ADS - Alliance Data Systems

[flesh]

I'm nearly positive there's another thread for this but couldn't find it.

I have a small position here and am still digging. Valuation, if you're okay with garp looks good, not great or could be great if there's a moat. Main reason it's not a larger position is because I can't figure out why card services has a moat or what it is, it's growth suggests a moat. I don't like taking large positions when intrinsic value is based on growth wo understanding the moat better, hopefully this will start a discussion.

Why there may be a margin of safety and some reasons for obfuscation:

-Actual debt is 6b cash is 1.9b, 9b ish of ev is tied to deposits and securitizations (non recourse), unless I'm wrong which wouldn't surprise me.

-significant buybacks last two years, exhausted as of q2 17

-mgmt alluded to paying off debt in the q2 17' ec as the buyback auth is exhausted or maybe some m & a. I'm assuming they pay of the 235m revolver due mid 18' and the 400m 17' senior notes for savings of 28m for CY 18'

-interest margin growth, I believe the following is from q1 17' ec transcript:

“Lastly to clarify the topic of raise in interest rates, yes the benefits Card Services made interest margin. The APR we charged is variable rate tied to primary, an increase in prime rate will reset the APR to the card holder within 2 billing cycles. Conversely funding costs which are about 70% fixed rate will reset over a two-year period.”

If I'm correct the 70% is based on 9b=6.3b. The june 25bps rate increase isn't yet reflected, good chance of another 25bps later this year and outlook is 3 more 25bps increases next year. That's a total of 1.25% if it happens. If one assumes a 1% increase on the 6.3 b you get an additional 63m flowing to 2018/19 pretax earnings.

-Closing the wedge: Collections were 50% outsourced in 16' and 50% in house. The recovery rates were 25% in 15'. The FH 17' recovery rates were 18%. As of recently, recovery has moved from 50% outsourced to 80-100% in house. Mgmt claims that the in house division is still collecting at a 25% rate hence the obvious move.

Annualizing the FH 17' provision for loan loss = 1.2b. At a 18% recovery rate you get 216m and at a 23% recovery rate you get 276m. If that's correct, you could add 60m to 2018 EBT.

-Mgmt claims that delinquencies foretell/track with charge offs, they put out a "wedge" last october based on this and thus far it has tracked perfectly with it. Because of delinquency trends they are claiming flat to lower charge offs in 18'. If flat, yoy growth will reflect.

-Absent charge offs, most of the company has been growing at a steady clip and mgmt is claiming 19% growth in core eps. Guiding to some core eps growth in 17' as well, back half weighted.

-Company pays 38% tax rate

I've valued this a few different ways and don't have a lot of confidence at this point that I'm accurate. Still messing around with it.

Here's a simple, crude, way to do it, bridging 16' with 18'.

2016 Ebitda = 1778m

+15% exclusive of below listed benefits, much lower than actual/guided growth 
= 2045
+28m interest savings
+63m net interest margin
+60m closing the wedge

= 2196m ebitda

-550m interest
-260m capex, mgmt claims capex is 3% of revenue and is guiding to 8.7b 18' rev

=1386m EBT

55.8m shares outstanding currently. Company pays 38%. Mgmt claims 1.1 fcf in 17' described as 200m divi, 500m buyback, 400m growth capex (q1 or 2 17' ec).

[no_free_lunch]

I am going to do some homework on this over the weekend.   I have had them on my watchlist for a few years.  Seem like a solid compounder.  Certainly getting more attractive at these levels.

In addition to buying back shares, if you look at their longer term history they have alternated between share buybacks and share issuance depending on the stock price, which is what you would hope for.  They don't just blindly buy back shares regardless of pricing level.  In 2014 calendar when their stock shot up, they issued another 25% shares for instance.  Then the price moved down and they have been repurchasing for the past couple years.

[vinod1]

I have taken only a brief look so could be wrong in my understanding:

- Their moat seems to be to be based on modest switching costs that its customers face along with being a low cost operator due to it having larger scale in its market segment (non-visa/mastercard/Amex/discovery) where there are the largest player.

