IP and the Art of "Forensic" Accounting

This is another in my series designed to stimulate thought and dialogue around the conundrum that is IP valuation.  I previously talked about the balance sheet ramifications of Google's transaction to acquire Nokia and will return to another large, albeit somewhat historical transaction that further elucidates this point.  From the Microsoft 2014 Balance Sheet, quarterly data:

I have highlighted the line for “Intangibles, Net” and you will probably immediately notice that between 2014-3-31 and 2014-6-30 there is a really big jump from just over $2.9 billion to almost $7 billion.  As I have discussed before (What every CEO…), this is yet another example of one of the very most interesting aspects of intangibles and the associated accounting rules.
As you probably already know, this jump is due to acquisitions, most notably Nokia.  The intangible assets associated with the transaction were listed as just over $4.5 billion in their 2014 Annual Report.  Further, they allocated about $2.5 billion to “technology-based” intangible assets (e.g., patents).  By the way, this example is a bit out-of-date due to more recent events surrounding the acquisition and Microsoft’s write-off of $7.6 billion related to its acquisition; however, the accounting issue is still illustrative.
Intangible assets generated internally are NOT reflected on the balance sheet.  So they have what value?  ZERO.  It is only intangible assets that are acquired that fall to the balance sheet.  Some clever enterprises and CFO’s have their own models and balance sheets (calculating value through various means) – I suspect that in the future this will become more common and we will see improvements in internal intangible valuation, as well as better communication to the market.
This is further complicated by the fact that after acquisition, the intangible assets get absorbed, combined, and enhanced by the new enterprise over time, and the value of those assets on the balance sheet surely goes up then, right?  No, actually the very opposite, as explained here by CFO.com in the article “The Case of the Vanishing Intangible Assets.”
“The first standard calls for an acquired asset such as a brand to be identified and valued immediately post-transaction. The acquiring firm then takes control of the asset and uses its marketing skills to improve the brand. It generates new value by keeping existing customers, gaining new ones and introducing improvements and innovations to achieve continuing growth. As the firm brings this about, the nature of the asset passes from the first accounting standard to the second. Instead of being acquired, the asset increasingly is internally generated, and that class of intangible asset is explicitly excluded from being recognized as an intangible asset.”
IP Valuation – Market Analysis
Interestingly, there have been numerous broad attempts to analyze IP, patent portfolios, and assess their relative strengths in order to glean important insights into market opportunities (e.g., buying and selling stocks). Very often the approach goes something like this:
1) Develop sophisticated data analysis techniques and algorithms to understand and “rate” patent portfolios in “strength” – by proxy, this attempts to also understand how well an enterprise is doing in generating and protecting its IP through its operations.
2) Analyze relative performance and place bets (i.e., pick the ones you think have the strongest portfolios and best innovation/IP management practices)
Just to be clear, what I am talking about is distinct from the type of analysis associated with single patent identification and valuation (value metrics, network analysis, etc.).  They aren’t completely unrelated obviously, but the ability of an analyst to intensively evaluate a single patent is quite different from understand 100’s, 1,000’s, or 10,000’s of patents and their value proposition.  I will return to the issue of single patents later as part of the series.
The basic premise is that by doing this, you are unlocking market inefficiencies.  That is, the market simply doesn’t know what to do with IP – and frankly, this is largely true.  It is a very valid approach when powered by good data, great analytics and algorithms, and most importantly solid execution.  However, it is missing a key component (at least long-term) – and that really defines the distinction between the two following questions:
“Which companies have the strongest IP (you could apply this more generally to intangible assets overall) portfolios and demonstrate market-leading behaviors in innovation and IP development?”
