Description

Recommendation
HMH is a business that is finally recovering and more importantly is amidst an accelerating digital transformation.  The market doesn't fully understand the accounting effects of this digital adoption, and you can create the company at 5.2x cash post plate EBITDA.  I estimate the company will generate $2.25 of FCF/shr this year and growing going forward and will have greater than $3.50 of net cash per share at year end.

Description
HMH is the largest K12 publisher of educational materials.  PSON and McGraw Hill are the next largest players in the market.  HMH primarily sells into the Basal educational market in the US.  These are the textbooks, workbooks, and digital tools that your kids are likely issued by the schools they attend.  

Market
Although certainly smallish considering the gross amount of money spent educating our 50mm plus kids in the US each year (est at approx. $640B) it's estimated to be over $3B this year.  The market as estimated by the American Association of Publishers (AAP) was growing and averaged plus or minus $4B from 2000 to 2008.  Starting in 2009, municipal funding pressures slowed spending and the market fell back to the mid to high $2B range.  To best understand the dynamic here, one must be familiar with two issues.  First, how do schools select and pay for educational materials?  I will be brief and just highlight the most critical issues.  There are adoption states and so called open territories.  For the purposes of this point, what is relevant, is the sales process stretches well over 1 year.  In 2008 the business was fine, it didn’t really start to struggle until 2009 – point is this market doesn’t turn on a dime. Decisions are made to open up for an adoption, then proposals are requested and made, samples reviewed and evaluations are completed prior to a "win".  Depending on the state, that just means the provider is then on an approved list.  In most states that still means that districts could actually make selections and purchases over a several year period.  Second, post 2008, schools districts had reduced funding and were predisposed to deferring spending on new educational materials.  Making it easier to rationalize was the known upcoming emergence of the common core.  Common core is the new initiative that essentially all of the states have now adopted (44 of 50 by some measures, but some states have adopted it just renaming the standards for political reasons) which attempts to raise the bar for our K12 students and narrow the education gap that has emerged vs other countries throughout the world.  So imagine the ease of rationalizing the deferral of spend in 2011 and 2012.  Muni budget are just starting to improve, first extra dollars are allocated to rehire teachers, the common core curriculum is on the verge of really rolling out in 2013 and 2014, is it that hard of a decision to defer one more year? 

Based on commentary from HMH and PSON the market has finally started to bounce back this year.  HMH has said their addressable market, which is not identical to the K12 AAP market, but trends similarly, will be over $3B vs the $2.6B in 2013.  Through August, AAP corroborates, suggesting that the market is up ~17% this year (importantly, July, August, and June are, in order, the three largest months of the year).  Further HMH claims to have picked up very significant share in the "new adoption" market.  They estimate this portion of the $3B plus mkt to be around $900-950mm this year.  What is most interesting is that in the 2nd quarter, HMH, with 90% of decisions made by that time, claimed to have won 50% of the new adoption market this year vs 37% last year.  Further commentary also seemed to suggest that they were slightly outperforming in "open territories" (the areas where adoptions are not made by the state level, which includes the likes of NYC).  This is important because if one was to assume ONLY that HMH picked up 1300bps on 90% of the low end of the new adoption market of $900mm, that is $105mm in sales above the market growth rate.  $105mm in incremental education sales over the $1208mm of education segment sales in 2013 is 8.7% above market growth.  So this implies nearly 25% revenues growth yoy (8.7+ 17% of market growth alone ignoring any share growth in open territories, or better new adoption state performance).  How does my math compare to 1H results.  Well the company reported 40% order growth yoy in the first half of the year and billings were up 31% in 2q yoy.  Sounds directionally similar, right?  Well why is their guidance only for 5-8% revenue growth and why does the street remain at the high end of the range? I’ll come back to that.

