|Shares Out. (in M):||15||P/E||n/a||n/a|
|Market Cap (in $M):||120||P/FCF||2.8x||2.8x|
|Net Debt (in $M):||74||EBIT||23||23|
Asta Funding (ASFI) currently has the most favorable risk/reward balance of any of our holdings, probably the best one we have seen in several years. We own other stocks that will be ten-baggers if they work out favorably, but they have non-trivial changes of large losses, while Asta has very high odds of a 100% gain and very low odds of even a small loss under any scenario we can envision. The value comes from the following:
This $20 bill is still lying on the sidewalk, waiting to be picked up, for two reasons: The stock has a $120 million market cap and is thinly traded ($435,000 average daily volume over last 90 days), so it is off-limits to some funds. More importantly, our simple valuation calculation to get to 100% upside - net cash on hand plus three years of cash flows - is well-hidden behind a unique situation and GAAP financials that bear no relation to the economic reality, so the quant trading robots and most human fund managers can't see it. If you misanalyse any of the pieces or try to use valuation multiples instead of focusing on the discrete pieces of cash value, not only might you overlook the $20 bill; you can end up embarrassing yourself by telling the world that the stock is a short. (See http://www.sumzero.com/postings/4489/guest_view.)
THE BUSINESS AND ITS STORY
Asta acquires and manages distressed consumer debt receivables, primarily credit card receivables. These are debts that have long ago defaulted, been through the collections mill, and been written off by their originators. Asta pays 3-4% of the debt's face value to acquire it and then squeezes a bit more blood from the stones over a several-year period, using a network of third-party collection agencies and attorneys. The company is 28% owned by the Stern family, which runs it. Founder Arthur is still chairman emeritus and active on the Board; his son Gary is chairman and CEO; his daughter Barbara owns her piece but is not in management.
Asta's story over the last few years follows a familiar arc. The simple steady-state business model is to generate cash from collections of receivables on the books, spend a bit on SG&A and taxes, and use part or all of the cash flow to acquire new receivables. Like so many other financial services businesses, Asta over-reached at the bubble's peak: In 2007 they paid $300 million for a single large receivables acquisition, which roughly doubled their total receivables, and they paid for it with bank debt. (They refer to this as the "Great Seneca portfolio" in their press releases and as the "Portfolio Purchase" in their SEC filings.)
Soon thereafter, the credit bubble collapsed and all hell broke loose. Collections dried up, their modeling assumptions underlying their receivables accounting were suddenly far too optimistic, and they were forced to make massive write-downs of their existing receivables. The Great Seneca acquisition became a dud. Then the market for acquiring distressed receivables moved against them. Although the supply of distressed receivables for purchase increased, the demand increased even more, thanks in part to all the newly-formed vulture funds. Buyers started bidding up the purchase prices of distressed receivables to levels that seemed - at least to Asta's management - unlikely to produce a satisfactory return. The combined result was to decimate Asta's financials (all years are fiscal years ending in September):
But the Sterns were smart, in three ways that are relevant for why the stock value is hidden.
First, when they acquired the Great Seneca portfolio, they put it and its debt into a subsidiary and made the debt non-recourse to the parent.
Second, they refused to overpay for new receivables and essentially put the business into run-off mode. (Think of Buffett's insurance business managers refusing to write new premium too cheaply, taken to extremes.) Their investments in new portfolios fell from $441m in 2007 to $50m, $20m, and $8m in 2010. Total receivables balance fell from $546m in 2007 to $122m in the most recent quarter. They have been steadily accumulating the cash they collect from their remaining receivables, and sitting on it.
Third, in June they did what any good Buffett-following management team would do and announced they would deploy $20m of that cash to buy back their own massively-undervalued shares. That was 20% of their market cap at the time and is still 17%.
