Beauty is in the eye of the beholder.   
-  Margaret Wolfe Hungerford in Molly Bawn, 1878

And the same can be said for financial reporting as practiced by internet companies given their “new business models” that require “new accounting.”  Internet company financial statements seem to mean different things to different people, not unlike a piece of artwork.  Unfortunately, some of this accounting “artwork” is junk, as we have recently reported in the case of Groupon (First 10K: April Fool’s!).  At times like this beauty rests in the I of the artist.

How can management and directors and auditors see one thing, when the complete opposite reflects reality?  And why do internet IPOs seem particularly vulnerable? Well, we think the problem is with the accounting “standards” (and we use that term loosely) that apply to these companies.  As we stated in an earlier post:

Internet company accounting is suspect given all the unsupported assertions and assumptions that must be made to comply with generally accepted accounting principles…

Think about it.  The internet company balance sheet is generally dominated by intangible assets whose values are based on assumptions that are works of art themselves.  And then there’s revenue recognition in these companies with management making all kinds of assumptions about primary obligors, selling price hierarchies, and virtual sales.  Yes, what makes internet company accounting “special” is that so many of the applicable accounting rules require major assumptions, many of which could be better characterized as “giant leaps of faith.”  Clearly, the accounting rules used for internet companies should not be called “standards,” as their many judgments make any meaningful comparison an impossibility!  Enough pontificating…

Given Groupon’s recent accounting struggles we thought it might be interesting to see if there were any other internet company accounting issues lurking within today’s “hot” internet companies.  So, we looked at the most recent 10K filings of Demand Media, Facebook, Groupon, Linked In, and Zynga, focusing primarily on revenue and expense recognition, “unusual” accounting issues, and of course some of our favorites: intangible assets, cash flows, and non-GAAP financial metrics.  Here is what we found.

 

Revenue

Two of the five companies (Demand Media and Facebook) generate a significant amount of their revenue from advertising. The way these companies record revenue appears to be relatively straight-forward.  Generally, ad revenue is recognized either when the ad content is delivered, or for click-based ads, when a user clicks on an ad.  Nothing very interesting or complicated here.

Linked In, on the other hand, has a much more subjective revenue recognition method for its hiring and marketing solutions.  Most of the Company’s contractual arrangements include multiple deliverables, i.e., several products packaged together which Linked In swears can’t be pulled apart to record revenue separately.  Gee, if the Company’s cost accounting system keeps track of product and service costs separately, why can’t revenue be estimated separately? Interesting question, huh?  Anyway, Linked In uses convoluted GAAP criteria to record revenue, the relative selling price method, based on a selling price hierarchy.  In short, management decides what revenue will be based on vendor specific evidence, third party evidence, or management’s best estimate of selling price, in that order of priority.  Which one do you thing management likely favors?

Then, there’s our poster child for bad internet company accounting, Groupon.  As you may recall, the Company was busted by the SEC for improper revenue recognition last September. See “Groupon Finally Restates Its Numbers.”  Basically, Groupon ignored accounting guidance (that’s a much better word than “standard”) in Emerging Issues Task Force (EITF) 99-19, as well as SEC Staff Accounting Bulletin 101 (question 10), and recorded the gross amounts it received on Groupon sales as revenues. Since being forced to restate its financial statements, the Company now records revenue at the net amount retained from the sale of Groupons (gross collections less an agreed upon percentage of the purchase price due to the featured merchant excluding any applicable taxes), since it is acting as the merchant’s agent in the transaction.

It should be noted that Demand Media also faces the “gross vs. net” revenue issue discussed in EITF 99-19.  For revenue sharing arrangements in which the Company is considered the primary obligor, it reports revenue on a gross basis.  But for those situations where it distributes its content on third-party websites and the customer acts as the primary obligor, it records revenue on a net basis.

