DFS - Discover Financial Services

[siddharth18]

I was browsing SeekingAlpha when I stumbled upon one of the best and most researched thesis I've read so far on SA:

 

http://seekingalpha.com/article/1558232-discover-financial-services-set-for-growth-with-new-strategies-and-network-partnerships

 

We've see the remarkable appreciation in shares of V and MA in last few years enriching Todd Combs at Berkshire. The incredible toll-booth type, inflation immune, asset-light based business model produces supra-normal returns on capital and has a huge moat around the metaphorical castle.

 

Well, DFS - Discover Financial Services - is in similar business but it serves upscale clients compared to V and MA. More like AXP.

 

The nub of the thesis is that DFS is smaller compared to V and MA and is poised for a lot of growth (in emerging markets, B2B payments, PayPal partnership), has competent management with decent capital allocation plan and undervalued even using conservative assumptions.

 

Why are shares undervalued? The current P/E doesn't take into account the the strength of the business model or the growth that lies ahead. Even on comparative basis, they are grossly undervalued compared to V and MA.

 

Even if the business/industry of payment processing doesn't tickle your fancy, it's a thesis is very well researched so I suggest you read it.

[Guest wellmont]

dfs has been public for years. investors understand the "strength of the business model" very well. it's always traded at a discount to v and mc because it doesn't have as good a franchise. having said that it may well be attractive at these levels. but Discover financial has already been "discovered". :)

[siddharth18]

dfs has been public for years. investors understand the "strength of the business model" very well. it's always traded at a discount to v and mc because it doesn't have as good a franchise. having said that it may well be attractive at these levels. but Discover financial has already been "discovered". :)

 

Thanks for comment. Yes it has been "discovered" a while ago (a 10-bagger since the 2009-lows).

 

Could you elaborate on the "it doesn't have as good a franchise" ? The merchant acceptance gap has been shrinking after it won the antitrust case against V and MA. Plus, it has a strong brand loyalty and attracts the "high-end credit card demographic."

[CorpRaider]

Yeah I wish I had held it.  I sold after a triple. 

[Guest wellmont]

dfs has been public for years. investors understand the "strength of the business model" very well. it's always traded at a discount to v and mc because it doesn't have as good a franchise. having said that it may well be attractive at these levels. but Discover financial has already been "discovered". :)

 

Thanks for comment. Yes it has been "discovered" a while ago (a 10-bagger since the 2009-lows).

 

Could you elaborate on the "it doesn't have as good a franchise" ? The merchant acceptance gap has been shrinking after it won the antitrust case against V and MA. Plus, it has a strong brand loyalty and attracts the "high-end credit card demographic."

 

it doesn't have the brand or the scale. there are only so many cards people can use. they use visa mc and axp. there are limits. and winning and losing market share isn't going to happen much. it's mature. the players are pretty much set and in place. it's a good business. but I would buy it in a bear market. it's up 36% in a year. up 30% ytd. not exactly the time to be buying in my book.

[Philip Morris IV]

DFS is more comparable to AXP than V/MA because they issue their own cards vs. institutions issuing V/MA cards.  It's the closed- vs. open-loop model.  They can also lend to cardholders.

 

My take is in general, Amex's cardmember base is the crown jewel of this industry.  Their credit quality is supreme and they are big spenders.  People put cars, boats, jewelry on Amex cards.  And these cardmembers are valuable to merchants so as Discover may gain market share, it won't be at AXP's expense.  Discover has traditionally catered to the middle class whom get likely more value from not paying the annual fees of Amex cards.

 

You can probably reason that by virtue of being the smallest and least mature player, they may outperform, but it is likely they will always be the distant laggard.  They are up against some serious muscle.  I like this industry but they've all been on relentless tears the past few years.  The premium is warranted but I just can't pay it.

[kab60]

Anyone invested here? Seems cheaply priced: asset light, strong cf, high roe. Growth, buybacks and strong us economy seems like a free bonus. It is not a bargain per se, but in this environment it does seem cheap. What am I missing (other than missing the party the last 5 years?) Will dig deeper, according to transcripts it seems like there is a concern about how some of their loans will perform but without being very financial savy their chargeoffs doesnt seem bad - around 2-2,4 percent.

[no_free_lunch]

I think the biggest risk here is that the charge-off's are very low.  It's been a few months since I looked at them, but it seemed that much of the growth in earnings is just the charge-off rate dropping.  You need to figure out whether the current charge-off rate is above or below historical averages, as I recall it is below average.

