Description

Long DFS - Discover at $112,  short SYF - Synchrony Financial $36

Summary

Discover - DFS is an attractive long on a relative and absolute basis, and I think it’s an attractive pair trade vs. Synchrony - SYF.  

DFS held in quite well on pretty bad guidance and earnings on much higher credit losses and DQs than expected.  It’s usually a good sign when stocks don’t go down on rather bad news anymore.  

DFS (and most consumer lenders) have all guided and reserved for a 'mild' recession - which they basically assume is ~5% unemployment and around 0% to slightly negative GDP growth in 2023.    Of course, unemployment is the key macro metric for consumer lenders, since people need jobs to pay their debt. 

As you're all well aware, unemployment has remained remarkably strong, and the consumer overall still has much in savings (yes the lower end is getting squeezed from inflation, etc.),  but on the whole, the consumer and credit metrics are holding up well. Discover skews more prime, and so their customers are less sensitive to inflation.  DFS is more a general purpose card, which borrowers tend to prioritize in paying during tough times.  DFS is 85% prime, 15% subprime

Synchony on the other hand is more subprime (24% Subprime) and in a recession, people usually don’t care about defaulting on their TJ Maxx credit card, Gap, or Lowes credit card, so they take higher losses. 

Indeed, credit metrics are 'normalizing' after the abnormally great times during the COVID era when stimulus and savings kept building up on consumer balance sheets

Also, the Consumer Financial Protection Bureau (CFPB) is going after late fees.  Basically they want to cut it down to $8 from $40.  It hard to get an exact picture,  but for DFS late fees are only 3% of revs and ~5-10% of EPS.  For SYF, late fees going to $8 could possibly be on the order of a 30-60% of EPS (although there may be mitigating offsets

And yet, DFS and SYF trade at similar multiples - SYF is at 7.3x 2023 P/E, DFS at 8.2x, despite SYF being a worse biz as well with a worse track record (so it deserves a lower multiple).  

Discover also has a lot of excess capital and has recently restarted its buyback (after suspending it during a student loan investigation).  They have 13.3% Common Equity Tier 1 capital ratio, vs. a target of 10.5%  They have plenty of room to increase share buyback andf dividends.  This is roughly a $4bn capital surplus,  vs. a market cap of $30bn. 

Meanwhile,  SYF is already done a lot of capital return and is down to 11.8% tier 1 common equity, vs. its target of 11%.  So they’re already pulled that lever

Also, DFS has steadily created shareholder value and grown book value and EPS over time, because of their niche in prime revolving credit card customers.  E.g. book value per share has gone from $11.59 in 2006 to $54.64 in 2022.   EPS has gone from $2.27 to $15.50.  That’s quite impressive growth in what is supposed to be a pretty bland biz. 

On the other hand, SYF has grown as much as DFS because they have a lot of competition from other credit card issuers who want to get co-branded credit cards, and they have little bargaining power from their large retail partners (e.g. they lost Wal-Mart in the past).  Their 5 largest retail partners are Belk, Fleet Farm, JC Penney, Sam’s Club, and TJX companies.  Clearly, it’s not great that SYF has a lot of exposure to brick and mortar retail either.  Their 5 largest partners overall are Amazon, JCP, Lowe’s PayPal, and Sam’s Club, which in total account for 53% of their total interest and 52% of loan receivables.   Clearly, these partners have a lot more bargaining power and can switch to someone else who is willing to the volume

DFS’s guidance for 2023 is also notably more conservative than SYF’s: leading to more opportunities for DFS to beat and SYF to miss. 

• DFS guidance this year around NCOs is also much more conservative.  DFS is guiding to NCOs of high 3%s range in 2023, which is the higher end of their historical range.  

• SYF is guiding to NCOs of actually 4.75-5% , which is below their through the cycle NCO target of 5.5-6%

DFS is also assuming 5% unemployment in their models and guidance, vs. SYF assuming 4.2% unemployment

Upside / downside / targets

So I like DFS as a long here, and SYF as a short, either individually or as a pair trade

For DFS, I think there’s around 30%-35% upside, using 

• 10x P/E  (more middle of the historical range) on 2024 EPS ot $14.50 (EPS should bounce back after 2 years of depressed earnings from rising NCOs and provision / NCO headwinds) - that’s 30% upside.  

• Plus around 10% more upside from excess capital return

• Plus a 2.1% dividend

• Minus 10% hit from what the CFPB may do

For SYF I think there’s 30%+ downside, mainly from risk to losing late fees from whatever the CFPB does.  They pay a 2.5% dividend
Charts

DFS since 2007

SYF since IPO 

Clearly,  DFS has done a good job creating shareholder value and stock price performance over time and it shows up in the stock prices and fundamentals. 

This has to do with DFS having somewhat of a niche in prime credit cards (similar to ALLY having a niche and leading position in prime auto loans). 

