[SQ – Square; PYPL – PayPal] Comparing Business Models and Strategies

PayPal

From its origins in peer-to-peer transfers, PayPal has evolved into the ubiquitous online payment gateway that it is today.  Consumers use PayPal because it is a secure, reliable, and fast payment method that is widely accepted by merchants; merchants accept PayPal because it eases checkout friction for ~240mn potential customers, converting 90% of checkout visits to transactions vs. less than half for other payment options.  One Touch, which enables PayPal users to buy goods across online, offline, mobile channels without entering payment credentials in each instance – so you aren’t squinting, pinching, and expanding a smartphone screen as you tap in your name, address, and credit card details – has been a major driver of spend propensity and perhaps the single most significant flywheel accelerator in recent memory.  Just 4 years since launch, over 40% of PayPal’s 240mn users and nearly half of its 19mn merchants have enabled One Touch.  A comScore panel of 1mn consumers found that not only did 47% of PayPal shoppers make more purchases than non-PayPal shoppers, but that among PayPal shoppers, 55% of OneTouch adopters made more purchases than non-adopters.  Engagement, moreover, has generated tons of proprietary data, enhancing PayPal’s security and underwriting efforts in an iterative loop.  For instance, the company fuses insights gleaned from the ~10bn transactions it sees each year with risk models used by card issuers in order to boost transaction approval rates, and analyzes the sales activity of merchants to underwrite loans, spawning still more transaction activity and more data.  Network effects all over the place.

That old bit about enduring sub-optimal short-term profits for the sake of bolstering long-term value is especially pertinent to two-sided payments networks, which require continuous investment to stoke engagement given the reflexive nature of merchant <-> consumer adoption.  This is especially true if you’ve got multiple assets and engagement channels that reinforce one another, as PayPal does.  Incremental adoption in one channel feeds growth in another, producing larger outcomes than you might initially expect.  For instance, P2P, which accounts for ~20% of PayPal’s TPV and around a quarter of its users, has its own one-sided network effects but is also an important on-ramp for commerce.  Two-thirds of PayPal’s most active shoppers also use PayPal for P2P transfers and those who come to PayPal through P2P are twice as likely to checkout with PayPal than those who don’t.  Initiatives like Choice and One Touch and Instant Transfers bring incremental same-store volumes, sure, but by fostering consumer engagement, they also compel more merchants to join PayPal’s network, and that growing merchant base can then be leveraged to make money on additional services.  Like Venmo.

Venmo was acquired in 2012 for just $26mn by Braintree¹, which in turn was bought by PayPal a year later for $800mn.  By my estimate, Venmo was processing less than $2bn of volume/year when PayPal acquired it vs. $40bn in the last 12 months:

If it were a standalone asset, Venmo would probably worth at least $4bn today, several times what PayPal paid for both Venmo and Braintree just 5 years earlier.  And that gives no credit for the knock-on benefit of One Touch, which evolved directly from Venmo Touch and has become the most rapidly adopted product in PayPal’s history.  So, after years of running Venmo as a free P2P transfer service, PayPal recently started monetizing Venmo via commerce (just as it did with core PayPal back in the day), rolling out Venmo acceptance across its merchant base.  There is almost no setup overhead for a PayPal merchant to turn on Venmo and ~2mn PayPal merchants in the US so far have done so, gaining access to a coveted millennial demo.  Any merchant with an integrated PayPal button can now run a separate Venmo button alongside it to accept payments, like so:

Venmo commerce transactions are still in early stages.  Nearly all of its $50bn TPV is P2P, but just as PayPal began P2P but now derives ~80% of its volume from commerce, Venmo has a sizable monetization opportunity ahead of it.

But Zelle, the real time P2P mobile payments service owned by a group of 60 US banks (up from 30 last summer), bears watching.  Zelle has the unfair advantage of being natively embedded in the mobile apps of its banking sponsors, so it is perhaps no surprise that, according to eMarketer, after growing 73% y/y, Zelle’s userbase of 27mn is projected to exceed Venmo’s 23mn in the US by the end of this year.

Per TechCrunch, Zelle’s total payment volume grew by 45% y/y in 2017, to $75bn, well exceeding Venmo and approaching the ~$100bn of P2P volume across PayPal + Venmo.  Dividing volumes into users suggests that Zelle users transfer more than twice as much money as Venmo users, so Zelle’s user numbers aren’t just a flattering zombie stat.

Despite PayPal’s track record of putting its set of advantages to good use, the company has not always been granted the benefit of the doubt.

