, those dollars are put to work generating yield which Circle captures as revenue. PayPal will likely adopt something very similar, making PYUSD an easy avenue for PayPal to grow their balance sheet.

Conclusion

In sum, PayPal’s issuance of a stablecoin is bullish for the crypto space. Although custodial in nature and questionable in the ethos of crypto, PayPal is presenting itself as a pioneer in the crypto space. The issuance of PYUSD highlights how crypto can work in a regulated environment, and inspire other fintech giants to follow. Furthermore, PayPal’s massive user base and existing distribution as a payment processor will make them the largest crypto on-ramp in the space. If rolled out globally, 435 million users will have access to an on-ramp which can give them access to the broader crypto space.

We spent roughly 6-months at Peaze building a “user experience stack for crypto developers”. As part of this stack, we offered Account Abstraction tooling and experimented with various strategies: Smart Contract Wallets, Multi-Party Computation, ERC-4337, etc. After building various solutions and delivering to customers, we concluded that Account Abstraction is too immature for feasibility.

Although crypto’s UX challenges are felt by both new and experienced users, UX rhetoric around wallets are mostly directed towards helping onboard users who are new to crypto. We’ve all heard the same phrases over and over again: “onboard the next billion users”, “crypto isn’t good until my grandma can use it”, “Metamask sucks”, etc. We were guilty of over-regurgitating these phrases at Peaze too.

Builders don’t care about new users. There’s two main reasons for this:

First, new users demand hand-holding beyond the wallet onboarding layer. What happens when they don’t store their backup key and get locked out of their wallet? Do they understand native tokens vs ERC-20 tokens vs NFTs? Do they understand signatures and transactions? New users don’t just need help with private keys and gas abstraction, they need help with everything. As a result, the new user journey is much more complex than the experienced user journey. Accounting for two user journeys in an app requires too many resources, so developers are opting out.

Second, new users will not drive volume/liquidity like existing users who are already putting up numbers. Why should developers build for a demographic that won’t be as financially lucrative? According to DefiLlama (as of August 9, 2023):

• DEXs had ~$69b of total volume in July ‘23

• Bridges had ~$7b of total volume in July ‘23

• DeFi’s TVL is ~$41b

• Stablecoin market-cap is ~$124b

These are massive numbers, especially in a “bear market”.

These two reasons make it very hard for developers to justify building for new users, making Account Abstraction and Smart Contract Wallet support a non-option. Not to mention the technical overhaul of implementing something like ERC-4337 which includes UserOperations, Bundlers, Paymasters, etc.

Now, it’s worth mentioning that these arguments are largely within the scope of Ethereum and its ecosystem of scaling solutions (Polygon, Arbitrum, Optimism, etc.). It’s undeniable that the Ethereum ecosystem is driving an overwhelming majority of the volume and liquidity on-chain. Therefore, the Ethereum ecosystem is feeling the “1st-world problem” of not catering to new users because their existing users are good enough. Other ecosystems like Cosmos, NEAR, etc. have disturbingly low numbers in comparison: Ethereum mainnet has ~$24b TVL, whereas NEAR has ~$34m TVL (0.14% of Ethereum’s TVL). Therefore, these “Tier-2” chains are likely juggling between two choices: take existing users from the Ethereum ecosystem, or onboard new users.

It’s not very clear which choice is the right one because ecosystems are sticky; they each have their own wallets, functionality, security, etc. When a user gets situated within an ecosystem, it becomes very hard for them to bridge out of that ecosystem, re-learn everything, and risk their security while they’re at it. For this reason, it becomes enticing for “Tier-2” chains to focus on new users, but new users aren’t as financially enticing.

So, why would developers implement Smart Contract Wallet support if they’re not widely adopted by users? Why would users adopt Smart Contract Wallet solutions if they’re widely unsupported by dApps? We have a classic chicken-or-the-egg problem.

Crypto likely need a breakout use case that convinces a large number of users to adopt Account Abstraction and Smart Contract Wallet solutions. However, as it stands, Smart Contract Wallets have a long way to go in convincing developers to support them. Embeddable wallet solutions that ease the use of an EOA, like Coinbase’s WaaS or Magic Connect, are more feasible as they’re much less intrusive for developers to build with. However, they are questionable in their non-custodial nature, but this tradeoff can be justified for many use-cases.