Dec. 16, 2015 1:06 PM ET
Disclosure: I am/we are long APOL, STRA, AGX. (More…)
Revenue and enrollments have not come in great lately and the trend will likely continue, but from a purely operational perspective, Apollo is highly attractive. It trades for net cash.
A lot could happen outside the operations to make Apollo a big loser including regulatory action or a massive reduction in student lending.
On balance, Apollo is still fairly attractive as a very tiny position.
Let me start by saying that I think Apollo Group’s (NASDAQ:APOL) University of Phoenix (“UoP”) has a fairly poor educational value proposition and Apollo is not the most ethical company in the world. It’s also by no means a low-risk investment and nothing in this article should be construed as saying otherwise.
I’m not sure it’s worthless though.
Operational Risks are One Thing
First, let’s look at Apollo as a going concern – just at the operations and financials. It may be hard to compartmentalize and do this, but it’s necessary in order to think about the company clearly.
First, I will walk through all the assumptions baked into the model and provide color on Apollo’s performance throughout.
Revenue has been consistently declining for 5 years now and will likely continue to decline over the next few years given recent enrollment trends.
Subpoenas issued over the summer by the California Attorney General and FTC as well as a related probation by the DoD in October likely played a large role in the horrible Q4 2015 new student start y/y decline; however, it was also a very tough comp as the table above shows.
How much will revenue decline over the next few years? The low end of the company’s FY16 guidance of $2.18B implies a y/y decline of 15%. Some of the enrollment decline impact to revenue has been mitigated by tuition increases and retention improvement. The retention improvement has been driven by an increase in continuation-based scholarships, which have acted to offset some of the tuition increase. On the whole, revenue is declining, but not quite as fast as enrollments at this point. I think a reasonable scenario is y/y declines of around 15% over the next 3 years to get to $1.5B in revenue. Are enrollments and revenue stable at that point? It’s possible. Apollo is being more selective with prospective students. A recent Glassdoor review quantifies this, saying the internal goal is to intentionally cut University of Phoenix enrollment in half. They want to get rid of all the unprepared students who won’t continue, won’t perform, and won’t get a job, because these are the students who will both:
- Have much lower lifetime value to the company – acquisition cost only brings in 1-2 terms worth of tuition versus 5 years’ worth for a full-time bachelors student.
- Hurt UoP’s graduation rate, cohort default rate, gainful employment metrics, etc.
The metrics are largely a function of UoP’s value prop, but student profile does play a role and the company is doing the right thing looking for higher quality students. I should point out this is not just a financial decision and it’s not unethical, at least in my opinion. Much of the problem at for-profits has been predatory marketing pulling in people who have no business getting a college degree and are unprepared mentally and financially for the proposition. Part of that is on the for-profits, but it’s also on the consumers who made those boneheaded decisions. It is just those students that Apollo and many other for-profits like Strayer (NASDAQ:STRA) don’t want anymore, effectively making a good decision for these students.
Of course, there is also the chance that Apollo’s revenue and UoP’s enrollments continue decline well beyond the next 3 years, but I address this in a bear case valuation with a much more severe revenue decline and exit multiples that imply further decline.
10 years ago, EBIT margin was 26%. Since then it has gradually declined to 4.5% in FY15 with a 10-year average of 18% and a 5-year average of 10.4%. The FY15 number is understated due to $82mm in restructuring and impairment charges and $6mm of acquisition and related costs. Adding those back, FY15 EBIT was just over $200mm and EBIT margin 7.9%. The low end of FY16 revenue and op. income guidance is $2.18B and $115mm respectively, for a margin of 5.3%. In the conference call, the company said their internal, normalized target is 15-20%. That may take many investors by surprise. “15-20%? Revenue is still getting crushed. They’ll delever fixed costs and be unprofitable in FY17 if not sooner.” The problem with this thinking is that it does not appreciate the amount of variability in the for-profit education cost structure. Apollo’s revenue has been cut in half since 2010 and they’re still substantially profitable. While the cost structure is highly variable, there is some delay in cutting costs because the company has limited visibility on enrollment trends. I’ve observed that there is at least a 1 year delay between when enrollment declines to a certain level and when the FPE is able to adjust its cost structure to that level. Want proof? Look at Strayer’s financials from 2011-2014:
The company’s “old” average op. margin was over 30% and in the new environment, it’s more like 15-18%, but margin dipped well below that level in 2013 in the face of revenue declines before normalizing.
There’s no way I’m going to assume Apollo management’s 15-20% because I’m trying to do a very conservative, “kitchen sink” valuation here and, frankly, that seems kind of aggressive given where margins have been historically. However, I do think 10% is possible a few years out.