[kab60]

Thanks, no_free_lunch. Charge offs seem to hover around 1,8-2,4 percent in the last couple of years, but I'll have to dig deeper: http://www.sec.gov/Archives/edgar/data/1393612/000139361214000048/exhibit99-1_113014.htm

(dec '13 - nov '14)

 

http://www.sec.gov/Archives/edgar/data/1393612/000139361213000006/exhibit991.htm

(jan '12 - jan '13)

 

[HJ]

I think the biggest risk here is that the charge-off's are very low.  It's been a few months since I looked at them, but it seemed that much of the growth in earnings is just the charge-off rate dropping.  You need to figure out whether the current charge-off rate is above or below historical averages, as I recall it is below average.

 

It is quite a bit below.  But that said, there are reasonable theories that suggest charge off rates may be structurally lower compared with pre-crisis.  Regulation forces them away from the lowest rung of sub-prime, and just in general be more cautious.  Card act has restricted their ability to re-price loan rates higher after getting an account, so you are significantly less likely to do teaser rates to gain shares.  And more importantly, consumers who are significantly more cautious in taking out the loan to start with, which results in a better charge off rates on the loans that does get taken out.  The manifestation is not just a lower charge off rate, but also a very slow industry balance growth rate. 

 

So it's growth vs. credit as far as trying to hit earnings expectations.  Earnings based valuation is reasonable, but P/B is quite high compared with the top tier names in banking, USB, MTB, etc.

 

Credit card lending is an excellent business in general, but I have this lingering question surrounding this odd duckling type of situation.  I'm guessing it's a slightly different demographic that choses to get a Discover card, rather than the other kinds of cards out there.  It's never great to be a significantly smaller #4 compared with V, MA and AXP.  I'm also guessing their retention rate may be lower than the other card types.  After college, the more premium borrowers "graduate" on to a different card.  But these are all guesses, haven't found hard numbers to support any of this.

[Jurgis]

I think HJ hits the nail on the head with questions about DFS 4th position in the market. I think a lot of investors are drawn to the company because it is cheaper than the other three. However, if there was a stock market drop, a lot would also flee to MA/V.

 

I have Discover card for ~20 years now. However, I only use it for their 5%-cash-back quarterly categories. At least they finally got onto a 1% cash back with Discover It. It was not competitive with sliding scale 0.25% to 1% cash back before.

 

Another issue with Discover is that it does not have international presence. MA/V/AXP are accepted everywhere (AXP maybe fewer places, but still). Discover does not have any international presence. So yet another business lost ... and a business where MA/V are growing a lot.

 

They might do OK, but comparable to them would probably be something like COF and not MA/V/AXP. Sure they have some advantages over COF - their own network. But they also lose to COF all the international business: COF zero-foreign-currency-charge makes them very attractive to use abroad.

[learner]

I think the biggest risk here is that the charge-off's are very low.  It's been a few months since I looked at them, but it seemed that much of the growth in earnings is just the charge-off rate dropping.  You need to figure out whether the current charge-off rate is above or below historical averages, as I recall it is below average.

It is quite a bit below.  But that said, there are reasonable theories that suggest charge off rates may be structurally lower compared with pre-crisis.  Regulation forces them away from the lowest rung of sub-prime, and just in general be more cautious.  Card act has restricted their ability to re-price loan rates higher after getting an account, so you are significantly less likely to do teaser rates to gain shares.  And more importantly, consumers who are significantly more cautious in taking out the loan to start with, which results in a better charge off rates on the loans that does get taken out.  The manifestation is not just a lower charge off rate, but also a very slow industry balance growth rate. 

 

Losses will certainly increase and in anticipation of this trend, a lot of companies have already started increasing their reserves (see last quarter earnings of Capital One, Chase and Discover).  Currently, losses are at an all time low due to some reasons mentioned here and due to a large proportion of balance transfer $s; due to the low interest rate environment, banks are handing out a lot of long teaser balance transfer.  This wont continue once interest rates increase.

[HJ]

 

Losses will certainly increase and in anticipation of this trend, a lot of companies have already started increasing their reserves (see last quarter earnings of Capital One, Chase and Discover).  Currently, losses are at an all time low due to some reasons mentioned here and due to a large proportion of balance transfer $s; due to the low interest rate environment, banks are handing out a lot of long teaser balance transfer.  This wont continue once interest rates increase.