And last, but not least, there is Zynga with its consumable or durable virtual goods! For the sale of consumable virtual goods (goods consumed by player game actions), revenue is recognized as the goods are consumed. On the other hand, revenue from the sale of durable virtual goods (goods accessible to a player over an extended period of time) is recognized ratably over the estimated average playing period of paying players for the applicable game.  Confused yet?  Basically, we have to rely on Zynga to provide us with a best estimate of the lives of both consumable and virtual goods to book revenue. As we indicated in “Zynga’s First 10K: Zestful Zephyrs,” by merely changing the game’s rules, the Company can change what it books as revenue! This is all too arbitrary. Are we really surprised?

So, when it comes to recording revenue, it appears that booking advertising income is relatively easy, compared to the management estimates needed for multiple deliverables (Linked In) and virtual good sales (Zynga), or deciding who the “primary obligor” is (Demand Media and Groupon).  We would not be surprised if some internet companies don’t intentionally complicate their product offerings to make revenue recognition a function of management guesstimates!

 

Cost Capitalization

Given that several of these companies are struggling to achieve or maintain profitability, it is not surprising that they would try to record as an asset what really is an expense.  And sure enough, we find several instances of this.  For example, Demand Media capitalizes many different types of costs including content costs, registration and acquisition costs for undeveloped websites and internally developed software, as well as intangible assets acquired in acquisitions.  How significant is this?  Over 72 percent of the Company’s $590.1 million in total assets are intangible in nature!  Now that takes cost capitalization to a new height…we’d probably try that too if we were losing as much money every year as they are (2011’s net loss was $18.5 million).

Linked In also plays this “game,” but with a new twist.  The Company does do something quite interesting…it defers expensing $13.6 million in commissions already paid on non-cancelable subscription contracts, presumably to match the commission costs with the related revenue streams.  Why stop there?  Couldn’t you make the same argument for a whole host of other expenses as well?  Maybe they did, but Deloitte didn’t buy it.

Groupon and Zynga also have played a slightly different version of the cost capitalization game, by recording tax assets that presumably will lower future tax liabilities.  In recording these tax assets, the companies reduce income tax expense in the income statement, thus improving the bottom line.  The only problem is that a company must have future taxable income in order to use these alleged tax assets!  Well, if the companies did this to mitigate their operating losses, the game has ended for Zynga, and soon will end for Groupon.

In 2011 Zynga recorded a $113.4 million allowance against its deferred tax assets, almost fully reserving these assets, and effectively wiping them off the books.  This suggests that the Company may have had a reality check as to its future prospects, given that it no longer projects a future that includes profitability, more specifically taxable income.

As for Groupon, we highlighted this same tax issue earlier in Groupon’s First 10K: Looking Under the Hood.  In 2011, the Company increased its valuation reserve for deferred tax assets by $72.3 million reducing reported deferred tax assets to $65.3 million.  Although Groupon gave no reason for the increased reserve, it likely was forced to record it for the same reason as Zynga, i.e., little likelihood of generating taxable income in the foreseeable future against which deferred tax assets could be used.  So, who would have thought…the income tax note might actually shed some light on what a company really thinks its profit forecast is (as opposed to the press release)!

 

Other Accounting Issues

Our internet company reviews also turned up a couple of interesting points, which give us insight into managements’ attitude toward financial reporting transparency…and believe it or not, Groupon is NOT involved!

The first involves cash, naturally, and how Demand Media “defines” cash.  You may recall that we first reported on the increasing trend of companies to manipulate reported cash balances in “What’s Up With Cash Balances?”  And, yes, Demand Media is overstating its balance sheet cash by including accounts receivable as cash even though it has yet to receive the monies.  Here is what the Company’s accounting policy note says:

The Company considers funds transferred from its credit card service providers but not yet deposited into its bank accounts at the balance sheet dates, as funds in transit and these amounts are recorded as unrestricted cash, since the amounts are generally settled the day after the outstanding date. (emphasis added)

What’s the rush?  Why the need to manipulate the balance sheet this way?  We don’t know exactly how much in receivables is included in cash, so we can’t assess the impact on the balance sheet or the statement of cash flows.  Nevertheless, this is troubling to say the least.  This is exactly the same scam that for which Orbitz was busted and forced to issue a restatement 8K!  SEC…heads up there!