Again, I don’t think Apollo is a great business and the market sure doesn’t either. For profits in general have many attractive financial characteristics, but I think the ones with subpar value propositions like Apollo will continue to carry stigma with them well into the future. If Apollo’s business stabilizes 3 years from now, the stock will trade at a low multiple, but not as low as where it’s at now. Perhaps 7x EBIT, which implies 9x FCF is appropriate. That’s a valuation that simplistically bakes in no growth – it actually assumes a slight decline terminally if you think about 10% as the return you’re looking for with a FCF yield greater than that.
Apollo has $750mm in net cash right now. Yes, that is almost the entire current market cap. It also excludes about $200mm of restricted cash that the DOE requires for-profits to keep restricted as it represents Title IV funds disbursed in advance of the related credits being earned by students – kind of like billings in excess of cost for EPC contractors, which I discussed in myrecent article on Argan (NYSE:AGX). By my calculations, Apollo could generate close to $300mm in FCF over the next 3 years, leaving them with over $1B if they do nothing with it. But Apollo has historically demonstrated a willingness to back the truck up in buying back their own stock. They’ve bought back a total of $1.8B over the last 5 years. They seem to be doing more overseas acquisitions now, but I think there’s a good chance they continue to repurchase shares heavily, to the tune of $500mm over 3 years.
With all of those assumptions, I’m getting the stock quadrupling over 3 years. What’s the risk (again, just operational risks for now)? Perhaps the revenue decline accelerates, operating margin declines further from the very low FY16 level, the business is still doing poorly 3 years from now and trades at a very low multiple, and they don’t do much in the way of repurchases. Even in that case, I can’t model a negative return, and that makes sense. Apollo’s cash is essentially equal to its market cap and, pending something disruptive outside the operations, it’s likely they’ll still be in business 3 years from now and cash flow positive.
Wow! An expect double or triple with no downside risk! We should all go 100% in Apollo, right? Right??
Unquantifiable Regulatory/Legal Risks are Another
The above is all based on the assumption that Apollo’s business is not disrupted by an outside force, which is a major assumption in this case. There are a lot of external events/forces that could really hurt Apollo, and the probability of at least one or two occurring is fairly likely. In reading Apollo’s latest 10-K, I honed in specifically on the Risk Factors section and the Commitments and Contingencies and Regulatory Matters notes to FS.
To be honest, I was expecting worse, but it is still very bad and there are risks not discussed in the 10-K too.
UoP’s 90/10 is at 80% so they’re good there. None of the current lawsuits seemed to be major risks and many detailed in the 10-K have already been dismissed or settled. Accreditation was approved until 2022-23 a few years ago and Apollo was removed from Notice status earlier this year. They will have an interim comprehensive evaluation in 2016-17 though. Gainful employment rules will probably affect programs with 10% of UoP’s students, but these programs have already been discontinued and are being taught out. Apollo and UoP’s ED Composite scores are both well above 1.5. 3-year cohort default rates are well below 30% and are now actually below the national average:
The probationary status from DoD announced in October only affects 1% of revenue. None of these seem like major concerns to me.
The biggest things that are concerning to me in the 10-K are the investigations by the California Attorney General and the FTC. The company received a subpoena from the Cal. AG in August, shortly after receiving a Civil Investigation Demand (similar to subpoena) from the FTC a month earlier. Federal regulators and some state attorney generals (certainly the Cal. AG office given their instrumental role with Corinthian) are now collaborating and it looks like they have Apollo on their radar. The Cal. AG subpoena requested information related to UoP business practices with regard to Cal. National Guard and US military students. It mentions the “use of U.S. military logos and emblems in marketing, for the time period of July 1, 2010 to the present.” The FTC investigation seems to be very broad and essentially asked for all the information UoP has.
I’m not sure what the regulators are looking for and if they’ll find anything significant, but if Corinthian Colleges (COCO, OTCPK:COCOQ) is any example, they have a lot of discretion in how they treat Apollo. With Corinthian, the DOE essentially issued a deathblow by delaying transfer of Title IV funds for 3 weeks at a time when the company already had liquidity issues. Apollo does not have liquidity issues and could handle a decent sized fine, but restriction of Title IV funds would likely be a deathblow as 80% of revenue is funded from Title IV and most students wouldn’t be able to continue without it. The decline in enrollments and revenue would be so violent that Apollo definitely would lose money.
Regulators might be a little reticent to be that harsh with Apollo because:
- they’ve already set a major example with Corinthian
- there is growing pushback in Congress led by Republican Senators John McCain and Lamar Alexander
- and putting Apollo out of business would be very costly in terms of student debt forgiveness – to the tune of $40B in costs to taxpayers.