 The only other player that I know of that is offering a somewhat similar service is Synchrony Financial. But they target larger customers like Walmart who have their own in-house marketing/analytics operations. ADS targets small to mid-sized where the companies do not have enough scale to do so effectively.

- I do not think it makes sense to look at FCF for a business which has nearly half its earnings from a credit card portfolio. When provisions are higher than charge offs, as is the case in the recent past, it results in high OCF and high FCF. But we know charge-offs and provisions are going to even out over the cycle. GAAP income is a pretty good indicator of owners earnings in this case.

- I would prefer to make bottom up adjustments from GAAP income statement as you know precisely what you are adding back. I think there are three main adjustments that need to be made to reported net income of $515 million for 2016 to get to normalized earnings.

1. Amortization of purchased intangibles. This is about $345 million. Not all of these expenses can be excluded. These are partly from purchased credit card portfolios and others which do not have really long lives. Assuming about 2/3 of these are non-economic. Pre-tax earnings can be increased by $230 million.

2. One-off charge to adjust for change in rewards expiry due to changes in law. This has cost the company $242 million and can be added back.

3. Loan loss provisions are coming off from historically low levels and higher expenses can be expected in future. Loans are about $16 billion and assuming a 7% loss rate, loan loss provisions should be about $1.1 billion. Company has provisioned $940 million in 2016. Thus we need to reduce pre-tax earnings by about $160 million. Management has guided to a 6.5% loss rate in the past but has revised it down to 5.5%.

The adjustments above total to $310 million pre-tax or about $200 million after tax at 35% rate. Thus total net income is about $715 million on about 56 million shares as of Q2, 2017. Thus normalized earnings are about $13 per share in 2016.

Not looking as cheap on this basis.

Vinod

[Cameron]

Using EBITDA or FCF doesn't seem to make sense to me, this looks like a bank wearing a marketing services mask.

Its not even comparable to a captive financial arm like CAT, F, or GM has, when most of their profit comes from the non-financial business, ADS gets more than half their profits from the 2 banks. 

[Cameron]

Using EBITDA or FCF doesn't seem to make sense to me, this looks like a bank wearing a marketing services mask.

Its not even comparable to a captive financial arm like CAT, F, or GM has, when most of their profit comes from the non-financial business, ADS gets more than half their profits from the 2 banks.

Comenity Bank earnings: $222,317,000
Comenity Capital earnings: $100,975,000

Since a large portion of their assets are in securitization, and from reading through a couple, none of the bonds list any FICO scores, so I have no idea what kind credit quality the assets have. But the bonds have 15% overcollateralization which pretty much gives a sign that they are bad. To contrast, since Synchrony was mentioned, they did an offering around a week ago with mostly prime FICO scores with 5% overcollateralization.

To be completely honest this looks like a lawsuit waiting to happen, I'm about 95% sure that FICO's have to be disclosed since Reg AB II was passed but I'll have to reread it.

[kab60]

There's a short thesis by Citron. And a decent writeup on VIC for anyone interested. I think it looks compelling, since the short thesis is basically about this being a consumer credit biz and not a software company. I somewhat agree but you don't really pay a software multiple. One needs to adjust their reported numbers as someone wrote. What I think is important considering where se probably are in the cycle; they bought back a shitload of shares during the GFC and seem opportunistic. I also like that management has skin in the game and a long track record, plus ValueAct has a large stake. Would probably make sense to split the Company but that's just an option.

[Cameron]

There's a short thesis by Citron. And a decent writeup on VIC for anyone interested. I think it looks compelling, since the short thesis is basically about this being a consumer credit biz and not a software company. I somewhat agree but you don't really pay a software multiple. One needs to adjust their reported numbers as someone wrote. What I think is important considering where se probably are in the cycle; they bought back a shitload of shares during the GFC and seem opportunistic. I also like that management has skin in the game and a long track record, plus ValueAct has a large stake. Would probably make sense to split the Company but that's just an option.

