Cabela's (NYSE:CAB) is a specialty retailer of hunting, fishing, camping, and related outdoor merchandise that was founded in 1961. The company began as just a catalog business and then started opening retail stores in 1987. The company operates through three segments: Retail, Direct, and Financial Services. The Retail segment operates 59 stores across North America as of June 30, 2014, the Direct segment comprises the catalog and ecommerce business, and the Financial Services segment operates the "World's Foremost Bank" and issues Visa-branded credit cards. In addition, the "Corporate Overhead and Other" segment is largely shared services of Direct and Retail.
To provide some context, segment revenues and operating income, as presented by the company, for the last five years is set forth below.
Based on the data above, one might think the Financial Services segment is relatively unimportant. However, upon closer inspection, the Financial Services segment takes on greater importance.
CAB went public in June 2004 at $20 and earned $1.03 per share that year. The company then struggled to grow earnings for the next five years despite rapidly growing retail sales:
*Normalized EPS; Data from CAB 10-K filings
Then, CAB's retail sales growth slowed (fewer openings mainly), and earnings accelerated.
*Normalized EPS; Data from CAB 10-K filings
Based on the above tables, it is clear that CAB's earnings grew more quickly when the Company's retail and overall sales were growing slowly (2009-2013) than when the Company's retail and overall sales were growing rapidly (2004-2009). Therefore, the key to CAB's EPS growth does not appear to be retail or even overall sales growth...so what's really going on here?
Earnings Management 101
Companies have discretion in their accounting assumptions, but sometimes management gets hooked on the drug that is earnings growth and/or beating estimates. I think CAB management got hooked.
The table below sets forth CAB's adjusted earnings relative to sell-side estimates and shows that after a change in "accounting principle" at the beginning of 2010 (discussed in detail below), CAB did not "miss" earnings for 14 straight quarters. In fact, they beat estimates by an average of $0.06 per share (12%) from 1q10-2q13.
Data Source: Capital IQ
This is clearly an impressive track record, particularly because CAB did not provide specific quarterly guidance, so they couldn't "sandbag" the number.
There are three main theories that could explain CAB's newfound ability to consistently beat estimates:
- Retail store performance and new store openings (management's explanation)
- Bumper sales of guns and ammunition (investors' main focus)
- Cookie jar accounting with the provision/allowance (my explanation)
I will focus on #3 and answer any questions or comments others might have on #1 and #2 in the comments section, but first we need to go back in time a few years.
The Cookie Jar is Created
Effective January 3, 2010, CAB began consolidating when FASB updated ASC Topic 860, Transfers and Servicing, which eliminated the qualifying special purpose entity ("QSPE") concept. FASB also updated ASC 810, Consolidations, which requires the primary beneficiary of a variable interest entity ("VIE") to consolidate it.
In brief, CAB had to bring back onto its own balance sheet credit card loans it had "sold" to a bankruptcy-remote entity it controlled along with that entity's debt. As a result - and this is the REAL kicker - CAB had to begin reserving for loan losses at the "World's Foremost Bank" like other banks.
Prior to this "change in accounting principles," CAB's allowance for loan losses was only for un-securitized credit card receivables. In contrast, the securitized receivables placed into QSPEs (yes, of Enron notoriety) ignored the total assets and total liabilities of the QSPE. Instead, CAB merely recognized a "retained interest" on a mark-to-model basis.
Apparently, CAB's model for marking was too aggressive because CAB took a $114.6 million pre-tax charge ($92.8 million after-tax) as a result of the accounting change in 1q10. Notably, this charge did not go through the income statement (it was charged directly to equity):
Source: 1q10 CAB Form 10-Q
So CAB wrote-off these previously recognized profits and, in the process, CAB management created a cookie jar that would enable them to grow earnings and beat expectations with the stroke of a pen.
