J2 Communications (JCOM) - $30/share, $1.5B mkt cap, trading $12m/day

Short.

First – I should point to an earlier post on JCOM by Scott 737 a year and a half ago.  I agree with his thesis and think the post is a good read.  The short has performed reasonably well with JCOM gaining only 15% in an up market.  This lackluster performance came despite JCOM making the final large fax acquisition available to be made (which increased revenues 40%) – the acquisition was highly accretive since JCOM can pay a strategic buyer premium and still purchase a fax business at a ~50% discount to JCOM’s inflated market valuation.  

JCOM looks like an attractive short.  JCOM is a marketing company selling a commoditized (but sticky) product that is already obsolete and slowly being switched off.  JCOM charges the highest prices amongst its competitors (50-100% premiums for the same product), is reportedly very aggressive with customers (overcharges, refusal to cancel), and has a number of accounting red flags.  Management is very salesy in promoting this fax business business as a “cloud play” to investors, while refusing to answer the most basic questions about organic growth.  JCOM has maintained growth in the downturn by making overpriced acquisitions of competitors and essentially capitalizing subscriber acquisition costs into goodwill.  J2’s acquisition prices of perfectly comparable businesses (at strategic takeout prices) imply JCOM is only worth ~$22/share, and I think JCOM overpaid for those businesses. 

The short thesis for JCOM falls into a few buckets:

• Secular decline on core business with very high incremental margins – 80% of revs are efax, which will go to zero, the only debate is over what timeframe – 5 yrs? 20 yrs?  While JCOM refuses to acknowledge this decline or discuss organic growth, it appears this business is already declining between 2-5% per year organically.  This decline is on a business with ~80% incremental margins, while JCOM tries to fill the revenue hole with subscale, high cost, international growth and other junk products at much lower margins.  Even if JCOM can keep topline flat profitability will erode, and I doubt they can keep topline flat. 
• Business practice questions – I think you tend to not get paid for these issues (see MLM companies), but it’s usually a good indicator of a weak or unwanted product offering.  Search Google or Youtube for terms like “Efax cancellation scam”, “Efax fraud”, “Efax overcharge” and you’ll see some of the complaints they’ve had.  Videos can show you interaction with customer service, inability to cancel accounts.  To be fair, we had no problem cancelling a test account a few days after opening it, but I think that ease of cancellation may be different after 4 mo (when JCOM starts to count it as churn), or around quarter end (when JCOM is trying to manage sub numbers), and we plan to test more accounts.
• The real kicker for me is a number of accounting, presentation and governance issues – quite a few red flag accounting issues that indicate JCOM’s earnings and key operating metrics may be significantly inflated.

In my opinion JCOM’s decline is inevitable.  Organic customer adds are still profitable, but there just aren’t enough new buyers for fax services to replace the number of people shutting their fax services off.  I don’t think this business is worth the 12x consensus earnings multiple it trades for.  I think a more appropriate valuation would be something like Gamestop’s 8x or Vonage’s 6.5x, providing opportunity for a ~30-50% gain on the short.  I should note ~20% of the float is short, but it’s still a widely available with cheap borrow. 

Note: JCOM has actively acquired internet fax competitors.  In FY10 it spent a total of $250m on acquisitions including the $210m acquisition of Protus (#2 efax competitor to JCOM), its largest acquisition ever.  2010 revenues were $255m and 2011 revenues were $340m after accounting for acquisitions.

I’ll run through a quick business description before getting to the accounting issues which I think are the most interesting angle to shorting JCOM.