THE ACCOUNTING, AND WHY IT MATTERS
When Asta acquires a new receivables portfolio, it books the receivables at cost. Remember, the cost is only 3-4% of the debt's face value, and although Asta will never recover anywhere close to face value, it can easily recover far more than the book value. The accounting splits the portfolio into two buckets. The "interest accounting" bucket contains receivables that, based on various characteristics, make their collection amounts and timing reasonably predictable. Asta assumes it will recover 100% of the cost basis over the first 24-39 months and will eventually recover 130-140% of the basis over seven years. Each quarter, Asta books a pre-set amount of finance income on each pool of receivables and amortizes the receivables on the books by a different pre-set amount. For example, in a given quarter it might expect to collect $7m cash from the "interest accounting" receivables and to amortize $5m of those receivables, in which case it would book $2m of "finance income" to the revenue line. Note that the income statement does not show all net cash collections in the revenue line; it only shows the finance income amount, i.e., the assumed spread between collections and amortization. If in a given quarter Asta collects more than expected, it books the excess as deferred revenue. Once a particular portfolio gets amortized to zero, the deferred revenue becomes finance income, and any collection beyond that in later periods is 100% finance income.
(The other side of collecting excess revenue is when collection experience - coupled with, say, a 100-year financial flood - suggests collections will be less than expected. Then Asta must take an impairment charge to the receivables all at once. Before the credit bubble popped, collection timing assumptions for the "interest accounting" bucket were 18-24 months to collect 100% of book value and five years to collect 130-140%; the impairments in 2009 were largely the result of extending this timing to 24-39 months and seven years.)
The second, "cost accounting" bucket contains receivables whose collections are too hard to forecast precisely. Every dollar collected is simply booked as a reduction of the receivable amount until that amount reaches zero. After that, every dollar collected is booked as finance income.
The key point for both buckets is that Asta can make collections from receivables portfolios that have already been amortized to zero: from "income accounting" receivables after they have been amortized over their allotted seven years, and from "cost accounting" receivables after actual collections have exceeded their cost. Asta highlights these "zero-basis" or "fully amortized" collections in their earnings releases. They also provide enough detail in the SEC filings to allow us to calculate that roughly 90% of the zero-basis income is currently coming from the "income accounting" receivables.
One final accounting-related note: Because of the Great Seneca portfolio's disappointing performance, bank covenants require that all of the cash collections (after expenses) be used to pay down the debt. The portfolio is now subject to "cost accounting"; it produces zero GAAP finance income for the company.
THE GAAP FINANCIALS, AND WHY THEY LIE
Let's start by summarizing the lies, all of which are good ones for prospective shareholders:
Because of those lies, here is how the company looks to the quant robots, in simplified terms and $m:
BALANCE SHEET as of 6/30/2011
Cash (including $1m restricted cash) 105
Deferred income taxes 17
Other assets 8
Other liabilities 6
BOOK VALUE 172
With a market cap of $120m, the stock is trading at 0.7x book - not bad, but not necessarily exciting for a run-off business. Now the income statement:
INCOME STATEMENT, trailing 4 quarters
Operating expenses -23
Operating income 9
Interest expense -3
Pretax income 6
Income taxes (40% rate) -2
Net income 4
So the stock is trading at 30x trailing earnings. Even adjusting for the impairments (including taxing the adjustment at 40%), net income is $14 million and the stock is at 9x, for a run-off business. Yawn.
Now here is the reality. We start by stripping Great Seneca out of the calculations. Asta is collecting $3.5m per quarter from Great Seneca and is steadily paying down the related debt. The portfolio may either turn around enough that Asta can change the accounting back to producing GAAP income, or it may steadily grind the debt down to zero and then keep collecting more cash. Either way, Great Seneca has some nice unquantifiable upside for Asta. For our purposes, we can conservatively value it at zero and recognize that it has real option value. (Actually, we value it at less than zero; we remove its cash inflows but retain any related operating expenses.)
In all scenarios for Great Seneca, the non-recourse debt can't hurt Asta; if things turn south, Asta can hand Great Seneca over to the banks and walk away. That means Asta's true debt level is zero, and its net cash as of 6/30/2011 is $105m - about 90% of its market cap.