CAB's Cookie Jar
The table below highlights why I think theory #3 (Cookie Jar) explains CAB's ability to beat expectations. Similar to the table of earnings surprises above, I include the estimated and actual adjusted EPS. However, this time I also note the benefit from drawing down the allowance (as a percentage of CC loans outstanding) and estimate its contribution to earnings (and earnings "beats").
Since the declining allowance has driven approximately over 100% of CAB's cumulative EPS beats since 1q10, I believe it's clear that the cookie jar loan reserves are what drove CAB's impressive 2010-2q14 performance relative to expectations.
In the table below, I calculate the EPS benefit and contribution to growth resulting from the decrease in the allowance as a percentage of CC loans receivable relative to the 2009 assumption (4.47%). In other words, it quantifies the benefit CAB's EPS and EPS growth received from using less conservative (increasingly aggressive) accounting assumptions with respect to credit losses.
While other factors contributed to earnings growth, the cookie jar appears to have been the largest. In fact, I estimate that CAB's reduction in the percent reserved has driven over half (13.2% / 24.1%) of the company's earnings growth since 2009.
Note: $1.15 is different from the $1.19 table above because the prior table incorporates the first and second quarters of 2014.
The remainder of the company's earnings growth can be attributed to (1) a depressed starting point and subsequent improvement in the retail environment, (2) all-time high guns and ammo sales, particularly in late 2012 and the three quarters of 2013, and, more recently, (3) aggressive cost cuts to manage expenses.
Clearly, the environment for consumer demand has improved since 2009, but that is a rear view mirror issue and we currently appear to be closer to a cyclical peak than cyclical trough. Additionally, guns and ammunition sales continue to tail off, but present more of a risk to CAB than opportunity since demand for these products remains elevated. Finally, we will let other investors draw their own conclusions about what the implications of a growing retail operation touting cost cutting opportunities such as re-negotiating contracts for doorknobs.
2q14 Earnings Call:
Answer - Tommy Millner: Absolutely. And specifically, as Ralph and our construction teams are looking at how to lower store cost, we're not talking about gutting customer experience. We're talking about construction techniques and that kind of stuff, not changing in-store experience, which is sacred ground here.
Answer - Ralph Castner: Yes. What we should do sometime, Matt, is walk through a store, and I can kind of give you our impression about how we think about it, and where we see opportunity. And I can give you examples, it would be silly, but it's literally $50,000's and $100,000 at a time, going with national contracts in some of our vendors.
One of the things we looked at is -- I know this sounds silly, but the hardware we use in all the parts of the store that the customer doesn't see -- door knobs, hinges, those kinds of things -- we are just taking every aspect of the design and seeing what we can do better. And by the way, that philosophy is, what has happened all across the Company, is how can we not impact the customer experience but do things better?
Source: Capital IQ
The Cookie Jar Deficit
Up until now, I've discussed what has happened. Now let's talk about where CAB is today because sources of growth and upside surprises can quickly become the cause of a decline in earnings or result in disappointing earnings. In CAB's case, the reliance on a cookie jar to generate earnings leads to the critical question "Are there any cookies left OR is it actually in deficit?"
In preview, it isn't pretty - the jar is empty at best and more likely in deficit.
As discussed above, the largest driver of earnings growth and upside surprises has been the decline in the loan loss allowance - from $114.6 million as of 4q09 to $47.4 million in 2q14 (a 59% decline). While the allowance fell 59%, credit card receivables rose from $2.56 billion (4q09 gross) to $3.95 billion, or 54%! In other words, the allowance as a percentage of managed receivables fell from 4.47% to 1.20%.