JCOM Business

• J2 Communications provides internet fax and related services.  Efax is 80% of revenues and a higher percentage of profit.  Efax is a service that allows you to send and receive faxes as PDF’d email attachments (seriously). 
• J2 provides 13.1 million DIDs (Direct Inward Dial phone numbers) to customers.  2m DIDs are paid for by customers and 11m are provided for free with limited services.  DID-based revenues (fax and phone together) are 93% of J2’s revenue
• Paid DID customers generate $315m of Jcom’s FY11 revenue
• Free DID customers generate $3m of revenue (headed to zero)
• J2 provides most customers a JCOM-owned DID – this is the key stickiness factor, if a customer wants to leave they can’t take the fax number with them, which is a barrier to exit only if the customer intends to keep a fax line
• This is the most attractive selling point for J2.  However this only applies to “retail” customers coming to J2 through the internet, which is a shrinking segment.  Most existing businesses coming to J2 will port in their own existing numbers, which they can take with them when they leave.
• 80% of J2’s subscriber revenue comes from fixed monthly fees, 20% is variable based on usage
• Geographic split – formerly 85%US/ 15% ROW (mostlyCanada), now 60%US/ 40% ROW due to Protus’ large Canadian presence ---Canada& US together are likely more than 90% of the business. 
• J2 value proposition
• For individuals / small businesses – JCOM provides fax lines for $16.95/mo, cheaper than the $30/mo telco charges for a fax line
• Corporate – Some IT departments do not want to replace old fax servers at end of life because fax is becoming obsolete.  Usually they can just extend life or maintain declining capacity on working fax servers.  In circumstances where they cannot some corporates consider internet fax as an alternative.
• JCOM offers corporates ~$10/line to corporate customers and the former head of corporate accounts said they’d do $4/line for really high volume customer, but they don’t tend to get those customers
• J2 competitive positioning
• Retail- J2 is the high price point in a commodity market with a terrible reputation.  They charge $16.95 for an indistinguishable service the competition prices at $5-10 
• DID-line stickiness has been their fulcrum – in the past they have jacked up prices on their customers in large increments (Feb ’01 increased from $4.95 to $9.95, June ’03 to $12.95, Aug ’06 to $16.95), and the general analysis on the price increases has been 'we’ll increase prices 30%, see 20% customer churn, and come out ahead'. 
• This stickiness allows JCOM to squeeze the last dollars out of a customer base that hates them, but certainly does not allow for much growth or attract new customers. 
• Corporate -
• SME – Similar to the retail market, JCOM is the high price competitor with a terrible reputation.  I called a few companies to price a 5 fax line need, JCOM quoted me $10/line and gave no sell job, RingCentral quoted $4.50/line and worked for the sale.  I tried to give JCOM a chance to match the price, it took an entire week for the salesperson to call me back then they would only come to $8/line.
• Large Corporate – I don’t think JCOM has a serious position in the corporate market, but competition is stiff in this market as well.
• Other “simple” service competitors have better prices and more inroads to corporate markets (Easylink, Retaris)
• Most real “corporates” would want the customization they can get from the fax server manufacturers that now provide outsourced/cloud services (Biscom, Sagem, Esker, Captaris). 
• Technological barriers to entry?  None. 
• A capable college programmer can set up a this service in a few weeks.  DID numbers are easily obtained from CLECs.  The industry is scattered with 1-2 person operations. 
• Patent barriers to entry? Seems more like a hassle factor than a real legal position, but the hassle is probably an effective enough deterrent to keep people from trying to enter a market with an obsolete product in terminal decline. 
• J2 has brought patent lawsuits against a number of competitors.  J2 hasn’t won or lost a case outright yet.  The most high profile case was Venali, which dragged on for 7 years and ended with J2 purchasing Venali (at what appears to be a high price)
• The credible threat of patents lawsuits might be a deterrent, but it is clearly not a real barrier judging by the number of small competitors on the web.  It is easy to set up an internet fax shop.  The greatest barrier to entry is that no one wants to enter a declining fax services market
• Customer perception?
• JCOM has questionable customer practices – the internet is littered with accusations that JCOM adds charges to customer bills, refuses to let people cancel. 
• My personal experience was not as bad -- I signed up for an account and cancelled 5 days later.  It was not easy, but I was able to cancel. 
• I was charged $26.95 versus my expected $16.95 ($10 setup fee not clear when ordering). 
• Can this market grow?
• The entire fax market is certainly in decline, though participants argue fax lines will decline slowly as companies want customers to be able to contact them in any format they choose. 
• The question is can internet fax convert traditional fax customer fast enough to offset the decline in fax lines.  I haven’t been able to find any good data on the fax market or internet fax market, but working with Gartner I’ve been able to make some assumptions that I think should give reasonable sizing:
• I think a reasonable estimate for the high end of market size is ~25m US fax numbers required, possibly 18m candidates for a paid service, with internet fax already penetrating ~50% of that market (see appendix)
• I do not think the rate of conversion from traditional fax to internet fax is high.  Most companies just don’t care about fax.  I also think my estimates of 50% penetration are surprisingly high, this does not have blue sky growth potential.  Most importantly, taking share in a declining market will eventually be a losing proposition (by definition its negative 2^nd derivative growth).  I think the internet fax market is still growing at this point, but that all the growth is going to competitors with market-rate pricing (one guy I spoke with estimated he’s growing accounts at ~5%/year, others mostly flat or slight decline)
• Is JCOM growing organically?