Same question above with the added “and the market doesn’t already factor that into its valuation of the enterprise(s)…”
One may argue that for the moment the second question is unnecessary because this is an area that the market does a poor job with across the board.  I’d tend to agree somewhat; but will play devil’s advocate and ask if anyone would be surprised if Apple was considered a market leader in innovation (assuming the data told us that) – which suggests there are companies for which their innovation (and perhaps IP) prowess are factored well into their valuation.  This is NOT the same thing as saying their IP prowess is well understood – I would still suggest that it isn’t… but for the purposes of a superficial investment analysis it doesn’t matter in this case, does it?
So where does that leave us?  We are begining to understand the scope of the challenge but need to be a little more circumspect about indicators of performance as direct rationale for buy/sell decisions in the marketplace.  Its like saying that a company’s revenue (substitute any traditional financial metric) is trending up and somehow having enough cognitive dissonance to believe the market hasn’t somehow already consumed that nugget and priced it into the stock.
I would not for one moment suggest the market is perfectly “efficient.”  It isn’t – there are countless examples of this: “black swans” and others.  The massive mortgage derivatives debacle in the last decade and subsequent recession should have taught us this certainly.  But the market does do a pretty good job most of the time, so one should interpret signals carefully and understand them in a larger context.  Of course, one of the key assumptions of efficient market theory is that stock prices “should reflect all available information.”  I am sure you see the problem with IP here (more on that later)…
However, my main point here wasn’t necessarily to tantalize stock-pickers and investors (although this is a perfectly valid objective – for another time), but to more broadly address the issue of market ignorance and how it is relevant to many parties.  Ultimately, the objective is more intelligent generation and management of IP and maximizing the return on that investment, not precise algorithmic or exhaustive manual valuation processes.
As the title of my article suggests, proper “intangible” valuation is a bit of a “forensic” exercise.  There is no easy way to objectively and precisely determine what the value of intangible assets are for an enterprise, unless the assets are 100% acquired externally (even then, there are challenges, as discussed above).  
Yes, there are numerous techniques designed to do just this, but I would argue that NONE of them are adequate and most suffer from the weakness of needing data/time of the intangible ALREADY active in the market (e.g., value the intangible based on its licensing revenue stream).  
I don’t think it would come as a shock to anyone that the patents backing the superstar drug patents already generating hundreds of millions in annual revenues are “valuable” and estimate a reasonable back-of-envelope valuation using the income flows.
Or alternatively, some valuation approaches are simply an accounting (usually not very accurately) of the R&D expense to develop the intangible (does anyone assume that “I” by definition equals “R”?).  Plenty of researchers and accountants have studied financial statements and market capitalization and made estimates based on reported financials.  For example, by looking at market cap and financials delta (complicated by the fact that market cap builds in investor expectations as well).  
While the actual conceptual analysis is surely flawed (you can’t simply take market cap and subtract out assets to derive “intangibles”), the conceptual framework is highly useful in meta-analysis.
Baruch Lev concisely summarized the intangibles problem in his wonderful paper “Intangible Assets: Concepts and Measurements.”  There are many valuable insights in this paper, but one in particular that resonates was his statement about the measurement of intangibles and why we should care about it.  He cites the old adage that “what’s not measured is not managed.”
The challenge is daunting and taking on the task in light of an entire product and patent portfolio is really difficult.  But building in processes and analytical steps into the IP and product lifecycles can lead to future mastery just by focusing on best practices for emergent products and IP.  Over time, an enterprise will have a much higher level of understanding of key data for decision-making and competitive positioning.
I’ll address this more in a subsequent paper, but by following some modest and integrated steps, an enterprise can significantly improve their understanding and stewardship of their intangible assets.  Some of the shroud of mystery surrounding valuation of intangibles should begin to clear.  If you are already tackling this problem successfully, congratulations to you – you are ahead of the game.  If you aren’t, the next decade will reveal it and you won’t like the consequences.
I won’t get deep into the financial aspects of assigning $ to various assets during the process, but I hope this series at least stimulates some thoughtful dialogue about intangibles and their management.  I will address aspects of portfolio analysis and valuation and plan to include contributions from licensing managers and transaction professionals as well.