Below is CAB's disclosure of what the allowance represents (emphasis added):
Allowance for Loan Losses on Credit Cards
The allowance for loan losses represents management's estimate of probable losses inherent in the credit card loan portfolio. The allowance for loan losses is established through a charge to the provision for loan losses and is evaluated by management for adequacy. Loans on a payment plan or non-accrual are segmented from the rest of the credit card loan portfolio into a restructured credit card loan segment before establishing an allowance for loan losses as these loans have a higher probability of loss. Management estimates losses inherent in the credit card loans segment and restructured credit card loans segment based on a model which tracks historical loss experience on delinquent accounts, bankruptcies, death, and charge-offs, net of estimated recoveries. The Financial Services segment uses a migration analysis and historical bankruptcy and death rates to estimate the likelihood that a credit card loan in the credit card loans segment will progress through the various stages of delinquency and to charge-off. This analysis estimates the gross amount of principal that will be charged off over the next 12 months, net of recoveries. The Financial Services segment uses historical charge-off rates to estimate the charge-offs over the life of the restructured credit card loan, net of recoveries. This estimate is used to derive an estimated allowance for loan losses.
Source: CAB 2013 Form 10-K, page 62
So charge-offs, net of recoveries are expected to be 1.20%, as set forth in the table below:
I have drawn a line in between 1q12 and 2q12 in the table above because of comments by CAB's CFO, Ralph Castner, indicate an allowance of approximately 2.25% is "normal" and "to really expect anything is sustainable below 2%...is just unlikely."
2012 Investor Day (March 22, 2012):
Ralph Castner: A lot of guys believed us and saw some of the differences and educated themselves through this, but we've clearly seen that fall to sort of historical levels, and I'll tell you, before the run up we saw in '08, '09 with the great recession, this business pretty much like clockwork had charge-offs between 2% and 2.5% and we're well within that range. To really expect anything is sustainable below 2% is just unlikely. I fully expect charge-offs will be -- maybe even tick a little above 2.5%, but somewhere in that 2.5% range is a sustainable ongoing level for charge-offs.
1q12 Earnings Call:
Question - Aaron Goldstein: Lastly on the bank side on the charge-off provisions you talked about historical rates been in the 2% to 2.5%. When you are looking at your provisions, is that what it would currently -- if you currently unwound at the current rate, that is what the charge-offs you would need or is there a buffer in there for kind of what the average historical rate that you think it going to be?
Ralph Castner: We'll, I'm not sure I understand your question. When we calculate the allowance, we use the existing delinquencies and that's in correlation to a charge-off level that we expect to see. So as we go forward, what is going to happen with the allowance, we've ‒ over the last two or three years, we have had the benefit of dramatically lower charge-offs and delinquencies. Sometime soon that is going to start to get mitigated because of growth in the portfolio. So we are going to need a bigger allowance just because the portfolio is growing. And at least I feel like -- as I sit here today, I feel like we are near -- pretty close to that starting to cross. In other words, the growth of the portfolio is going to require a bigger and bigger allowance, which will offset any benefit we get from continued improvements in the macro credit environment.
Ralph Castner: I said in my comments we don't expect a significant reduction -- I want to make sure we are talking about the same thing, in the balance sheet allowance. So yes, I would expect charge-offs and the provision to be more closely aligned in the last three quarters of the year.
2013 Investor Day (March 13, 2013):
Ralph Castner: As you go back and look at the photo sign, this is charge-offs, I don't want to say from the beginning of time, but I'll say we hardly had enough accounts back in 2000 to make it relevant for today's portfolio. I had somebody graph our charge-offs the last 13 years, and the takeaway over long periods of time is we've seen basically two spikes and chart drops. In 2000, '01 levels, charge-offs got up just above 3%. And obviously, in the '08, '09 timeframe, they got up around 5%. Really excluding those, charge-offs have always been between 2% and 3%.
2014 Investor Day (March 26, 2014):
Source: Capital IQ
- Allowance for loan losses = expected charge-offs over the next 12 months
- CAB's CFO expects charge-offs to be 2.0% to 2.5% and suggests that any period where charge-offs drop below 2.0% is unsustainable.
- In spite of 1) and 2) above, the allowance is currently 1.20%...
This points to a significant problem with the allowance, as indicated by the fact that CAB's net charge-offs over any 12-month period have never been below ~1.65%, let alone 1.2%!
The speed with which CAB changed their commentary around the allowance this spring is demonstrative.