I think JCOM is organic decline – though it is difficult to see through the acquisition noise.  Management won’t discuss organic growth or give good disclosure on acquisitions)

• I believe paid subs have been roughly flat once acquisitions are stripped out
• Since 2006 subs have increased 998k to 2m, I estimate 950k subs have been acquired
• That estimate might substantially undercount acquired subs if mgmt is managing the numbers as I suspect (see below)
• ARPU is declining organically, accelerated by acquisition mix
• Since 2006 ARPU has declined from $16.45 to $13.30
• JCOM’s stated service price is still $16.95, but mix shift is moving towards customers paying a discounted ~$10 level that can be negotiated with JCOM (still priced too high versus competition priced as low as $4). 
• Management says they are considering no longer reporting ARPU.  This is a bit of a red-flag given 93% of revenue is still directly tied to DIDs
• Other operating metrics
• Churn has hovered between 2.5 and 3% and is currently at 2.5% (though this number appears managed by mgmt as discussed later)
• Margins have improved, but JCOM has boosted margins by using acquisitions to capitalize SAC. 
• Est. subscriber acquisition costs have stayed flat just under $100/sub (after spiking to $110 for a few years), however it is difficult to discuss SAC with certainty because of acquisitions. 
• Increasingly subscribers have come through acquisitions priced as high as $400/sub
• Gross margins have improved from ~80% to ~83% since ‘06
• Total EBIT margin for JCOM has improved from ~43% to ~47% since ‘06
• Operating cost per sub has declined from ~$7.27 / mo in FY06 to ~$5.30 currently, offsetting some ARPU declines
• Reported operating numbers appear massaged
• JCOM’s sub counts and churn reflect management estimates and are footnoted that the subcount includes “reserves for anticipated product migration, price increases, one time adjustments”.  Despite the simplicity of counting paid subscribers at quarter end, management has decided instead to massage this number and use their discretion in choosing what paid subscriber count to report to investors. 
• Paid DID base remained exactly 1.274m throughout 2009.  IN 2009 JCOM was churning ~10% per quarter (3%+ per month), I find it hard to believe the ending counts on the last day of each quarter never deviated by as little as 0.078%.  This appears to be a managed number. 
• In 2006 JCOM raised prices 30%, but churn only increased from 2.6% to 2.9%.  I’ve heard from a competitor that the price increase was a huge windfall in new customers, that doesn’t seem reflected in churn. 
• Management takes large reserves on subscriber counts for acquired companies (5-30% they say if you ask them directly), those reserves then allow JCOM to under-report churn, and to release undercounted subs back into the DID count over time to give the appearance of greater subscriber stability and lower churn. 
• JCOM reports churn only if a customer has been with them for 4 months.  So real churn is much higher than what is reported to investors.  However they count all customers including those under 4 months. 

• Customer model
• I do think organic adds around $100 / sub are profitable for JCOM, I think the problem is JCOM has difficulty attracting new subs.  On my customer models I think JCOM earns ~60% cash IRR on a successful new sub acquisition, but cannot organically find enough new subs to replace churn, so it’s a moot point. 
• I can’t see how acquired adds are attractive at $350-400/sub.  For instance the Protus acquisition cost $207m, and I assume it added 620k subs, despite the fact management says it added only 500k subs (500k subs appears impossible if you look at implied revenue/sub vs. Protus’ product prices --- it appears management undercounted subs significantly to give themselves a cookie jar from which to draw fake sub growth in future periods and under-report churn). 
• But even using 620k subs it is hard to see a run-off valuation @ Protus’ ~$10 ARPU that nets more than $177m in undiscounted FCF, even assuming low 2.3% churn and 80% margins (JCOM margins ~48%).  Even assuming management undercounted subs by 100% and there were really 1m subs adds at extremely low churn and high margin you cannot justify a return on the purchase price
• Also interesting that not only does the Protus valuation appear unjustifiable, but applying that valuation to JCOM suggests JCOM stock should only be ~$22/share, ~25% downside.