To look at Asta's cash flows, we take the last three quarters and multiply by 4/3; we don't have quarterly numbers for their fiscal 4th quarter (September) 2010, so we can't use trailing four quarters. The results for the last three quarters have been remarkably consistent and give us confidence that the following numbers show a good run rate:
ADJUSTED CASH FLOWS, annual run rate
Cash collections from fully-amortized receivables 36
Cash collections from amortizing receivables 50
Total cash collections 86
Less Great Seneca collections -14
Adjusted cash collections 72
Operating expenses -21
Cash operating income 51
Interest expense (excluding Great Seneca) 0
Income taxes (40% of GAAP net income) -8
Cash from operations 43
We make one final adjustment to get to free cash flow, for the sake of simplification. The company is not quite in run-off mode. Over the last three quarters it has purchased $7m of new receivables, for a run rate of $9m. These are very different receivables than past purchases; many of them are performing debt, and the company is paying 40% of face value for them. This shift just occurred three quarters ago and has not impacted their recent cash collections much. We exclude the new purchases on both the cash cost side and on the future cash inflows side. In conceptual terms, we are making the conservative assumption that the new receivables purchases being made by this (smart) management team have at least a zero NPV.
Because there are no other material cash inflows or outflows, for our purposes free cash flow to equity equals cash from operations - a $43m run rate. Even without stripping out Asta's big cash balance, the stock is trading at 2.8x free cash flow.
VALUING THE PIECES, OR "HOW LONG CAN THESE CASH FLOWS LAST?"
We value the business as follows:
Cash on hand as of June 30 was $105m. Asta is consistently churning out $11m more per quarter, using our simplified approach. Thus by September 30 they should be at $116m - just slightly less than their entire market cap, and the primary reason this is such a safe stock. (If you want, you can subtract $2m for their new portfolio purchases, but only if you add $2m of future cash flow value below. It's immaterial either way.)
The hardest question is how to value the cash flows, but the safest approach is to simplify this analysis down as well. We are Buffett's basketball scout looking for seven-foot guys who can shoot; we know we've found a seven-footer and don't particularly care whether he's 7'0" or 7'1". We are willing to say the cash flows are worth at least $120m based on the following:
Finally, we can estimate an additional value for the buybacks, but only roughly. If we are right that the stock is worth $16 (ignoring the Great Seneca option value), and if Asta were magically able to buy back all $20m of its authorization at today's $8 price, it would create $8 of value for every share purchased. Spread across the remaining ~80% of shares, that is roughly $2 of additional value per share. However, given the stock's low trading volumes and restrictions on what percentage of total volume the company's trading can be, it would take the company at least a year to purchase $20 million. We do not expect the stock price to stay at $8 for the next year. A higher price means less value creation from the buybacks. Even so, let's SWAG the buyback kicker at another $1 per share - not bad on an $8 stock price.
WHAT IS GOING TO HAPPEN TO ALL THAT CASH?
Whenever a company is sitting on a big pile of cash, one must weigh the odds that management will squander it in an ill-conceived acquisition or business expansion or grossly overpay for even a strategically sound one. (See, for example, almost every cash-rich large tech company over the last few quarters.) That risk is certainly at play here: Management makes clear during their earnings calls that they are looking to spend cash on one or more acquisitions within the financial services industry (not just in receivables collections). As of their last earnings call they were about to step up their efforts once their earnings quiet period ended and, because they were complaining about the asking prices of businesses, they may have stepped up their efforts even more now that asking prices have presumably come down along with public market prices.
However, we have faith that an acquisition would not be materially value-destroying, because of management's ownership stake and their smart stewardship of capital over the past several years. First, the Sterns own 28% of the company; our loss is their loss. Second, although they made a large portfolio acquisition at the bubble peak, the acquisition's timing merely suggests they were not yet listening to Nouriel Roubini (they didn't call the top), and the deal terms speak well of them: they did it with non-recourse debt in an industry that typically operates as cash-and-carry. Third, they have been sitting on a lot of cash for quite a while now and have undoubtedly let acquisition opportunities pass them by due to pricing. Fourth and most tellingly, they refused to chase distressed debt prices when their competitors bid them up, essentially consigning themselves to run-off mode when the overwhelming institutional imperative is to self-perpetuate by continuing to do what you've always done. Indeed, we wonder whether management has been overly conservative. We strongly suspect that other purchasers have bid prices higher after appropriately lowering their cost of capital assumption, thanks to Helicopter Ben driving down the cost of debt. So far the Sterns have refused to take and use any of that cheap debt.