The problem is because of the prevalence of the 2 banks figuring out the software business multiple means nothing. This all depends on the credit quality of the two banks as well as the health of the credit card securization market. Simple research shows that we can assume the credit quality is below par. They buy subprime credit card receivables package them and have to pocket 5% of each transaction. The fact that management  throw the word EBITDA around like it's going out of style shows they have little idea how their business really makes money, their trust now has something like $23 billion receivables in it.

[kab60]

There's a short thesis by Citron. And a decent writeup on VIC for anyone interested. I think it looks compelling, since the short thesis is basically about this being a consumer credit biz and not a software company. I somewhat agree but you don't really pay a software multiple. One needs to adjust their reported numbers as someone wrote. What I think is important considering where se probably are in the cycle; they bought back a shitload of shares during the GFC and seem opportunistic. I also like that management has skin in the game and a long track record, plus ValueAct has a large stake. Would probably make sense to split the Company but that's just an option.

The problem is because of the prevalence of the 2 banks figuring out the software business multiple means nothing. This all depends on the credit quality of the two banks as well as the health of the credit card securization market. Simple research shows that we can assume the credit quality is below par. They buy subprime credit card receivables package them and have to pocket 5% of each transaction. The fact that management  throw the word EBITDA around like it's going out of style shows they have little idea how their business really makes money, their trust now has something like $23 billion receivables in it.

Appreciate it. I still have a lot of reading to do. Usually stay away from companies like these, but it seems like they handled the GFC well. How do you explain that if credit quality was total crap (and how do you know?). Why would management run the risk of blowup when they own 2,5 pct. of the Company? It's possible you're right, I havent't dug in. Thanks

[Cameron]

There's a short thesis by Citron. And a decent writeup on VIC for anyone interested. I think it looks compelling, since the short thesis is basically about this being a consumer credit biz and not a software company. I somewhat agree but you don't really pay a software multiple. One needs to adjust their reported numbers as someone wrote. What I think is important considering where se probably are in the cycle; they bought back a shitload of shares during the GFC and seem opportunistic. I also like that management has skin in the game and a long track record, plus ValueAct has a large stake. Would probably make sense to split the Company but that's just an option.

The problem is because of the prevalence of the 2 banks figuring out the software business multiple means nothing. This all depends on the credit quality of the two banks as well as the health of the credit card securization market. Simple research shows that we can assume the credit quality is below par. They buy subprime credit card receivables package them and have to pocket 5% of each transaction. The fact that management  throw the word EBITDA around like it's going out of style shows they have little idea how their business really makes money, their trust now has something like $23 billion receivables in it.

Appreciate it. I still have a lot of reading to do. Usually stay away from companies like these, but it seems like they handled the GFC well. How do you explain that if credit quality was total crap (and how do you know?). Why would management run the risk of blowup when they own 2,5 pct. of the Company? It's possible you're right, I havent't dug in. Thanks

In terms of the GFC the securization part of their business wasn't as large as it is today, as far as I can tell they didn't issue any bonds from 2004-2008.

For their credit quality, I have a previous comment on the thread that gives an explanation but they don't list the FICO scores of any of the receivables that they have in their trust which means we are mostly left in the dark in that respect. But each of these bonds have to have overcollaterialization. So in their most recent offering they have class A with 550M in principle class B with 42M in principle and class C with 27.5M. Then they have excess collateral of 115M, these are used as a buffer, the bigger the buffer the worse the credit quality of the underlying asset. If they thought they could get away with 45M in excess collateral they would as they would make more money on servicing the receivables. It equates to something like 15% of the bond being excess collateral which is the equivalent to some of the subprime auto offerings that I've looked at.

Plus they are non-recourse loans.   

https://www.sec.gov/cgi-bin/browse-edgar?company=World+Financial+Network+Credit+Card+Master+Note+Trust&owner=exclude&action=getcompany

[Cameron]

Also their treasurer Michael Blackham has experience in subprime lending servicing that he was able to pick up leading up to the GFC at Fairbanks Capital Corp So it doesn't surprise me that what they are offering has become worse since 2014 when he took over.

Good luck to those invested.