 

I think this is where the debate on the stock is, how long does the current charge off rates last, how much lower is a post crisis "normalized earning" vs. the current run rate, offset by how much bigger can the portfolio get.  The stock trades mostly in line with consumer finance companies ala ALLY or SC.  The card network is basically thrown in there for free on current valuation metric.

 

I happen to believe the current charge off rates in the 2's is reasonably sustainable.  Even if one were to use 100 bps more as a "normalized" charge off rate, on a $55 billion portfolio, that's $550MM pre tax, $350MM +/- after tax lower in earning power, against 455MM shares, which is somewhere around 0.75 per share in earnings.  So you are at 13-14x this "normalized earning", rather than 12x headline.  The earning is pretty much all cash, and  doesn't have any gain on sale magic, with all securitized receivables properly reserved for.  Not a very aggressive metric in my mind, when the company is still growing loan balance around 5% a year and shrinking share count somewhere around 5% a year. 

   

The competitive intensity in the industry is nowhere near where it was in the late 90's, when you have First USA, MBNA, COF, Metris, Providian all aggressively pushing growth to meet their quarterly growth targets.  The industry has basically matured.  Balance transfers is much less emphasized today as a competitive tool vs. cash back rewards, which is targeting a more premium credit customer base to start with. 

 

It's not a V or MA or AXP, but also not trading like those.  I believe at the current metric, it's a solid investment in its own right.

 

[learner]

I happen to believe the current charge off rates in the 2's is reasonably sustainable.  Even if one were to use 100 bps more as a "normalized" charge off rate, on a $55 billion portfolio, that's $550MM pre tax, $350MM +/- after tax lower in earning power, against 455MM shares, which is somewhere around 0.75 per share in earnings.  So you are at 13-14x this "normalized earning", rather than 12x headline.  The earning is pretty much all cash, and  doesn't have any gain on sale magic, with all securitized receivables properly reserved for.  Not a very aggressive metric in my mind, when the company is still growing loan balance around 5% a year and shrinking share count somewhere around 5% a year. 

Thanks for posting the analysis.  I'd guess that normalized losses will be 100bps higher than the current losses, in line with your expectations so agree with your valuation.  Most card portfolios were at 8-9% loss rate in '09-10 and they are now ~2-2.5%.  Macro improvements contributed to about 40% of this improvement and another 60% to conservative underwriting.

 

What are your thoughts about their $4bn personal loans business?  Many issuers have tried this business in the past and couldnt make it work but Discover seems to keep pushing through.

[Spekulatius]

I have been following DFS for a while , but haven’t owned the stock since 2007 , but decided to buy a starter position after today’s drop. Itis interesting now this has evolved since the last post in 2015. Card losses are indeed up in thr 3% ish rate from the 2%, but it hasn’t hurt their performance. DFS has had ~25% ROE and bought back a significant amount of stock from ~450M shares in 2015 to ~315M shares now, while growing their receivables in the mid single digits.

 

It’s all unsecured credit of course and definitely will get squeezed in a decision, but the cash generation is quite impressive when times are good. I consider the stock cheap, even baking in some headwinds from CECL (current expected credit losses) accounting, which forces to build a reserve when a loan is created - my understanding as I am no expert.

 

http://www.rocketfinancial.com/Financials.aspx?fID=5705&pw=188223

[Gregmal]

Anecdotal perhaps, but Ive been told Discover has hands down the best, and most convenient balance transfer system. The cards I typically use send you bank checks every so often or offer temporary promos. Discover just has a continuous and rolling balance transfer option where you just click whether you want 12 months no interest with a 3% fee or 18 months @5% with no transfer fee. You can then just electronically enter the pay off account number or even send it to your bank account, effectively borrowing cash at those rates.

[Spekulatius]

Perhaps related to above, I feel that DFS could be a great takeover target for A Fintech company like  SQ. They would acquire a substantial customer base, a profitable business and Discovers Pulse network that they possible could use to nudge transactions over Pulse rather than Visa or MC and/or enhance it with their tech layer.

[abitofvalue]

Perhaps related to above, I feel that DFS could be a great takeover target for A Fintech company like  SQ. They would acquire a substantial customer base, a profitable business and Discovers Pulse network that they possible could use to nudge transactions over Pulse rather than Visa or MC and/or enhance it with their tech layer.

 

but they would also acquire $$$$$ of consumer loans and see their valuation get re-rated to crappy consumer fin valuations.  i agree DFS could be an interesting acquisition but other than a SYF type merger of equals dont see how/why it would be a target..  international networks looks for US presence would find it interesting but i doubt it would get regulatory approval. the big issue is their network is prob their most interesting asset (for a tech company)..but it has almost 100% of earnings from the lending business.  no tech company wants to deal with Fed regulation.. better to partner with an issuer and let them deal with that..