DOWNSIDE RISKS - OR RATHER, POTENTIAL LIMITS TO THE UPSIDE
We see only two real risks - one that could reduce the value of existing cash that currently equals 100% of its market cap, and one that could reduce the value of the future cash flows that are worth another 100% of its market cap. However, even both risks in combination are highly unlikely to drive the company's expected value down by half, to below the current stock price.
First, we could be wrong and Asta could overpay badly for an acquisition. The margin of safety here is large. The acquisition bull case is that Asta buys well and creates positive NPV. Our expected case is that an acquisition is NPV-neutral. The bear case, if it occurs, would still likely be a modestly NPV-negative acquisition, which would reduce but not eliminate the 100% upside. Eliminating all the upside would require taking all of the current cash and flushing it down the toilet by spending it on something that is worth, literally, zero. We have seen acquisitions worth zero or less than zero in the past - Bank of America / Countrywide comes to mind currently - but they are exceedingly rare.
The second risk is that collections decline more than expected, either because the economy falls into a new recession or because as-yet-unreleased Dodd-Frank regulations restrain collection efforts. (This second cause seems unlikely to be material given that $900m of face value receivables have already been reduced to legal judgments against the debtors.) As with acquisitions, to cause true downside, the collection declines would need to be so bad that they drove Asta's cash flows negative and started eating into its existing cash balance. Asta is currently collecting $18m per quarter (excluding Great Seneca) against only $5m in operating expenses, and cash earnings were strongly positive even during the worst of the financial crisis. Even taken in combination, the acquisition and collections bear cases seem more likely to limit the stock's upside than to create downside from the current price.
One final interesting note is that these two risks are in part self-balancing: If the economy slows further and makes collections harder, potential acquisition prices would likely fall and the company would be more likely to create value through an acquisition.
|Entry||09/16/2011 05:32 PM|
Nice work here. I've followed ASFI for years and made a good bit on them back when their business model had more visibility.
It seems like you're valuing this on asset recovery to shareholders. My understanding is that right now it would cost ASFI more to borrow money than they make return on available portfolios they can purchase (see discussion in previous posting of ASFI). That means the endgame is all important. The one thing keeping me out of this stock is that I've lost count of the number of times I've seen a beautiful sum-of-parts turn into a mess as management plays to their own interests in the endgame and tries to continue operations or acquire their way into continued CEO status. I agree the owning family has a lot of stock ownership, but they have also demonstrated a willingness to take money out of the company via 10% interest related party loans. Game-theory-wise, there are plenty of ways that the value might not end up in shareholder hands.
I would appreciate if you could perhaps comment on your theory of how you expect the endgame to play out and what events might result in you keeping or ditching the stock on a path to realizing the value.
|Entry||09/17/2011 08:29 PM|
Good question, 4maps. I haven't followed the company for years as you have, but management has said how they view the end game, and I have no reason to doubt them. They announced a $20m buyback over the next year, and they do indeed intend to "acquire their way into continued CEO status." Their statements on recent earnings calls make clear that they are vigorously pursuing an acquisition search. However, for the reasons expressed in the writeup, I view the bad-acquisition risk as low; our baseline assumption is that their acquisitions would be NPV-neutral. Just to pick a number, if they pay $120m (current market cap) for one or more acquisitions, I would expect the market to value the acquired businesses at $120m, e.g., they might pay 10x earnings for something with comps of 10x earnings.