Acquisitions

• JCOM is highly acquisitive.  These acquisitions are necessary to maintain a modicum of topline growth, and they also offer ample opportunity to massage earnings and key performance metrics.
• Management has started paying increasingly high prices per sub for acquisitions (~$400/sub now vs. $60/sub 5 years ago by my estimates).
• What is most telling about these acquisitions is that they are perfect JCOM comps at strategic, synergistic prices.  And the implied value for JCOM is still only ~$22/share including cash, 25% below market value. 
• Management would tell you they got Protus cheaply because of the ongoing patent lawsuit (though they were 6 years into the lawsuit with no clear judgment)
• I would argue JCOM wildly overpaid as they can’t even recoup purchase price running off that customer base on very aggressive churn and margin assumptions. 
• You could argue these acquisitions make JCOM substantially overstate earnings because they are essentially capitalizing subscriber acquisition costs into goodwill (see model)
• All JCOM is buying is a subscriber base, which is then moved onto JCOM’s system.  So money spent on acquisitions is essentially just SAC, which you could argue should be amortized over the ~3yr average life of a customer
• But what JCOM actually does --- over the past 7 years, ~80% of the $340m JCOM has sepnt on acquisitions has been stuffed into goodwill where it will never hit the income statement.  The remaining acquisition price was put into customer relationship intangibles where they are shockingly amortized over 7-8 years which makes absolutely no sense in a 2.5% monthly churn business which implies a 3yr average customer life. 
• This is similar to what Tyco did with ADT during it’s accounting scandal (the mgmt compensation was just headlines, the real accounting scandal was around ADT) – buy customer accounts so you can capitalize subscriber acquisition cost instead of expensing it, which overstates profit. 
• Most importantly – JCOM is now running out of acquisitions which will be a huge problem for them
• Previously was able to buy fax businesses at half of JCOM’s market valuation, going forward there are no more good targets.  JCOM now is buying other junk service companies with no barriers to entry or differentiated product, and they’re buying in spaces that command higher TEV/revenue multiples with much lower margins – online data storage, email marketing, etc
• JCOM needs to spend these dollars because despite having >80% of revenue in US & Canada JCOM has a ~25% tax rate, with most cash flow trapped offshore.  Protus was Canadian based so it could utilize offshore cash, most other recent acquisition were also offshore. 

Non-Fax Services

• ~20% of revenues are non-fax services
• ~13% voice – automated menu services and call routing (press 1 for Jill, 2 for Frank). 
• This is not a VOIP service, just an automated system that routes calls to traditional phone lines, you still need to pay the phone company now you just pay JCOM as well
• This is a crowded business space with no barriers to entry and the same product offered by most telco’s.  JCOM’s voice has grown through acquisition not organically
• ~3% email – enables @yourbusiness.com email addresses, spam filtering etc.
• Most customers appear to be buying virus protection, doesn’t seem to be any value proposition at all, just a marketing business
• ~3% campaigner – email campaign manager, management likens to constant contact, this one just doesn’t grow like constant contact b/c it is overpriced and JCOM doesn’t pay to market and grow it
• <1% data storage.
• These services are commoditized and similar to efax are kinda “junk” services offered by a lot of small players.  Generally I think these businesses have low barriers to entry and high competition.  They are growing off a smaller base, but I do not see a hidden gamechanger here.  These prouducts come with much lower margins than the fax business JCOM is churning and I don’t expect they will provide enough growth for JCOM to organically grow topline. 

Mgmt’s growth strategy

• JCOM’s growth strategy seems to shift on every conference call.  In 2Q11 it was about growing in enterprise, that didn’t seem to work, next conference call it was about growing internationally --Japanup 100%! (Of course that’s on a base of 10k customers with negligible revenue). 
• On the latest call JCOM spent a lot of time talking about cross-selling --- Selling existing fax customers other JCOM services. 
• First, ask management where the revenue from cross-selling is and they’ll tell you – “cross-selling” is really “cross-giving-away”.  They are just trying to give away other free services hoping they can force customers to upgrade later, that’s what they are counting as a “cross-sell”.  Amazingly they have a low hit rate and largely can’t give away these junk services for free. 
• Second, JCOM is not a vendor with a real relationship with their customers, it’s not “Hank our JCOM rep is in the office again this week, let’s see what new innovations he has to sell us”.  I think the customer relationship is more like receiving spam and telemarketing, for clients who may not really want JCOM fax anymore in the first place trying to sell them extra services.  This does not seem like a gamechanger to me, but I’m sure on the next conference call there will be a new growth plan. 