One good question is to ask whether or not this is an overshoot, or 1x anomaly. I think historical context for the number of years of below trend spend and a couple of conversations with people in the industry who are familiar with the state and age of the “stock” in our schools will quickly dismiss you of that notion, however I suggest the following thought experiment. The K12 population in this country is known and predictable, and continues to grow and is projected to grow for the foreseeable future (~60-70bps/yr). So there is the quantity. Next up is pricing. A conversation with anyone in the industry will tell you that pricing is very stable and does not generally determine outcomes, its down the list as a determining factor. The market is consolidated, there aren’t new entrants vs the big content players, barriers are still quite high, and pricing pressures are not apparent. So no meaningful new entrants winning share, corroborated by lack of pricing pressures, 5yrs of well below trend market sales that for reasons that are intuitive, coupled with a secular move to more digital content. Where is the long term bogeyman here? Further, digital provides incremental sales opportunities both inside and outside the school, software upgrades, more features, related training and services that HMH will sell to school systems etc all providing more revenue opportunities. Also there is increased focus and opportunities to take this content and tools abroad.

Accounting confusion
This gets right at what I think is the poor understanding of HMH's business and its results this year to date. Principally it’s the misleading nature of GAAP for this company.  One of the basic tenants of GAAP is the matching principle, which is where revenues and expenses are to be "matched" together so as to reduce distortions in reported results.  This is breaking down when it comes to digital revenues for HMH.  They recognize revenues straight line over 6 to 8 years while expensing their costs upfront (sales commissions and royalties are expensed with the cash payments).  These digital revenues manifest themselves in the change in deferred revenues.  And critically they collect all of the cash upfront under regular collection terms.  This has become a bigger deal this year with digital sales ramping.  So not only are GAAP revenues understated, the GAAP margins are reduced further do to full period expensing. The company will tell you that if their business was to ever get to 100% digital (which it won’t, but to demonstrate the point as it grows), they might pick up 20 points of margin or more! The company has said as much when asked why they didn't bump up their guidance, they said with so much of their growth being digital, their GAAP results weren't going to keep up with their order growth.  In addition, when asked, they didn't provide "guidance" per se but said that the pace of deferred revenues was going to be similar for the rest of the year.  This makes sense as the third quarter is the biggest of the year. Keep this in mind when looking at street estimates that imply negative 2^nd half deferred revenue growth.

Similar issues
Cengage and McGraw Hill Education both serve the higher education market.  This truly is a different market, but they also are starting to sell more digital products all the time.  They both report "cash" revenue and "cash" post plate EBITDA.  There are two big differences though.  One, they recognize their digital revenues over one year (a code purchased by a freshman expires prior to sophomore year).  Two, their stakeholders view "cash" post plate EBITDA and FCF as the relevant metrics to measure the economic value add in any time period.  The definition used for "cash" revenue is to take GAAP revenue, and add the change in deferred revenues over the relevant time period.  "Cash" post plate EBITDA is defined as GAAP post plate EBITDA (post plate just reduces EBITDA by regular and ongoing expenses that are identified and capitalized as “plate” capex), and again add in the change in deferred revenue over the relevant period.  Why are these adjustments used for these other companies?  Because its a better proxy for cash flow, and a better estimate of the level of economic activity in any time period.  Inarguably these adjustments are much more important with respect to HMH as the gap between GAAP and reality are much greater due to significantly slower recognition of these revenues in HMH's case of approx 7 years on average vs only 12 months for the higher education businesses.  This gap is logically greatest during significant changes in trend of the percentage of digital content, such as what is occurring in 2014. 