In a year they will have generated another ~$40m in cash from the existing receivables book, half of which they say they will spend on buybacks. Sticking with my $120m acquisition number, I can envision owning the following at that point: (1) an ongoing business worth $120m, (2) a run-off receivables book still worth at least $80m, (3) $20m in cash, (4) on a share count that is ~15% lower. If the stock price were still at current levels, the market cap would be only $100m, and the acquired business alone would be worth 20% more than that.
As cash continues to come in after that (another $40m/year from the old receivables book for a few years plus perhaps $12m/year from the acquired businesses), one could ask, "now what's the endgame for the new incoming cash." I don't know but would not be surprised to see, at some point, an increase in new receivables purchases or a dividend increase (it was cut by 50% in September 2008).
What might cause us to ditch the stock? Either a truly horrendous acquisition or the discovery that they had, without a strong reason why, not repurchased anywhere near $20m in stock after a year. (Recent trading volumes mean that their purchase pace must be slow so far.)
|Entry||09/18/2011 12:10 AM|
Thank you for the very direct and considered response and sharing the milestones you might look out for, that's definitely how I think about investing - what might happen and how should I respond as an owner if it does.
I really hadn't thought about ASFI before as a simple acquisition vehicle. If one accepts the NPV as greater than market cap, then it comes down to how efficiently the management will make a new business acquisition and whether one is willing to join them at a discount. There are some additional concerns with some of their related party transactions perhaps worrying one over fairness of the series of transactions but that's a model that people are familiar with for acquisition vehicles.
|Subject||Endgame (again) and IRS audit|
|Entry||09/19/2011 12:17 PM|
I would agree that one can think of ASFI as an acquisition vehicle, but not an ordinary one. We have invested in a SPAC (special purpose acquisition vehicle) in the past, so I can compare the two based on experience. Someone buying the shares of an ordinary SPAC pays $1 for $1 in cash, with the expectation that the SPAC manager will come up with a deal that will create, say, $1.30-$1.40 of value and skim $0.10-0.15 of carry for himself, leaving 0.15-0.25 of profit for the shareholder. It is explicitly a bet on the jockey; if the jockey isn't very good, you can lose money from the carry alone. With Asta, you are paying only $0.50 for $1 of cash and FCF, and there is no carry skim for the jockey. We are assuming that a mediocre jockey is sitting on the world's best horse, and the jockey would need to completely screw up to hurt us.
On the IRS audit: shame on me for not mentioning it in the writeup. I didn't do so because it does not appear to be a material risk and did not stay top-of-mind for us. Management isn't saying anything more than that an IRS audit of their 2008 and 2009 financials is ongoing. When I look at their financials, they have had a steady effective rate of ~40.5% for at least the last 6 years. The only material issue I can envision, which is strengthened by the IRS's choice of 2008 and 2009, is that the IRS claims Asta made their writedowns too big in those years. Those write-downs benefitted their taxes at a ~40.5% rate. If they collect more than their lowered assumptions that drove the write-downs, then they will pay 40.5% taxes on the "excess" collections when they occur. This appears to be an issue of tax payment timing plus perhaps some penalties and interest, which I view as immaterial to the big-picture investment case.
I did realize in the course of thinking further about the tax issue that taxes may have caused us to slightly overstate the future cash flows: If we are right that amortizations of their receivables book will decrease faster that cash collections in future periods, then their GAAP pretax income will increase, and taxes paid will also increase, which will make FCF modestly lower. Again, our valuation of future cash flows is imprecise and conservative enough that I do not view this issue as material to the investment case.
|Subject||Confirming information from Asta's CFO|
|Entry||09/20/2011 11:16 AM|
I just had a follow-up phone call with Asta's CFO, who said the following:
(1) Although they are scrupulous about not making forward-looking statements, there is no reason to expect a change in the pattern of the past several quarters in which cash collections decline only gradually while the receivables book value is declining (being amortized) much more rapidly. (My interpretation: that is support for our assumption that the receivables book is worth at least three years' worth of current cash flows.)
