PYPL: Credit Risk and Disclosure

nicklipinski
Print Friendly
Share on LinkedIn0Tweet about this on Twitter0Share on Facebook0

SEE LAST PAGE OF THIS REPORT FOR IMPORTANT DISCLOSURES

Howard Mason

203.901.1635

hmason@ssrllc.com

May 3, 2016

PYPL: Credit Risk and Disclosure

“We are not in credit for credit’s sake; we are in credit because it drives both consumer benefits and merchant benefits and the overall flywheel [i.e. network] effect” PYPL CEO Dan Schulman, Apr 2016

  • We believe PYPL is a juggernaut in the payments industry and will ride the integration of payments into merchant pay-and-save apps for use by consumers both online and offline to 25% volume growth for the foreseeable future (see note of March 20th titled “PYPL: Juggernaut”). And we understand the role of credit, even if driving only 2% of total payment volumes, in supporting firm’s merchant value-proposition by reducing cart abandonment costs and raising average ticket-size. However, we believe PYPL’s disclosure practices can make it more difficult for investors to assess credit quality and trends in the resulting loan portfolio.
  • While PYPL is pursuing an “asset-light” approach to its credit business, leveraging the funding and risk capacity of SYF for the PayPal-MasterCard cobrand card for example, the loan book[1] associated with PayPal Credit (for consumers) and PayPal Working Capital (for merchants) grew nearly 20% year-on-year and now stands at $4.2bn, net of the loan loss allowance, to represent ~30% of equity. Furthermore, we estimate fees and finance charges on the loan book contributed ~one-third of year-on-year growth in “value-added” revenue.
  • In 2016Q1, PYPL disclosed for the first time the net loss ratio on the consumer book (representing nearly 90% of total loans) as stable year-on-year at 6.1%. This suggests inferior credit quality to both SYF’s loan book and COF’s domestic card book, with 2016Q1 net loss ratios of 4.7% and 4.2% respectively, and PYPL appears under-reserved relative to SYF given reserve-to-loan ratios of 5.7% vs. 5.5% respectively. More importantly, as is standard in the industry, this net loss ratio includes only principal amounts that are deemed uncollectible and not finance charges and fees so deemed; these finance charges and fees are significant raising total charge-offs on the consumer book in 2016Q1 to $97mm, net of recoveries, versus ~$70mm implied by the principal charge-off rate.
  • The difference matters because the transaction loss rate which, along with the take rate and transaction funding rate, is followed by investors (Chart 1) includes charge-off amounts for both principal and finance charges/fees. Furthermore, the bundling into the transaction loss rate of loan losses with fraud losses, along with the presentation as a ratio of payment volume, tends to reduce visibility into credit trends. For example, commenting on the year-on-year increase in the transaction loss rate to 31bps from 28bps, CFO John Rainey noted it was “a result of increased [fraud] risk pressures that we saw in the quarter as opposed to anything that we saw specific to credit”. This is true in the sense that the loan loss rate (expressed as a percentage of payment volume) increased only 0.8bps in 2016Q1 from the prior-year quarter while the fraud loss rate increased 2.4bps. However, if expressed more conventionally as a percentage of average balances, the loan loss rate increased to 8.8% from 7.5% (Chart 2) and more obviously merits attention; in particular, given the stability of the principal net loss ratio for the consumer book at 6.1%, it raises the question of why more finance charges/fees are being reversed out.

Chart 1 – The Transaction Loss Rate at PYPL (orange text)

Source: Company Reports, SSR Estimates.

Chart 2 – Loan Loss Rates at PYPL as % of TPV and Loan Balances

Source: Company Reports, SSR Estimates.

Overview

Credit is an important part of PYPL’s business model albeit not in the sense of driving financial results, as in the case of a traditional lender, through net interest income. As CFO John Ramey notes, “if you were to just take the growth rates of credit and the rest of our business, it would take a couple of decades before credit even got to 20-25% of our overall revenue”. Still, we estimate fees and finance charges on PYPL’s loan book generated ~one-third of the year-on-year growth in “value-added” services revenue in 2016Q1.

Regardless, consumer credit (through PayPal Credit and the MasterCard co-brand) drives spending on the PayPal network with the firm reporting that the average purchase for a credit customer is 20-30% higher larger than for a non-credit customer (after adjusting, presumably, for social and economic circumstances) and reduces cart abandonment; both effects are important to PYPL’s merchant value-proposition of lifting conversion rates and sales. And merchant credit (through PayPal Working Capital) provides direct support to some merchants in financing their businesses.

For this lending business, PYPL is committed to an “asset-light” model which leverages the funding and risk-capacity and, in the US, works with third-parties to act as lender-of-record since it is neither a chartered financial institution nor licensed to make loans in any State. These third-parties are ADS (for PayPal Credit and PayPal Working Capital) and SYF (for the PayPal-MasterCard co-brand card). PYPL purchases PayPal Credit or PayPal Working Capital loans from ADS after ADS has advanced funds to the consumer or merchant respectively, and consequently has built up a loan portfolio of $4.2bn which has grown nearly 20% year-on-year (Chart 1). PYPL used to purchase loans from SYF as well but, in 2015Q2, sold loans back to SYF and amended its relationship so that it was not required to make on-going purchases so that we believe the current consumer loan book is substantially all related to PayPal Credit.

Chart 1 – The Consumer and Merchant Loan Books at PYPL

Source: Company Reports, SSR Estimates. Numbers may not total exactly because of rounding/reconciliation errors.