The street
The sell side equity research community isn't as familiar as the credit markets with these accounting issues.  There really aren't good public comps for HMH.  Some point to PSON.  There is definitely a division of PSON that is a good comp, but it winds up being pretty futile to attempt to tease it out.  HMH has a larger K12 business, but PSON has a TEV over 4x larger.  PSON has a huge K12 assessment business, has a higher Ed business, owns and runs entire schools all over the globe, has a LMS offering, does ERP like software for online programs at colleges, and a traditional publishing business among others.  Their disclosure is very poor and not at all granular.  Using this as a comp is like using AMZN to comp Barnes and Noble (AMZN sells a bit more than books now).  SCHL is a poor comp.  It's mostly a traditional publisher and there is no sign that their business is going through anything like the changes that HMH is going through (their deferred revenue is stable).  Some point to Wiley, which is really a stretch.  It's a collection of scholarly and professional titles with a higher education business.  Again, not the same drivers at all, nor the similar divergence with reality and GAAP.  So rather than just do the unthinkable and think about valuation on an absolute basis, the street has continued to value HMH off of misguided math and to poor comps, preferring EV to post plate GAAP EBITDA. Which frankly up until this year was fine as there weren’t meaningful changes in deferred revenues over the course of the last couple of years – thus GAAP was a decent proxy for reality. This is no longer the case – see 2Q results, whereby billings grew by 31% yoy, deferred revs are up over $100mm in the first 6mos, and A/R has exploded – but importantly the A/R will be collected back to normal levels by year end.

Valuation errors
There are a number of them, but the two largest and most common are the following.  1) is just an utter reluctance to deviate from official guidance and recognize the "cash" revenue the company is going to generate, which can be estimated using only 3rd party information and the company's own disclosure and an understanding of their business cycle/model.  When adjusted for this (ie adding back change in deferred revs for the year of 2014 to modeled GAAP revenue), most street models suggest 13.5-15.5% "cash" revenue growth this year.  I find this hard to reconcile with market growth exceeding that, and even harder to understand with their demonstrable share growth above and beyond that.  Nearly every street model has deferred revenue for the year that is less than the growth in the first two quarters!  I believe the street is at least 700-1000bps too light on implied "cash" revenue and over 100mm short on FCF. Push this threw a model with a lot of operating leverage and impact will become apparent. The street’s estimates for FCF generation are either way too low, or everything I just went through is off, there isn’t a middle ground.

2) the shocking confusion of the company's regular, predictable working capital cycle.  The company ended 2013 with 425mm of cash and 246mm of debt.  I am unaware of any analyst, even the most conservative or bearish, who doesn't have the company generating meaningful FCF this year, yet most published reports calculate TEV using the reduced cash balance of the 2Q of and thus inflating TEV by 230mm (or a whole turn on GAAP PP EBITDA) due just to the normal and predictable working capital cycle of the first 6 months of the year, which when recalculated at 3Q will shrink if using the same method due to the cash flow that is about to pile up. 

Model
The 80/20 rule applies here. So rather than share my model, I’m going to just make a few “corrective” adjustments off of the numbers out of some published research, effectively the consensus model. The margin for error if one believes these corrections are justified, overwhelm and should be more than plenty to make a neutral person quite bullish. When reviewing the numbers, ask yourself if the street estimates make sense next to the company’s view of 17% market growth, with them taking significant share. Does it makes sense that a business with so much operating leverage should only be growing post plate EBITDA low double digits on a yoy basis? Ask yourself, should FCF be up more with orders up 40% in the first half and billings up 31% in the 2^nd quarter, which scenario seems more likely?

Price target

I believe 12x FCF/shr is well on the conservative side of reasonable for a company like this. That would still make it more cash generative per share then all those so-called comps that I dismissed earlier and is very significantly on the low end of anything in regards to historical context. On the very low end and rounding down, 12x my 2014 $2.25/shr is $27, and when adding back the over $3.50 of net cash get you a bit over $30. With essentially no fundamental downside to the consensus view, and the likely initiation of some amount of capital being returned to shareholders after year end, one can easily get to higher valuations with a more efficient capital structure and/or improved visibility to growth in the B2C endeavors. If they company was to put just a turn or two of net debt on the company, this gets to well over $30 quickly.

Catalyst

I believe this will become more apparent this thursday 11/6 on the 3Q earnings call.