And then there is the “stealth” restatement made by Linked In.  Accounting errors are supposed to result in a restatement of financial statements and disclosed by public companies in an 8K filing. However, “stealth” restatements occur when a company “hides” restated financial figures in a current financial report, say a 10Q or 10K, instead of filing an 8K announcement.

So what was Linked In’s accounting error?   In 2010 the Company erroneously reported trade payable obligations totaling $10.8 million in other accrued expenses within accrued liabilities instead of accounts payable. So, who cares; after all there is no P&L effect, right? WRONG!  We care because this is just another example of how cavalier internet company managers are with the rules. It won’t surprise you to learn that the Company called this error correction a reclassification (note 2 of 2011 10K financial statements).  Sounds so much better doesn’t it?  We are left wondering why Linked In even bothered at all to report this.

And what would a Grumpy Old Accountant blog piece be without us ranting again about non-GAAP pro-forma reporting?  Guess what?  All five of the internet companies in our sample report such metrics, and except for Facebook (which reports free cash flow only and whom we gave an A to earlier), all provide convoluted performance measures that paint a better operating picture than reported under GAAP. Here’s what we mean:

-      Demand Media uses something called “Adjusted OIBDA” which adjusts GAAP operating income by excluding depreciation, amortization, stock-based compensation, as well as the financial impact of acquisition and realignment costs, and any gains or losses on certain asset sales or dispositions. Oh, by the way, look at what’s in acquisition and realignment costs…non-cash GAAP purchase accounting adjustments for certain deferred revenue and costs; legal, accounting and other professional fees directly attributable to acquisition activity; and integration and employee severance payments attributable to acquisition or corporate realignment activities. The Company also has a non-GAAP Revenue metric, ex-TAC, which measures consolidated revenues net of traffic acquisition costs.

-      Groupon uses both free cash flow and the now infamous consolidated segment operating (loss) income, or CSOI, as key non-GAAP financial measures. Free cash flow is defined as we would expect (operating cash flows less purchases of property and equipment), but CSOI excludes acquisition-related costs and stock-based compensation expense from the consolidated operating (loss) income of the Company’s two segments.

-      Linked In reports “Adjusted EBITDA” by adding the provision for income taxes, other (income) expense, net; depreciation and amortization, and stock-based compensation to net income.  By the way, look at what’s included in other income (expense): interest income, foreign exchange gains and losses, investment gains, and other non-operating expenses.

 

-      Zynga uses two pro-forma metrics: bookings and adjusted EBITDA.  Bookings is a non-GAAP revenue measure equal to revenue recognized in the period plus the change in deferred revenue during the period. The Company also uses “adjusted EBITDA” which it defines as net income (loss) plus change in deferred revenue, stock based compensation, depreciation and amortization, other expenses, and less gains (losses) from legal settlements, interest income, and provision for taxes.

Now for the big surprise, NOT!  All four companies have a history of operating losses under GAAP during the past four years. During the same period, Facebook reported operating profits.  So, you tell me why the big push to report these metrics.  Regardless of what management tells you about doing so to be “transparent,” it is exactly the opposite…they do it to obfuscate!

So, there you have it…lots of revenue and asset valuation assumptions in these internet companies.  Do you really believe that these companies with their limited operating histories, and often inexperienced management, are able to make the kind of assumptions and judgments that drive their accounting?  Scary thought, huh?

What are we left to conclude?  Internet company bean counters are not accountants!  After all, it’s tough to count those virtual beans.  If they are not accountants, then what are they?  Artists, of course, and generally bad ones…the paint by “numbers” type.

 

This essay reflects the opinion of the authors and not necessarily the opinions of The Pennsylvania State University, The American College, or Villanova University.

 

 

8 Responses to “THE “BEAUTY” OF INTERNET COMPANY ACCOUNTING”

  1. Lew Domke says:

    Really enjoy your blog and appreciate your viewpoints. As a Villanova Alumni (’99) and CPA who works at an Internet Company – can you provide some balanced insight on ways to improve the financial disclosure for a start up Company that has spent significantly to capture the “leader” space in an industry with huge potential growth? Our investors and analysts ask for the metrics described above, so ignoring that and leading with the points that you make in your blog would seem awfully defeating and doesn’t reflect the positive outlook we have for our Company and the industry.

