“The companies we’re short have a glaring, glaring hole in their economics – they are not spending on R&D… If you look at a Samsung or an IBM, they’re spending 6%, on average, a year, on R&D to put out new products or services. You look at a company like Dell or Hewlett-Packard, and its 2-3%; and that difference on a low margin business makes them marginally much more profitable. The difference is Dell and Hewlett are out acquiring their R&D; that’s capitalized – it’s not expensed.” I think it would be interesting to take a deeper dive; let’s look at what Generally Accepted Accounting Principles (GAAP) say about accounting for research and development:
“Statement #2 - This Statement establishes standards of financial accounting and reporting for research and development (R&D) costs. This Statement requires that R&D costs be charged to expense when incurred. It also requires a company to disclose in its financial statements the amount of R&D that it charges to expense.”
In addition, we have a specific statement to address computer software (emphasis added):
“Statement #86 - This Statement specifies the accounting for the costs of computer software to be sold, leased, or otherwise marketed as a separate product or as part of a product or process. It applies to computer software developed internally and to purchased software... This Statement specifies that costs incurred internally in creating a computer software product shall be charged to expense when incurred as research and development until technological feasibility has been established for the product. Technological feasibility is established upon completion of a detail program design or, in its absence, completion of a working model. Thereafter, all software production costs shall be capitalized and subsequently reported at the lower of unamortized cost or net realizable value. Capitalized costs are amortized based on current and future revenue for each product with an annual minimum equal to the straight-line amortization over the remaining estimated economic life of the product
I add that because I’m going to talk about Microsoft (MSFT) at first, and want to make clear that a relatively small percentage of the company’s R&D spend is accounted for in this fashion; amortization of capitalized software was $117 million, $114 million and $97 million for fiscal years 2012, 2011 and 2010, respectively.
Here is Microsoft’s R&D spend by year, and as a percentage of sales:
|Year||Dollar Amount||% of Sales|
As Mr. Chanos noted, some companies purposely look to acquire their R&D rather than develop it internally because of the benefit to GAAP earnings (capitalized over time, versus expensed in the current period). Inverting this, there’s another question worth asking: Which companies are overstating the economic costs associated with research and developing?
For the sake of conservatism (rightfully so), GAAP says that the entire bill for R&D spend must be expensed in the current year. Knowing that Microsoft spent about $820 million in R&D each month during fiscal 2012 (on average), this brings up an interesting question: Does the expense for any one month (for the sake of argument, the final month of the fiscal year) align in any way with the economic value of that spend? Is there any reason to believe that the construct of the calendar (12-month periods) is an accurate reflection of the useful life for Microsoft’s R&D?
Determining whether or not this is the case is easier said than done. However, let’s think about the consequences if it were so; essentially, what would be the end result of capitalizing R&D to account for it’s useful life (as with capital expenditures), rather than expensing the cost in the period incurred?
Let’s assume that we use a two-year useful life on the R&D spend: In fiscal 2010, we would account for half of the cost ($8.7 divided by 2 = $4.35 billion), and include the half that had been set aside in the previous year. Here’s the actual figures for fiscal year 2011 and fiscal year 2012:
|Year||2-Yr Life||As Reported||Boost To NI|
To be clear, this is a timing issue: If we didn’t spend a penny on R&D in fiscal year 2013, conventional accounting (expensed) wouldn’t show any R&D costs; if we had utilized the metrics presented here, the second half of fiscal year 2012’s $9.8 billion ($4.9 billion) would roll off and hit the income statement.
$400 million doesn’t mean much for Microsoft – it would have only increased reported earnings by about $0.05. Where this can become meaningful is for a company of marginal profitability – and when we start extending that useful life well beyond the two years used in this example.
On to Hewlett-Packard (HPQ). here’s the company’s R&D spend by year, and as a percentage of sales:
|Year||Dollar Amount||% of Sales|
The three year average (as a percentage of sales) is less than 2.6% of sales; in the comparable period a decade earlier (2000 TO 2002), R&D spending was equal to 5.7% of sales – or more than 2x the current figure. The critical section in this discussion comes from page 89 in the 10-K:
“HP includes the results of operations of the businesses that it has acquired in HP's consolidated results as of the respective dates of acquisition. HP allocates the fair value of the purchase consideration of its acquisitions to the tangible assets acquired, liabilities assumed and intangible assets acquired, including in-process research and development ("IPR&D"), based on their estimated fair values. The excess of the fair value of purchase consideration over the fair values of these identifiable assets and liabilities is recorded as goodwill. The primary items that generate goodwill include the value of the synergies between the acquired companies and HP and the acquired assembled workforce, neither of which qualifies as an amortizable intangible asset. IPR&D is initially capitalized at fair value as an intangible asset with an indefinite life and assessed for impairment thereafter. When the IPR&D project is complete, it is reclassified as an amortizable purchased intangible asset and is amortized over its estimated useful life. If an IPR&D project is abandoned, HP records a charge for the value of the related intangible asset to HP's Consolidated Statement of Earnings in the period it is abandoned.”
Here is Hewlett-Packard’s “amortization of purchased intangible assets” expense by year, and as a percentage of sales:
|Year||Dollar Amount||% of Sales|
By the way, this doesn’t account for the $160 million that HP has spent (on average) each of the past five years in acquisition related charges; let’s dig deeper into the amortization line item:
“The increase in amortization expense in fiscal 2012 was due primarily to amortization expenses related to the intangible assets purchased as part of the Autonomy acquisition. This increase was partially offset by decreased amortization expenses related to certain intangible assets associated with prior acquisitions reaching the end of their amortization periods…. The increase in amortization expense in fiscal 2011 was due primarily to increased amortization of purchased intangible assets from acquisitions completed during fiscal 2010. This increase was partially offset by decreased amortization expenses related to certain intangible assets associated with prior acquisitions reaching the end of their amortization periods.”
Financial accounting leaves room for flexibility. In many instances – as I believe is the case for R&D expense – it will err towards conservatism (as it should). As always, investors must carefully read through SEC filings to see what lies beneath the headline numbers; in the case of HP, I think Mr. Chanos may have found something that should cause HPQ shareholders to take a long, hard look at the current business practices of the company.
About the author:I'm a value investor, with a focus on patience; I look to buy great companies that are suffering from short term issues, and hope to load up when these opportunities present themselves. As this would suggest, I run a fairly concentrated portfolio by most standards, usually with 8-10 names; from the perspective of a businessman rather than a market participant / stock trader, I believe this is more than sufficient diversification.
I hope to own a collection of great businesses; to ever sell one, I would demand a substantial premium to the average market valuation due to what I believe are the understated benefits to the long term investor of superior fundamentals and time on intrinsic value. I don't have a target when I purchase a stock; my goal is to replicate the underlying returns of the business in question - which if I've done my job properly, should be very attractive over a period of many years.