These loan balances represent ~30% of equity and the question confronting investors is the extent to which they represent a credit risk and can impact valuation. Beyond emphasizing that PYPL is not in the credit business for credit’s sake and committing to an asset-light model, management has provided assurances that it does not expect credit losses to have a material impact on earnings. In the Q1 call, John Rainey stressed that PYPL is targeting a steady earnings stream and enjoys “a lot of ways that we can manage any volatility irrespective of whether it comes from credit or other aspects of our business” and that the year-on-year increase in the rate for transaction and loan losses was “a result of increased risk pressures that we saw in the quarter as opposed to anything that we saw specific to credit”. To back up this last assertion, PYPL disclosed the net credit loss rate on its consumer book for the first time and it was stable year-on-year at 6.1%.

Analysis

While PYPL does provide a FICO breakdown of its consumer loans, it unfortunately does not use the same rubric as peers. We know that 49% of loans to customers with FICO scores are to customers with scores of less than 680 but this is not directly comparable to the disclosure of, say, SYF that 28% of credit-card loans and 22% of installment loans are to customers with FICO scores of 660 or less or with no FICO score or of, say, COF that 34% of domestic credit card loans are to customers with refreshed FICO scores of 660 or less. PYPL explains in its reports that FICO 680 is generally considered “prime” by the consumer credit industry which is true but not as helpful as it could be; the standard used by large peers for sub-prime is a FICO 660.

The net loss ratio suggests that PYPL’s loan book is of inferior loan quality to both that of COF’s domestic credit card book and SYF loan book (with a 2016Q1 ratios of 4.2% and 4.7% respectively). Furthermore, PYPL’s allowance for loan losses, at 5.7% of loans, compares with 5.5% for SYF despite the difference in 2016Q1 net loss ratios (6.1% versus 4.7%). Finally, as is standard in the industry, PYPL’s net loss ratio consists of unpaid principal on loans that are deemed uncollectible and does not capture finance charges and fees that are deemed uncollectible and that, we estimate[2], represent ~30% of total net charge-offs.

Our broader concern, however, is that PYPL’s headline disclosures around the transaction expense rate and the transaction margin blur the lines between fraud losses and credit losses. We understand PYPL is in the credit business to boost network-spending rather than for its own sake but the management insists credit drives only about 2% of payment volume so that it seems odd to represent the cost of credit losses across the entire payment volume (Chart 2) particularly given that credit losses are driven by loan balances accumulated over time rather than the payment volume in any particular period. Hence, for example, a deterioration in credit quality will have a disproportionately adverse impact on the transaction expense rate (i.e. transaction expenses as a proportion of payment volume) which is typically compared to the take rate (i.e. revenue as a proportion of payment volume) to analyze the dynamics of the transaction “margin” (Chart 2).

Chart 2 – Typical Presentation of Transaction Margin Dynamics at PYPL

Source: Company Reports, SSR Estimates. Numbers may not total exactly because of rounding/reconciliation errors.

PYPL is properly downplaying the transaction margin as a performance metric given the pressures arising from a mix in payment volumes to P2P services (such as Xoom and Venmo) from C2B payments and, within these “merchant” payments, from small to large merchants. For example, in the last report, John Rainey comments that “as a management team, we are focused on margins, our generation of free cash flow, and growing EPS and not getting focused on just one aspect of the P&L being the transaction margin”. However, the funding and fraud costs of transactions, and the credit losses on loan balances, are important financial drivers particularly for large merchants with the buying power to negotiate a below-average take-rate.

More specifically, it is easy to lose sight of credit costs given they are bundled in with fraud costs to generate a transaction loss rate that is small relative to the transaction funding rate (31bps of payment volume in 2016Q1 versus 93bps) and appears relatively stable (up 3bps year-on-year). The tendency from the disclosure format to underplay changes in credit performance is illustrated by John Rainey’s remark that the rising transaction loss rate was “a result of increased risk pressures that we saw in the quarter as opposed to anything that we saw specific to credit”. This is true in the sense that the loan loss rate (expressed as a percentage of payment volume) increased only 0.8bps in 2016Q1 from the prior-year quarter while the fraud loss rate increased 2.4bps. However, if expressed more conventionally as a percentage of average balances, the loan loss rate increased to 8.8% from 7.5% and more obviously merits attention.

Chart 3 – Loan Loss Rates at PYPL as % of TPV and Loan Balances

Source: Company Reports, SSR Estimates.

©2016, SSR LLC, 1055 Washington Blvd, Stamford, CT 06901. All rights reserved. The information contained in this report has been obtained from sources believed to be reliable, and its accuracy and completeness is not guaranteed. No representation or warranty, express or implied, is made as to the fairness, accuracy, completeness or correctness of the information and opinions contained herein.  The views and other information provided are subject to change without notice.  This report is issued without regard to the specific investment objectives, financial situation or particular needs of any specific recipient and is not construed as a solicitation or an offer to buy or sell any securities or related financial instruments. Past performance is not necessarily a guide to future results. The analyst principally responsible for the preparation of this research or a member of the analyst’s household holds a long equity position in the following stocks: JPM, BAC, WFC, and GS.

  1. Given PYPL is not a chartered financial institution in the US, nor licensed to make loans in any State, the lender or record for PayPal Credit and PayPal Working Capital are the “Comenity” banks subsidiaries of ADS. However, PYPL purchases the loans from ADS after they have been made. The lender of record for the PayPal-MasterCard cobrand is SYF and, until 2015Q2, PYPL had been buying loans from SYF as well; in 2015Q2, however, SYF bought back these loans and released PYPL from the obligation to buy loans on an on-going basis.
  2. PYPL discloses net charge-offs for 2016Q1 at $102mm while principal charge-offs are ~$70mm if we apply the disclosed net charge-off rate of 6.1% to average gross loan balances of ~$4.4bn.
Print Friendly