SYF on the other hand is in a more competitive space of private label credit card, where the merchant partners have a lot of bargaining power.  Synchrony likes to tout that their partners have been with them for many years (their 5 largest partner average tenure is 15 years and Lowe’s is 43 years).  But when these contracts come up for renewal, they are having to give up more and more economics to partners.  And despite these long-standing relationships, they’ve lost a lot of marquee clients.  E.g. in 2021, SYF lost Gap, who was a partner for more than 20 years, to Barclays.  In 2018, SYF lost their Wal-mart partnership to CapitalOne.   Clearly, other big banks are eager to undercut SYF and get large partners. 

Financials

Overtime, DFS has actually done a great job growing and creating value.  

• EPS has gone from $2.27 in 2006 to $15.50 in 2022. 16% CAGR

• Book value per share has gone from $11.59 in 2006 to $54.64 in 2022.  13% CAGR.  

• In normal times, they steadily generate 20-30% ROEs

On the other hand, SYF has been held back due to large retail partner losses, and continually having to give retail partners more economics.  This is reflected in their Retailer Share Arrangement, which keeps rising, from 3% of loans outstanding to 5% now

• EPS has gone from $2.81 in 2012 to $6.15 in 2022,  9% CAGR

• Book value has grown from $8.45 to $27.87,  14% CAGR

• ROEs are usually more in the teens to low 20%s

Valuation 

DFS valuation is on the lower end of its historical range of 8x NTM P/E (vs a usual 7-12x)

And P/B of 2x, vs. usually 1.75-2.5x.   

So there’s multiple expansion upside as well.  

It’s a decent entry point, but you can probably get a better entry point on dips given how volatile the market is

DFS P/E since 2010

Since 2007

DFS P/B since 2007

Synchrony valuation is also on the lower end of the range,  but given the massive risk to earnings, that is seemingly not being priced in.  Hence, I think it’s a good hedge / relative short to DFS.  

Again, SYF EPS hit to late fees could be around 30-60%,  it’s a very wide range since it’s hard to tell how their retailer share agreements (RSAs) can offset it.  DFS has relatively minimal exposure (maybe around <10% EPS hit) to late fees. 

SYF NTM P/E since IPO in 2014

SYF P/B

SYF’s lower P/B is from their lower ROEs. 

Also, keep in mind DFS has a lot of excess capital ~14% vs. their target of 10.5%

SYF has already deployed a lot of their excess capital over the years.  

This means DFS will have a lot more capital return, buybacks, and dividend upside to support shares 

SYF is also assuming around 4.2% unemployment their guidance, but their NCO guidance wasn’t that high (actually 4.75-5% , which is below their through the cycle NCO target of 5.5-6%, while DFS is guiding at the higher end of their historical range.  So SYF’s guidance is less conservative than DFS

DFS on the other hand gave very conservative guidance and relatively high NCO guidance relative to their historical levels and 5% unemployment.   So I think SYF is more likely to miss this year, while DFS has more room to beat. 

CFPB Going After Late Fees

https://www.consumerfinance.gov/about-us/newsroom/cfpb-proposes-rule-to-rein-in-excessive-credit-card-late-fees/

This has been well known and in the works for a while,  but the initial announcement that the CFPB is looking to cap late fees to $8, (down from ~$41), could be a very big deal

It is still early in the regulatory process, and this may get watered down or tied up in courts.  

It’s likely that the CFPB will come out with a final rule by late 2023, so that’s a major catalyst for downside in SYF stock 

Under Dodd-Frank, the CFPB has been granted authority and has made annual adjustments to maximum allowed late fees.

Credit card companies say that late fees are a type of deterrent and a way of keeping people current on their credit cards and other borrowings.  If they reduced late fees, they claim this would increase defaults and losses broadly, causing lenders to pull back, and cause a diminishing of consumer credit.  

Also, credit card companies would likely have to try to make this lost revenue & earnings in other ways, like lower rewards, or higher interest, etc.   So in a roundabout way, the industry will figure out something to get their ROE to be pretty similar.  However, it it’ll take a long time, while the possible hit to EPS from lower late fees would happen first. 

Estimates for the impact of a reduction in late fees are all over the place,  but generally in the same ball part.  For DFS and AXP it’s pretty minimal. 

For SYF it could be huge:

However, supposedly 60% of the late fees flow through to retailer partner revenue shares and they could be allowed to revisit pricing changes.  That being said, clearly it’s a very big risk for SYF, which arguably isn’t baked into SYF’s current multiple

Borrower Credit Quality 

Clearly, credit quality is normalizing from abnormally low levels as stimulus put a ton of savings in consumer bank accounts over the years. Now inflation and a weaker economy is normalizing those loss rates

DFS guided 2023 to be 3.5-3.9%, i.e. back to historical highs.  This is a positive in an ironic way since it’s basically 1) conservative; 2) back to prior peaks,  so arguably already guiding to a pretty weak outcome.   

DFS has a history of being conservative with guidance, so it’s likely going to beat over the year. 