Like that time 2 years ago when PayPal rolled out “Choice”, partnering with card processing networks and issuers to make it easier for consumers to select Mastercard and Visa credit and debit cards as their funding instruments inside the PayPal wallet.  Investors sort of freaked because these relatively more costly funding instruments were expected to deplete PayPal’s transaction profits.  But it turns out that PayPal users really wanted this feature [evidenced by the 40mn users who eventually set preferences under Choice] and 70% of Choice adopters selected lower cost funding options, ACH and PayPal balances, anyways.  It has also so far proven a significant engagement lever, with adopters experiencing lower churn [PayPal has seen 2mn fewer customer contacts from Choice adopters over the last year], larger average basket sizes, and 10% greater spend compared to non-adopters.  Banks, too, are happy because their cards are now more likely to be used in online/mobile transactions, and a card wielding customer is twice as likely to complete an online purchase when PayPal is offered as a checkout option.

Or how about earlier this year, when eBay, which is currently using PayPal as its exclusive payment processor, announced that it would be running most of its transactions through Adyen by 2021.  Today, regardless of whether you pay with a “branded” PayPal option or with a credit or debit card, those funds make their way to the PayPal account of the individual seller from whom you are buying.  Under this new deal, Adyen will process unbranded (i.e. credit and debit card) transactions and PayPal payments will be funneled to eBay’s PayPal account.  In either case, eBay, not the seller, is now the merchant of record [the party responsible for returns and chargebacks initiated by the customer].  As a consolidator of payment streams across individual sellers, eBay gets a processing volume discount.  PayPal shed ~9% of its market cap the day this news broke as some speculated that the company would experience a “big blow to its total payment volumes“.

But as it turns out, the consequences are far more innocuous.  First, since its separation from eBay, PayPal’s eBay revenue has grown at a fraction of the rate of its non-eBay revenue.  By the end of the Operating Agreement, PayPal’s management expects eBay to account for just 4% of TPV, which maybe translates into a high-single percent of profits.   Second, the “branded” part of the eBay relationship has been renewed through 2023 (though I suspect at lower unit profits than before).  And the far less significant unbranded processing volumes that are falling away apparently yield little to no profit for PayPal.  Third, and far more importantly, management claims that had the Operating Agreement been extended in full, PayPal would have been disallowed from acting as the merchant of record for the “largest and fastest growing marketplaces in the world” (referred to as “Competitive Platform Operators” in the OA), which I think we can infer to mean Facebook, Google, and so forth.  That interpretation, by the way, appears saturated with positive spin.  Under Article XIV, “Merchant of Record; Non-Compete”, the OA actually states:

“If PayPal provides payment processing services to a Competitive Platform Operator as a Merchant of Record for third party merchant transactions effected by or on such Competitive Platform Operator’s platform, eBay shall also be permitted to become a Merchant of Record for transactions effected by eBay Merchants on eBay Covered Properties”…

So, PayPal’s claim is that the Operating Agreement restricted PayPal from engaging with competitive platforms and that it, PayPal, chose not to extend in order to pursue those platforms.  But, it seems a more accurate read is that PayPal could have chosen to provide processing services to a competitive platform as an MOR but had it done so, eBay could have also assumed the role of MOR for its own properties.  Under that interpretation, eBay was the restricted party, not PayPal.  Also, even with the OA restrictions, PayPal processed $65bn of volumes from small businesses selling on the top 20 platforms outside of eBay, a figure that is not only greater than eBay volumes but growing substantially faster too (40%+ y/y vs. high-single digits). 

I don’t know that the difference matters all that much now, except that it maybe begs the question of why only now, with the OA was reaching its senescence, did PayPal deny itself the opportunity to strike expansive agreements with other significant platform operators just to keep a business that, in management’s own words, was “undifferentiated” and “commoditized”?  Well, the Operating Agreement doesn’t delineate branded from unbranded volumes.  So perhaps had PayPal “violated” the spirit of cooperation with eBay and chosen to MOR with competing platforms while the OA was in effect, it would have risked losing the profitable branded part too, which it is now keeping under a separate agreement after having been a good partner over the OA’s 5-year term.  Just speculating.  But, that said, given PayPal’s ubiquity on the consumer side – management suggested that it had maybe 50% share of eBay’s checkout – and the 2x higher checkout conversion rates seen by PayPal adopting merchants, I find it highly unlikely that eBay, as an MOR, would have disallowed the branded PayPal gateway in any case.

PayPal has long outgrown the online marketplace platform of eBay that had taken it beyond P2P and was so critical to its early growth.  Payments is a horizontal function that cuts across online and offline properties, carried along by the purchasing behaviors of consumers, whose channel preferences have proven fluid and ever expansive.  As such, it made little sense for PayPal to wed itself to a single online marketplace when it could better optimize value by serving as an unrestricted neutral third party platform for a whole universe of merchants across multiple venues.  And sure enough, PayPal has expanded its relationships with Google and Facebook, making it easier for users to shop wherever they happen to find themselves (“contextual commerce”)…so, by linking your PayPal account to Google Pay, you can fund purchases on YouTube and Google Play, or make P2P transfers in Gmail.  In Facebook Messenger, PayPal users can fund P2P transfers while sellers’ bots can accept PayPal payments and deliver invoices.  Baidu’s ~100mn+ wallets are accepted by PayPal merchants in the US.