Some investors may argue that the cost to taxpayers of all that debt is sunk because the debt already exists, but it’s not because if Apollo stays in business, many students will still graduate, get jobs, and pay off their debt – in fact the majority will (3 year CDR of 13%), and none of the students will have their debt relieved – just some of it will go unpaid. The true cost to taxpayers on Apollo’s existing debt now is probably $5-10B, not the $40B that would be relieved with major regulatory action.
However, UoP and Apollo could be really hurt by another external force that does not involve regulators at all.
There is quite clearly a student loan bubble in the US.
Student loans have doubled as a percentage of GDP from 3.45% in 2006 to 6.79% at the end of 2014 (source: FRED data). If lending decreases materially in some shakeout where Title IV programs take a reduced role, Apollo will be hit much harder than most because its students are not very creditworthy. It’s official CDRs are one thing, but these numbers are likely understated b the short measuring period (many students default after 3 years) and metric-manipulating magic tricks like forbearance. According to a very prominent and comprehensive student loan study:
The University of Phoenix, one of the nation’s largest institutions, had the most outstanding federal student loan debt, with 1.1 million borrowers collectively owing $35.5 billion in 2014. Among a 2009 cohort of University of Phoenix borrowers, the researchers estimated, about 47 percent had defaulted within five years.
Many wouldn’t be able to get loans and those who do wouldn’t get them at anywhere close to the current rates (~3% subsidized, ~6% unsubsidized) given the credit risk they represent.
I could go on. Obama has proposed free 2-year community college, which would disrupt much of Apollo’s student demographic. Online/distance education is becoming very feasible for non-profits.
These scenarios are difficult to quantify, so let’s just say Apollo is a zero in any case. I wouldn’t say these are sure things. FDLP could continue for the foreseeable future. There are huge negative externalities to disrupting the stream of educated graduates in the US and that could cause inaction. The investigations could find nothing, and even if they do, it may not result in major damage. A recent FTC investigation of another institution only resulted in an $11mm fine. But let’s just pull the bear case out of the model and say there is a 50% chance of Apollo going to zero.
In that case, the stock still looks set to deliver good returns, but starts to look a lot more risky.
Charlie Munger proposes investors adopt a “two-track analysis.” The two tracks run simultaneously, constantly questioning:
- what are the factors that really govern the interests involved rationally considered?
- and what are the subconscious influences where the brain at a subconscious level is automatically forming conclusions?
With Apollo, the psychology part is challenging because it’s hard not to anchor on the initial model and ignore the risks common sense tells us are obvious, just because they aren’t quantifiable. One of my requirements for large prospective investments is that there’s no foreseeable (model-able) risk of loss. The risks mentioned above are foreseeable, but not quantifiable, which makes them very difficult to think about and deal with appropriately.
I’ve done a lot of thinking about Apollo, and don’t think I will ever own more than a tiny position in it because of the risks discussed above, but I’m okay owning a tiny bit because of my answer to this question: Would I buy the entire business?
Pulled directly from my notes:
If the purchase of the business represented a substantial portion of my net worth, then no way. There is definitely risk of capital impairment here even if the model doesn’t show it because there are regulators that can effectively shoot and kill Apollo with lots of discretion on whether to pull the trigger and why, but if it was a smaller percentage of my net worth like 3-5%, I’d probably go ahead because if I could buy the entire company, I’d be able to influence capital allocation and strategic decisions. I’d pull out more than half the cash in the form of dividend (share repurchase is better with it in highly liquid public market) and then my downside would be limited to like 20-30% of my initial capital and I’d have potential for a triple or quadruple on conservative assumptions. Further, I could greatly increase the probability of getting that triple, or even more, by making dramatic shifts to the company’s value proposition. I’d greatly increase continuation and merit-based scholarships, become much more selective with prospective students, focus on channels that bring in higher quality students like corporate partnerships, cut out any advertising and business practices that pose regulatory risk, and take other aggressive actions.
The idea that Apollo is attractive as a very small investment as part of a diversified portfolio is supported by some of the company’s largest shareholders. Prominent investors like Larry Robbins, Joel Greenblatt, Donald Yacktman, and Ken Griffin all have long positions in the stock, but in each case the position is a fraction of 1% of their portfolio:
Apollo is a very high risk investment. There is potential for complete capital impairment, but it also has multi-bagger potential, and on balance, it looks attractive. As such, it is not a good idea to hold any more than a tiny position in the stock, but that tiny position is worthwhile. Apollo is the kind of stock that drives the performance of systems like Magic Formula because no rational investor holds it in size, which distorts supply-demand dynamics until something happens. Let’s see what happens.
Editor’s Note: This article covers one or more stocks trading at less than $1 per share and/or with less than a $100 million market cap. Please be aware of the risks associated with these stocks.