Also, they’re guiding NCOs higher mostly because they had a lot of loan growth over the last 2 years originated during looser times, and now those receivables are seasoning,  so optically it looks worse than it is (i.e. growth math).    With cyclical industries, like consumer financials, usually you want to buy them when things look very bad and much bad news is priced in.  I think we’re closer to much of the bad news priced in

Company background

 Discover - DFS is one of the largest credit card issuers in the US.  They focus on prime revolvers, i.e. prime FICO score borrowers who borrow on their credit card

This is an interesting niche.  Most other credit card companies either focus on subprime (Capital One, pay day lenders, the private label guys), or they focus on prime transactors (e.g. AXP, the large banks etc.). 

Because of their niche, they have relatively less competition, and rewards inflation is more muted for Discover (only 3 bps increase a year) than it has been for other card issuers.   They do a lot of balance transfer promotions (which come with 3% fees) and low initial APRs. 

Their rewards are pretty competitive, but again they’re not trying to be the best travel card or rewards card.  E.g. they have 5% cash back on categories, or 2% card on gas stations and restaurants,  or a 1.5x miles card.  Pretty standard

They also cater to students as well, and get them while they’re young

While the ‘prime revolver’ space  does seem a bit of an oxymoron,  clearly there are a lot prime rated people who still have a balance on their credit card, and Discover has had a history of developing products, offers, and marketing to attract these customers. 

Discover has

• $90bn in credit card loans

• $10bn in student loans

• $8bn in personal loans

The Discover Network has 50mm merchants, and 2.2mm ATMs around the world

They did $550bn in total network volume in 2022. 

In the US, Discover has basically the same acceptance (10.6mm) as Visa and Mastercard (10.9mm)

Internationally though, Discover doesn’t have as much coverage 

Discover cards are frequently considered some of the best in rewards and has good brand perception. 

—------

Balance Sheet / Capital structure

$108bn in loans (at 13.5%)

$20bn in other earnings assets  (3.4%)

$67bn in consumer deposits (64% of total funding), at 2.5%

$19bn in brokered deposits, at 3.5%

$19.5bn in borrowings, at 3.8%

Total interest funding cost 3%

DIscover has a pretty simple, no frills online banking.  But it’s somewhat unique in giving 1% cash back on debit purchases as well

Currently, it’s offering 3.5% interest on online savings.  As the Fed’s rate rise slows, deposit costs will abate.  Many banks are already saying that deposit outflows have stabilized recently, and with Fed Funds nearing a peak, the pressure here will abate. 

Discover, like other card companies benefit from higher rates  (NIMs rise) as loan yields riser faster than their liability costs

Recent Quarter (Q4 2022) 

The big negative surprise in recent earnings was 2023 NCO guidance notably above consensus.  2023 NCO of 3.5-3.9%, was well above cons at ~3%. 

However, DFS is usually conservative,  they guided 2022 NCOs to be 2.4%, and it came in at 1.8%,  so that's a mitigating factor

Basically, DFS is mostly blaming the large increase in NCOs from all the loan growth they had over the last 2 years (~30% loan growth), which are now going through the normal seasoning process and hitting peak losses in year 2 as usual.  I.e. ‘growth math’ or the typical higher NCOs following loan growth. 

They do see some mild deterioration in low credit quality borrowers from inflation.  But their prime borrowers are holding up fine since employment is OK

They didn't take up their allowance for loan loss reserve rate because they don't see credit quality worse than expected and macro outlook hasn't changed / worsened yet.

In some ways, this is actually a positive.  DFS has had a lot of loan growth and hence has increased future earnings potential by a lot.  Yes there are some mechanically higher losses and DQs in the short term, but as those season through normal roll rates, it’ll come back down, and optically help EPS and credit quality on the other side.   So 2023 likely will be the a peak in NCOs and DQs (outside of an even worse recession than they’re already factoring in)

Interestingly, DFS stock held up quite well despite what many thought was a very bad quarter and very bad 2023 guidance.  Partly because this felt like kitchen sinking, and there are signs that their prime borrowers are holding up well 

Tidbits 

- loss rates are "normalizing" reflecting seasoning of new account vintages from the past 2 years, normalization of older vintages and mild deterioration and low credit bands, largely inflation-driven.

- has a lot of credit visibility for 1H 2023 from roll rates.  2H is less clear (depends on macro assumptions)

   -  "We have fairly significant vintages that are going through the normal seasoning process right now."

- Among prime revolver segment, don't see evidence of broader stress given the robust labor market

- assumptions:  "changes to employment conditions pose the most significant risk to our forecast. For the year-end 2022 reserve, our baseline assumption was unemployment in 2023 between 4.5% and 6% -- 5% and with alternative scenarios above 6%."

   - reserve rate didn't change because their their view of the macro environment didn't change

Credit metrics

Loan reserves 

Catalyst

- economy better than DFS guided to

- getting past DFS peak in charge offs 

- CPFB actions hurt SYF