• Management appears to manage subscriber count and churn as noted above
• Earnings may be materially overstated as subscriber churn is replaced with acquisitions, which capitalizes subscriber acquisition cost instead of expensing it.  Theoretically, if acquisitions were fully expensed over 3-year customer life it would add ~$80m of expense to a company with ~$120m earnings base
• In 1Q11 mgmt swept a $10m deferred revenue accounting charge under the rug without so much as a question from the sell-side.  This was because they had incorrectly realized $10m of pure-profit revenue in the past. 
  • Looking at the balance sheet (days deferred rev movements), it appears this improper recognition of revenue came largely in the past 3 years.  $10m of pure profit is very material set against an annual earnings base of $75-90m in that period. 
  • Management’s explanation of this charge was ridiculous – They said the $10m adjustment came after a systems upgrade made them realize they had been underaccruing deferred revenue on annual pay customers.  But the $10m adjustment was to a balance sheet account of only $12m total deferred revenue (look at deferred rev on balance sheet pre-Protus acquisition).  It is hard for me believe it’s an innocent mistake to be off by a factor of 2 times on such a simple calculation that just would have happened to inflate your earnings ~5% and prevent you from missing earnings estimates. 
  • ~15% receivables allowance is remarkably high for a business with automated credit card billing (indicative of high chargebacks, a lot of customers getting charged who dispute whether they should be charged)
  • Options backdating scandal uncovered in 2006, had to restate
  • Non-Big 4 accountant -- Singer Lewak.  LA firm, 85^th largest accounting firm in US according to Inside Public Accounting. $36m in revenue ($850k from JCOM)
  • JCOM saw a $13m tax settlement in ’10 for ’04-‘08 periods, with open audits on other periods.  I think this is notable as it is hard to justify a 25% tax rate on a business with 80% of sales in US &Canadawith most of the remainder in theUK. 
  • Related party trx:
    • JCOM leases HQ from a chairman-affiliated company for $1.3m.  Interestingly, that lease cost increased from $800k in ’06 to $1.2m in ’08 but the location of JCOM’s headquarters didn’t change.
    • Consulting arrangement with a chairman-related company to pay $300k/year for financial consulting. Has a consulting agreement with director John Rieley to pay $100k for consulting on public relations for voice services. 
  • Director compensation is well outside the norm - 7 directors earning $450-675k / year on a company with $340m of revenues.   This compares to average director fees for companies with sub $3B in revenue of $126k, average fees for companies with over $20B revenues of $203k/year
  • Insiders are selling lots of stock, but to be fair they have been selling lots of stock for a long time now, no acceleration in sales
  • Stock comp is substantial and real economic cost to the company, of course JCOM adds this back to earnings and free cash flow. 

Bull case?

• JCOM does have $221m cash on balance sheet, ~15% of market cap, but a large portion of that is overseas and I expect the majority of cash generation is overseas (JCOM just used a large slug of overseas cash on the Protus acquisition. 
• JCOM generates a 9.5% FCF yield (8.7% after stock comp expense)
• I think some view JCOM as a stable business, with decent cash flows that could potentially start growing if international/cross-selling/enterprise initiatives take hold.  This view assumes fax is a business necessity which customers will continue using forever.  JCOM’s service is sticky b/c it is a pain to change fax numbers, but it’s somewhat irrelevant as the secular headwind is customers shutting off their fax, not switching providors. 
• Low multiple for a cloud services company.
• Could fax demand accelerate with job recovery?

Bear case

• Difficult to trust JCOM’s accounting – a number of red flags we can see (my two favorites are capitalizing SAC cost and the $10m deferred revenue “mistake”), cockroach theory that there is potentially more going on than just what we can see from the public financials. 
• JCOM has masked their decline with acquisitions which are accretive to JCOM’s inflated market price.  It also appears that acquisition accounting reserves have also boosted profitability in addition to capitalized SAC cost.  Following Protus there are no more huge acquisitions to pursue.  JCOM now has to buy companies in more competitive spaces at higher prices and lower margins. 
• Fax is a declining market, organic growth seems to be negative despite management’s assurances they are growing but refusal to talk about organic numbers.  Fax decline will at some point hit a tipping point and declines will accelerate quickly ala Telex.  Exit of a couple key constituencies (notably healthcare) could be prompted by clarification of privacy rules (some rules were written to explicitly allow fax but have not yet been updated for current communication options such as email so a lot of healthcare companies still use fax).  Digital signature legislation should also pull fax out of the legal profession. 
• In addition to negative organic growth, margins almost certainly will come down.  JCOM is losing fax revenues with incremental margins around 80%, and trying to replace it with lower margin revenue from subscale businesses such as international, data storage, web marketing and email hosting.  Churning 80% margin revenues with low margin revenues should bring margins down significantly.
  • Can see this in JCOM’s FY12 guidance – revenue growth of 1-7%, flat EPS despite a 10% buyback program.  Also worth noting – JCOM says this guidance does not include any “large acquisitions”, which seems artfully phrased to allow JCOM to include assumptions of continued small acquisitions in their guidance in order to achieve some topline growth. 

JCOM is ~12x forward earnings (ex stock comp), 12.5x incl. stock comp, and 10.5x ex-cash.  FCF is inline with earnings.  This is overvalued for a product in terminal decline and I think earnings are overstated by a number of questionable accounting practices.  I think a more reasonable multiple for a secularly declining business is 6-8x.  As the cookie jar accounting from the Protus acquisition wears off and JCOM’s decline is easier for investors to see I expect earnings to deteriorate and JCOM’s multiple to retract to more reasonable levels.