(2) The IRS is indeed focused on Asta's tax refund that was generated by the impairments. Also, my assumption is correct that the issue is only one of timing: if collections came in ahead of the new assumptions, they would pay taxes them anyway.
(3) When I asked how large of acquisitions they would do in total in the near future and put out a $120m number, he responded "I don't want to put my own number out there, but I also don't want people to think it could be that big."
(4) An alternative explanation for why other distressed debt buyers have bid prices higher, apart from the availability of cheap debt, is that others have high-fixed-cost collections infrastructure while Asta outsources collections. Those infrastructures were set up when collections assumptions were much higher and have not been cut back sufficiently. Others must cover their fixed costs, while Asta can simply step back.
|Subject||RE: Couple Questions|
|Entry||09/20/2011 02:11 PM|
(1) Management was clear that the $5b number is specifically for zero-basis receivables, and as you say Great Seneca receivables have a $81m basis (not zero). From those two facts, I think it safe to say that the $5b number excludes Great Seneca.
(2) Different pieces to your question: Bankruptcy would discharge credit card debts (but if that happened the debt should be eliminated from Asta's face value calculations as well). Death would not technically discharge the debts, because the estate would have to pay them before distributing anything to the heirs. However, anyone with debt that has made it this far through collections probably doesn't have anything in the estate anyway. For garnishment, the unemployed are off-limits, some states prohibit garnishment for credit card debt, some states prohibit garnishment of single parents, and garnishment is limited to 25% of wages (less than 25% in some states). In short, there are several reasons why collections on judgments could be zero for a time or dribble in for years yet still ultimately be collected.
(3) Correct, any write-down of Great Seneca would be non-cash, and Great Seneca can't hurt Asta under our valuation approach because we have already stripped out all the revenue while leaving all the costs.
(4) To the best of my knowledge the Gelband contract is arm's length. It is also immaterial; they are paying him $45k plus 50k options at $11.50 to help them find acquisition candidates.
Finally, on "the dealings between the Stern family and company," the only recent dealings I can find are the Sterns loaning funds to the company in 2008 to temporarily bail it out of a jam in the teeth of the financial crisis. Those are the kind of dealings I like to see.
|Subject||RE: Value of portfolio|
|Entry||09/21/2011 11:21 PM|
The primary insight behind the investment case is that the amortization schedule is a terrible starting point for approximating future cash flows; future collections have almost nothing to do with either the receivables book value or the amount of that book value being amortized.
I do not expect the amortizing receivables to continue to deliver $9m per quarter; I do expect the total receivables to keep delivering $21m on a slow decline, with the mix shifting to zero-basis receivables. Once a particular portfolio is amortized to zero, its cash collections do not suddenly drop off, but Asta starts reporting its collections in the zero-basis bucket. So over time, collections from amortizing receivables will decrease and those from fully zero-basis receivables will increase. We should see a big jump in this direction in the December quarter (1Q) results, because Asta's amortization assumptions are made on an annual basis rather than a quarterly one.
|Entry||12/27/2011 11:04 AM|
Katana - what do you make of the last quarter? Collections dropped close to 20% sequentially. Are the future cash flows still likely to be $120mm? Still seems like the downside is non existant but trying to determine the upside. What is your read on the Sterns? Do you think they will issue a special dividend? Thanks
|Subject||RE: RE: Comments on 5 different questions/issues|
|Entry||01/02/2012 09:58 PM|
Haven't looked at asfi in years but i believe all of the collections that mgmt reports is net of collection costs. Most of what they used to collect was outsourced and the collection costs are all contingent and success based. That is at least how I recall it working.
|Entry||06/04/2012 04:30 PM|
Stock trading at $8.80 and company buys back 1 mm shares at $9.40? The $9.40 price is still a 21%+ discount to tangible book. How often do you see a large buyback at a premium?
$130 mm market cap, $116 mm in cash.
See the recent entry into matriomonial funding business? Seems like a natural fit with the personal injury funding business.
|Entry||11/15/2012 11:12 AM|
Anyone following this? Cash continues to grow but some regulatory scrutiny. Any further read on the Sterns?