A great investment can present itself with the combination of two things: one, a company operating at a competitive advantage relative to its peers and two, in an industry currently facing tailwinds. In these situations, the company will benefit from both attractive fundamentals as well as favorable perceptions/multiple expansion through its industry. That being said, I believe I have come across a company that exhibits both of these characteristics.
InterDigital Inc. (NASDAQ: IDCC) is a developer and licensor of wireless digital technology, primarily related to 3G, 4G, LTE and, most recently, innovative 5G technology. It’s also mining at the forefront of “Internet of Things” (IoT) technology, which is essentially technology that facilitates the connection of objects in our everyday lives (i.e. self-driving cars, house artificial intelligence systems, etc.) as well as sensor technology with the acquisition of Hillcrest Labs. IDCC boasts a 20,000-patent portfolio, which it licenses to large technology companies including Apple (NASDAQ:AAPL), Samsung (OTC:SSNLF) and Huawei, a large telecommunications company from China.
There are a few reasons that I advocate an investment in IDCC, and I will list them from order of relevance:
IDCC has an attractive patent portfolio that it is poised to monetize. Looking at recent market transactions of similar patents as well as transactions directly related to IDCC, we see that its patent portfolio is extremely valuable. The book value of this asset is ~$300mm but I believe the true value is closer to ~$2bn based on extremely conservative estimates.
You can buy the whole company for its reproduction value, making for a great asset-based estimate.
IDCC generates its revenue from two factors: recurring revenues from ongoing license agreements and “past patent” revenues, which licensees pay upfront after an agreement is reached, essentially compensating for using the technology before a license agreement was signed. If you completely disregard past patent royalties, IDCC still consistently generated a ROE>10% and a ROIC>15%. In addition, mitigating the impact of these “one-time” revenues still yields an upside of 20%.
All these numbers were generated using an extremely conservative analysis, using assumptions like “half of the patents are worthless.” Using more realistic assumptions further bolsters the ROI.
Operating income has grown 20% and net income has grown 30% CAGR for the past 5 years. Recurring revenue growth has been 7% for the same time period. If we consider growth in this analysis, the stock can yield an upside of 92%.
The company is pioneering in emerging markets of utmost relevance to today’s society. While developments of 5G technology is still in its infancy, it provides enormous potential for growth. In addition, with the purchase of Hillcrest Labs, IDCC is pioneering sensor technology that goes hand-in-hand with its IoT technology. If these technologies turn out to be big, the stock can enjoy much appreciation. The best part is that this thesis doesn’t rely on this trend - although it’s fairly predictable that the world is progressing to a more IoT world - and it’s just icing on the cake.
The Value of the Patents
On its books, IDCC has about ~$300mm in patents. I think this value is grossly understated. Because the asset-based valuation isn’t the crux of this thesis, we are going to do more of a back-of-the-envelope analysis of the patent portfolio, which will primarily be used to justify our earnings arguments.
There are a few indications that the book value of the patents is lower than their true value. First, in 2012, InterDigital sold a mere 1,700 patents to Intel (NASDAQ:INTC) for $375mm. That comes out to a valuation of $226,470 per patent. Secondly, InterDigital also bought a stake in Hillcrest Labs, which had a portfolio of 235 patents/pending patent applications, for $48mm. That’s $204,255 per patent. However, according to its most recent 10-K, the value of the patents in this purchase were $36.2mm, so if we use this number instead, we get $154,042 per patent.
Now, these numbers vary dramatically, but they have one thing in common: they are all significantly greater than the stated book value. Before we move on, we should note one thing: not all patents are created equally. The patents that Intel purchased aren’t the same as the ~19,300 left in the portfolio and the patents that were purchased from Hillcrest Labs aren’t the same as the patents that Intel purchased.
If we go into specifics of each deal, Intel bought patents that were primarily related to 4G and LTE technology, while the Hillcrest Labs purchase involved patents related to sensor technology and this probably explains why there are stark differences in the cost per patent of each of the situations. Nonetheless, let us move on.
If we use the Intel purchase as a basis for our valuation, using $226 thousand per patent, the value of the total patent portfolio should be $4.5bn. If we use the Hillcrest Labs as our starting point, the value is $3.1bn. But not all patents are created equal. It is not safe to assume that all 20,000 of IDCC’s patents are as coveted as the 1,700 that Intel purchased.
In addition, some of the patents may not be able to be monetized or may cover a general process/product, making them un-defendable. Already, there have been some cases in which many of IDCC’s patents have been deemed un-defendable by the court. To account for this, let us assume half of the patents are worthless. This gives us a value of between ~$1.5-2.25bn.
At this point, it is important to note that the vast majority of IDCC’s patent portfolio is made up of wireless technology; in other words, patents more closely related to those in the Intel sale. Therefore, I’ll stick with $2bn, keeping in mind, however, that the potential value can be much larger.
A huge contradiction should hit you just about now, and this is why this analysis won’t rely on the asset-based valuation. If all patents are not created equally, then the whole method of applying a single cost per patent and multiplying that for all the patents is inherently flawed. Therefore, alternatively, we can take an R&D approach.
In this approach, we take the R&D expenses for the past 5 years, as well as the number of new patents signed each year, and use that to find the cost per patent. This method estimates the total R&D expense used to create IDCC’s full 20,000 patent portfolio.
