Mastercard beefs up its defenses against first-party fraud

Mastercard is building more capabilities around fraud detection as part of a strategy to win more business from companies that might otherwise turn to fintechs for the same services.

"We want to enable transactions to be more successful," said Chris Reid, head of identity solutions for Mastercard.  "First-party fraud has become a big problem and we know we can eat into that." 

In addition to competing against fintechs, Mastercard is also in a tech arms race with Visa, which is similarly upgrading its fraud-detection capabilities. For Visa and Mastercard, managing identity for fast-moving digital commerce transactions is also a way to demonstrate utility beyond payment processing. 

"Battling identity fraud is a way to measure the effectiveness of our services," Reid said. 

Mastercard is adding Microsoft's artificial intelligence to the card brand's identity risk technology to identify suspect user behaviors early. That would provide a better chance to authorize transactions accurately as well as mitigate chargebacks and refunds. 

Identity fraud risk is accelerating, reaching $24 billion in 2021, a 79% increase over 2020, according to a March study from Javelin Strategy & Research. The number of adults in the U.S. impacted by identity fraud grew more than 50% to more than 15 million, with credit card-related identity fraud increasing 69%. The impact of identity fraud expanded, with the per victim loss increasing from $1,350 in 2020 to $1,551 in 2021. 

First-party fraud, which cost $50 billion globally in the past year according to Mercator Advisory Service, is also an emerging threat. 

Also called "friendly fraud," first-party fraud occurs when a consumer makes a legitimate purchase but then tries to collect the funds through the chargeback process. 

"First-party fraud changes the dynamic for fraud prevention," Reid said. "Traditionally fraud prevention was about proving it wasn't you who made the fraudulent transaction. Now we're trying to prove it was you who made the bad transaction, but still minimize friction." 

Mastercard combines its network data with the merchant's payment data to confirm the identity of the shopper, using AI to examine behavioral patterns that suggest a dishonest chargeback. 

Mastercard's partnership with Microsoft includes projects that go beyond identity risk to fuel the card brand's services strategy. Over the past several years, Microsoft and Mastercard have teamed on cloud research, augmented reality, small-business services and other security functions such as anti-hacking tools. 

"It's a multiyear and enduring partnership, and we're collectively trying to solve the same problems," Reid said. 

It's also part of Mastercard's broader expansion into technology-powered services for issuers and merchants, which CEO Michael Miebach has detailed in numerous recent earnings calls. In the most recent call, Miebach referenced the card brand's acquisition of Dynamic Yield, an AI company Mastercard purchased from McDonald's in 2019, as a way to improve the use of purchase data to produce personalized marketing and other consumer engagement. 

Visa, which did not return a request for comment by deadline, combats first-party and other identity fraud through a mix of identifying suspicious activity patterns, accessing the likelihood a person is who they say they are through layered authentication, and comparing data input by the consumer at the point of sale against external data sources. 

Stripe has a similar tool called Radar, which is built directly into payment flows and uses machine learning to detect patterns across transactions, assessing risk levels for each payment. 

"We see alternative payment processors branching out into providing fraud management tools," said Andras Cser, vice president and principal analyst at Forrester Research in Lexington, Massachusetts. That puts pressure on the card networks to up their own digital identity capabilities, he added.  

"This trend allows Mastercard and Visa to bring in their own products such as Visa's Cybersource and Mastercard's CypherTrace as well," Cser said.