[Gregmal]

some reasonable insider buying here today

[Spekulatius]

some reasonable insider buying here today

 

$0.0 purchase price. These are all stock grants.

 

Edit: some market buys from the CEO and others today ~$74

[HJ]

I think among all the card holders out there, the ones who pick Discover are somewhat a self selected bunch.  On the one hand they have demonstrated better credit stats all else equal, on the other, I wonder if the brand identity is limiting the addressable market vs. Visa / Master / Amex.

 

DFS trade mostly as a lender.  The network aspect should have better valuation if that business were to trades like the other networks, but that valuation require a qualitative step up in its network business, which doesn't seem to be in the cards.  It doesn't seem to be the focus of the current or recent management. 

[Jurgis]

Spek raised interesting question: why wouldn't someone buy DFS for their network? I wonder why nobody bought them for their network in all these years though. It would seem that the network would be valuable for any large fintech. E.g. why wouldn't PYPL buy DFS?

 

 

[HJ]

Spek raised interesting question: why wouldn't someone buy DFS for their network? I wonder why nobody bought them for their network in all these years though. It would seem that the network would be valuable for any large fintech. E.g. why wouldn't PYPL buy DFS?

 

Let's say PYPL were able to disentangle the network aspect of the Discover business from its lending aspect somehow, and acquires it.  The question is after acquisition, what do you do with it, and how does the combined business benefit from the merger strategically or financially?  Visa and Master are unusual businesses to compete with, part of the reason they are afforded such rich multiples.  The consumer facing aspect of Discover makes it feel like it's just a smaller version of V/MA, but it really isn't.  Merchant acceptance is just one aspect of the competitive dynamic which is already commoditized.  Thinking through what sustains each of those businesses and how they compete in the market place, I think one quickly reaches the conclusion that maybe the DFS network is not afforded much value for good reason .

 

PYPL already has its own network, the virtual one on line.  Much smaller than V/MA, but significantly more profitable.  That's why it's market cap is 1/3 of Visa, even though its transaction volume is nowhere near that ratio.    It's kind of like looking at Wikipedia which is not run for profit, and asking whether Priceline should pay a search engine multiple for it on its traffic.  Is the combined company now a better competitor to Google, even if you just look within travel? 

[Jurgis]

Spek raised interesting question: why wouldn't someone buy DFS for their network? I wonder why nobody bought them for their network in all these years though. It would seem that the network would be valuable for any large fintech. E.g. why wouldn't PYPL buy DFS?

 

Let's say PYPL were able to disentangle the network aspect of the Discover business from its lending aspect somehow, and acquires it.  The question is after acquisition, what do you do with it, and how does the combined business benefit from the merger strategically or financially?  Visa and Master are unusual businesses to compete with, part of the reason they are afforded such rich multiples.  The consumer facing aspect of Discover makes it feel like it's just a smaller version of V/MA, but it really isn't.  Merchant acceptance is just one aspect of the competitive dynamic which is already commoditized.  Thinking through what sustains each of those businesses and how they compete in the market place, I think one quickly reaches the conclusion that maybe the DFS network is not afforded much value for good reason .

 

PYPL already has its own network, the virtual one on line.  Much smaller than V/MA, but significantly more profitable.  That's why it's market cap is 1/3 of Visa, even though its transaction volume is nowhere near that ratio.    It's kind of like looking at Wikipedia which is not run for profit, and asking whether Priceline should pay a search engine multiple for it on its traffic.  Is the combined company now a better competitor to Google, even if you just look within travel?

 

I think this is too simplified. And I think the comparison to Wikipedia is not good. Better, though not perfect, comparison might be Booking acquiring TripAdvisor - which it may or may not. 8)

[HJ]

There you have it, the answer is somewhere between Wikipedia and Trip Advisors.  But all this discussion is really quite moot.  They are not selling unless it's an extraordinary number.

 

The question is whether it's a buy currently as is.  I kind of want to say yes.  Comparing to the last time this thread was active in 2015, you have $20 billion more loans, 100MM less shares.  Book value per share up by 50%, stock price up by 25%.  We are also deeper in to the economic cycle.  But maybe it has less bearing to the consumer credit cycle, because "This time is different!".  There!  Now I've jinxed it.