March 2014 Analyst Day, continued (March 26, 2014) (emphasis added):
Ralph Castner: I know that we don't expect a meaningful change in the actual charge-off rates that we're charging on our cards -- oh I'm sorry, our charge-off rate from bad debts in -- between 2013 and 2014. Now I know this has gotten a lot of attention, but there is a headwind as it relates to reserve releases between '13 and '14.
We had reserve releases of $12 million in 2013 (ed: absolute -- larger benefit as a % of CC loans), we've got receivable - we've got an allowance for loan losses of, I believe, about $60 million, that we will probably need to increase that reserve in 2014, not because it's under reserved at absolute levels, but we'll need to increase that commensurate with portfolio growth.
So if we've got a $60 million allowance and the portfolio is going to grow 12% to 14%, you can estimate there's some number between $5 million and $10 million. So we'll probably need to increase that reserve in 2014 just through the portfolio growth.
1q14 Earnings Call (April 24, 2014):
Ralph Castner: As I am thinking about the business today and particularly the allowance for loan losses, I still think for the full year we are going to have to increase that. Probably not to the extent of the $7 million that we shared with you at our Analyst Day that we have to, but I would still expect increases in the allowance as we move through the back half of the year.
Note: In 2q14, the allowance declined $3.7 million (again) sequentially.
Phases of CAB's Cookie Jar:
I think about CAB's Modern Era (i.e.: after the accounting change) in three phases.
Phase 1 (2010 - 1q12): "Free" Allowance and Improving Charge-Offs
Since CAB took a "one-time" charge to retained earnings for the change in accounting principles, the Company never had to bear those expenses in the income statement. As such, the relatively full allowance of 4.47% was "free" and reflected a weak consumer credit environment.
CAB's charge-offs declined from this fortuitous starting point, creating a two-fold benefit. First, lower charge-offs meant, for a given customer or set of customers, CAB could reserve less for future losses on new loans. Second, the lower charge-offs meant that they could reduce loan loss reserves (allowance) as a reduction in the provision for loan losses on existing loans outstanding.
Phase 2 (2q12 - Present): Going for Broke
Since the quarter ending March 31, 2012, the allowance has fallen from 2.26% to 1.20%. At the same time, trailing twelve month charge-offs have fallen from 2.15% to 1.78% (37 basis points) - a level the CFO has described as unsustainable (see above).
Had CAB not drawn down the allowance to this degree, the Company would have missed earnings and/or missed substantially worse for the last five quarters in a row -- a far cry from the 11.6% average earnings "beat" investors came to expect from 2010-1H13.
Phase 3a (3q14): The Day of Reckoning
CAB shareholders seem to think they own a retail growth company, but they are about to get a rude awakening because, in reality, they own an under-reserved bank wearing retailer clothing.
I believe CAB's investors will realize this unfortunate situation very soon. In particular, I expect CAB to report surprisingly weak results in 3q14 since (1) the allowance needs to start ratcheting up, and (2) 3q13 benefited from an egregiously low provision.
Another way to look at this:
Net Provision = Gross Provision +/- Adjustment for Prior Under-Reserving (Over-Reserving)
Net Provision % = Net Provision / Average Managed CC Loans
The net provision reached the second lowest level ever in 3q13 (0.95% annualized) despite having already released substantially all of the bank's "excess" reserves for loan losses (>2.25% of CC loans outstanding) by the middle of 2012.
Additionally, delinquencies and charge-offs have been essentially flat (TTM NCO % has remained between 1.78% and 1.83% for the last seven quarters), so CAB will have a hard time arguing that reserves should be falling... or have fallen a year ago for that matter.
Assuming a $3 million sequential increase in the allowance (as alluded to by management) and constant NCOs sequentially (down slightly Y/Y), CAB's provision will need to be approximately $19 million, or $10.6 million higher than a year ago. Since the provision is an item netted against gross revenues of the Financial Services segment, this will reduce segment revenues by approximately $10.6 million, all else equal. What's worse, this contra-revenue is pure margin. In other words, there is no cost reduction associated with this revenue decline.