    • edketz says:

      Tony’s Reply: Thank you for your interest in our ravings! Unfortunately, we see no need in providing a “balanced” perspective, since the markets are clearly “unbalanced” when it comes to financial reporting (i.e., the pro-forma reporting issue). Our intent is to restore equilibrium by reminding whoever will read and listen that financial statements (and securities registration statements) are NOT, and were never intended to be marketing documents.

      We can appreciate the pressure that young startups are under to peddle their stock, but financial disclosures in registration statements and public filings are supposed to give a realistic picture (not make believe) of a company’s financial condition and operating performance. When companies consistently incur GAAP losses and report negative GAAP operating cash flows, yet somehow magically through pro-forma financial metrics claim wonderful performance, something is amiss.

      As for the argument that analysts and investors need such metrics, you are correct…the sell side analysts who use anything to make a sale, and the pitiable investors who listen to them. Quality analysts don’t need companies to add back expenses to convert operating losses into profits…they can do it themselves if so inclined!

      We would hope that startup companies would spend more time on their business process and internal controls, and less on trying to “game” the financial reporting.

  2. Anuj Vij says:

    Your blog is very enjoyable. As a professional accountant and a MBA, i am often amazed at the lack if basic accounting knowledge displayed by a number of finance professionals these days. Basic accounting principles are often ignored and routinely flouted. A number of finance managers and CFO’s do not have solid accoutning fundamentals which leads to numerous disasters down the road. The situation is more prominent in technology and resource companies where the lack of clear cut standards allows managers to comeup with their own version to meet their short term interests.

  3. Jeffrey Donaldson says:

    I appreciate the blog. I’m a lowly student, but it appears to be that some of the negativity may be overblow. Perhaps you could explain to me what is wrong with Demand Media reporting “gross” when it is the primary obligor and “net” when it is not? That seems to be what EITF 99-19 wants them to do.

    And what is the problem with Zynga treating their virtual goods as durable if backed up by game-play data? Is the concern that they will use these “managerial decisions” to adjust time periods in an effort to “smooth” revenues?

    • edketz says:

      There is nothing wrong with how Demand Media is accounting for its revenues. We were just pointing it out as an example where it was the primary obligor in some cases but not in others.
      Zynga, on the other hand, recognizes revenue according to customer behavior. Our problem is the fickleness of humans; these consumers could change their buying habits in an instant and this revenue recognition method is no good. It would be more useful if the firm just recognized revenue as the customers bought the new features of the game.

      • Christopher Ferro says:

        I agree about Zynga – the fact is that the revenue was earned and should be recognized when the ‘goods’ were purchased. The way Zynga is trying to do it is like Frito-Lay recognizing sales of chips when they were eaten, not when they were purchased.

  4. Ah, the “beauty” indeed. Thank you for getting me to think differently about revenue recognition. I think this was a healthy wake up call for the people that read it.

    Ok, time for my substance. I’d like to just comment about Zynga and they’re wacky revenue recognition. I don’t really believe that we can justify accrued revenue based on an estimate that can ebb and flow according to usage rates. I believe this specifically because gamers can get turned off in a second, or the reverse can happen. A third party promotion that Zynga doesn’t realize will help them, suddenly can boost sales.

    Additionally, the internet has a viral effect which is extremely unpredictable and difficult to estimate.

  5. Just found your blog today. Am really enjoying it since I’ve learned a few new things. I even Linked to it from my Twitter and LinkedIn accounts. My background is definitely on the micro/small business level but I’m (finally!) finishing my degree in accounting and have been exploring public/corporate accounting and macro-level accounting issues.

    I really hope you’re giving internet companies too much credit here:

    “We would not be surprised if some internet companies don’t intentionally complicate their product offerings to make revenue recognition a function of management guesstimates!”

    Wouldn’t doing such a thing would harm sales, since the process would be more complicated for the customer? But perhaps you meant it tongue-in-cheek.

    Regardless, keep up the good work!

Leave a Reply

*First and Last Name Required

*Email Address Required

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>