The approximate EPS impact is: ~$0.10 = $10.6 million * (1-35% tax rate) / 71.7 million shares
A headwind of $0.10 may not sound like much, but it equates to more than 10% of estimated EPS and creates a greater than 10% headwind to EPS growth in the quarter.
CAB has not grown earnings in 3q or 4q by more than 25% since 3q11 when the baseline was much smaller ($0.31), thus causing more volatility. Even when guns and ammo were white hot in 4q12, CAB only grew earnings by 18% - and that was with the added benefit of the provision falling!
Phase 3b (4q14 & Beyond): The Long Slog Ahead
I believe 3q14 will be just the beginning of disappointing results for CAB's investors. As noted above, I estimate CAB received an EPS benefit of $0.23-$0.32 in 2010-2013. Starting with 3q14 and continuing forward, CAB will lose approximately 10% of EPS (using TTM EPS of $2.98 as a denominator) right off the bat since they can no longer release loan loss reserves (which count as a negative provision - a boost to revenue at 100% margin). So that resets the baseline for EPS to ~$2.70 per share.
Once again, it gets worse. Not only will CAB lose ~$0.30 of EPS, it will face a constant headwind as it needs to rebuild the allowance to at least the current (cyclically low) level of charge-offs (~1.8%) and more like the historical average discussed by the CFO (~2.25%).
Valuation & Scenarios
In the tables below, I estimate the EPS impact of various scenarios for the CC loan portfolio.
There are two important drivers of CAB's EPS impact. First, I show the impact to earnings from the cookie jar being depleted. In other words, this table answers the question, "What will happen to CAB's earnings if they provision at the same rate of NCOs under different NCO scenarios?"
For example, Scenario 1 assumes NCOs remain at the all-time low levels of 1.80%. Since last year CAB only provisioned 1.12% (raiding the cookie jar created the 68 basis point difference), CAB would have to either decrease the allowance again or increase the provision. Increasing the provision as a % of CC loans would decrease EPS by $0.25 ($27.8m lower pre-tax profits). In Scenario 3, I assume NCOs revert to the level suggested by CAB's CFO (see above), thus requiring the provision to increase 113 basis points (2.25% - 1.12%). A 113 basis point increase in the provision would cause EPS to decline by $0.42, all else equal.
Next, I show the impact to EPS from increasing the allowance to match the NCO (since allowance represents expected losses over the next twelve months). However, I assume that the impact is spread over six quarters to reflect the discretion management has with respect to how quickly they must recognize credit losses.
For example, in Scenario 1 below, I assume that CAB can maintain their allowance at 1.20% despite the 1.80% NCO rate. However, since the NCO percentage reflects realized credit losses on an annualized basis and the allowance represents estimated credit losses for the next twelve months, over time the allowance percentage needs to rise to match the NCO percentage (it should exceed it due to growth, but I'll give them the benefit of the doubt).
Rebuilding reserves (through additional provisions) can be spread out over more than 12 months (in the tables below I assume they spread out the additional provisions over 18 months), but if CAB waits too long, they risk the disastrous combination of being significantly under-reserved when credit deteriorates (the way a short hits a grand slam here).
Finally, I take the combined impact of these adjustments and CAB's run-rate EPS to estimate EPS and price targets under different NCO assumptions.
The P/E multiples used below are lower than CAB's recent trading multiple (~16.3x) to reflect the spurious nature of CAB's EPS (and EPS growth) in recent years. However, my assumptions are within its historical range and are well above the multiple at which it traded in 2008 and 2009.
Source: Capital IQ
Once investors realize CAB's earnings are dominated by the company's under-reserved bank instead of its retail operations, and that earnings are likely to decline in future years as a result of increasing provisions, I believe the market will assign a meaningfully lower multiple to CAB's EPS.
Price Target: $26.29 (58% downside).
Disclosure: The author is short CAB.
The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it. The author has no business relationship with any company whose stock is mentioned in this article.