Using this approach, the value of the patents is closer to $1bn. Again, there is a huge discrepancy, but one thing stands out: despite where the actual value falls, it is significantly greater than ~$300mm. Finally, this analysis of the patent portfolio was very crude and quick, but it gave a good idea as to how much the book value understates the InterDigital patent portfolio.
This exercise is futile, however, if the company cannot monetize this portfolio. In the past years, however, we have seen IDCC strike license agreements with Apple and Huawei, extend its agreement with Samsung and strike a deal with Intel. All in all, new patent agreements and patent sales have produced an extra ~$1bn of revenue since 2012. In addition, IDCC recently transferred 500 patents to its variable interest entity, Signal Trust Fund, whose goal is to monetize said assets through asset sales. Looking at the big picture, it seems like monetizing this patent portfolio is at the top of management’s mind.
The point of this exercise is to prove that the book value of IDCC is much larger than what is stated, and this sets us up to calculate the reproduction value in the next section. And while this thesis doesn’t rely on this, there is a chance that this patent portfolio appreciates in value if we are at the horizon of a 5G/IoT revolution, but I’ll discuss this more in the last section.
Great Company, Fairly Priced
Buffet once said that it is acceptable to buy a great company at a fair price, and if we solely consider this asset-based valuation, that is exactly what InterDigital is. To find the reproduction value of this firm, let’s use $1.7bn for the patents, which is the midpoint of the above analysis. Without doing anymore adjustments, other assets give us another $1.53bn for a total asset value of $3.2bn. Total liabilities are about $1bn, so the reproduction value of this firm is $2.2bn.
Could we buy this whole business at or less than its reproduction value? It has over $400mm in cash and $500mm short-term investments, which can easily be converted into cash if need be, and ~$200mm in debt. Since the market cap is $2.4bn, the enterprise value of this firm is also ~$2.2bn. What does this tell us? We can buy the whole company for its reproduction value; in other words, it is fairly valued based on a conservative asset-based valuation.
It is worth reiterating the major assumption that led us to this conclusion: we said that half of IDCC’s whole patent portfolio is utterly useless. Conservatively, the company is fairly valued. If we were wrong and only ¼ of the patents are useless, we could be looking at a huge upside from an asset perspective.
Before going into the earnings, it is important to understand how the licensing agreements work. When IDCC signs a licensing agreement with another company, the other company agrees to either a royalty-based or fixed-fee agreement depending on how many units with the patented technology are sold. This is recurring revenue. Aside from that, if the licensee had been selling the technology before the license agreement, the licensee agrees to pay an upfront “past patent royalty,” which covers all past sales with that technology. For example, when a license agreement was signed with Apple in Q4 2016, IDCC realized ~$300mm in past patent royalties.
Because of their opportunistic nature, past patent royalties cannot be counted on year in year out. They only arise after a new license agreement is signed into action. Therefore, as a starting point in this analysis, we take away all the effects of past patent royalties in the income statement (note that under past patent royalties in 2012, we include $375mm from the Intel deal):
As you see, past patent royalties make up a large portion of revenues. Equally important, however is the returns generated even after making these changes. The average ROIC over these past five years, after making these changes, is 17%.
Now, while we cannot count on huge past patent royalties every year, we can expect some sort of compensation. Over the past five years, IDCC has generated $1bn in patent sales/past patent royalties, or about $200mm a year. To be conservative, I will use $100mm, so we aren’t too dependent on new license agreements. No major license agreements were signed between 2013-2015 and past patent royalties were dabbling around the $100mm figure.
If we make the $100mm adjustment, we get a normalized EBIT of $228mm. The average EV/EBIT for companies in this sector is 12, so applying this multiple yields an EV of $2.7bn for an upside of 23%.
Again, the assumptions used in this exercise were extremely conservative. We downplayed the effect of non-recurring revenue by 50% and we didn’t consider growth in recurring revenue. This analysis also justifies the value of the patents we obtained in the previous exercise; the ~$500mm difference in the earnings power and asset valuation comes from the competitive advantage that stems from the proprietary technology and patents InterDigital owns. After making the $100mm adjustment, the resulting ROIC of 65% in 2016 (39% 5-year average) and 18% 5-year average ROE justifies this claim.
Value of Growth
I do not want to spend too much time on this section because in this opportunity, I see growth more of a qualitative margin of safety. In addition, small mistakes in assumptions can distort the value dramatically, so growth value should be taken with a grain of salt. Nonetheless, many qualitative favorable industry tailwinds give IDCC a long runway for growth. Most notably, the company signed an agreement with Apple and considering the proliferation of the Apple Watch and Cloud connectivity, wireless communication technology is in high demand.
IDCC is a pioneer in IoT and 5G technology and it has been recognized in many industry conferences for development in these emerging markets, most notably, the Mobile World Congress (where it predicted that 5G technology will roll-out as early as 2018. With the acquisition of Hillcrest Labs, IDCC is also poised to make strides in sensor technology, with applications most obvious in self-driving cars, which are slowly becoming more prevalent.
To the quantitative factors, recurring revenue has grown at a 9% CAGR for the past 5 years. Operating income has grown at a 20% CAGR and net income has grown at a 30% CAGR over the same time period. If we assume a 5% terminal growth rate and an 8% discount rate, the present value of the firm should be $4.6bn, giving us a 92% upside.
There are a few catalysts that may occur over the next few years. One, the company can capitalize on patent sales. In 2012, IDCC sold 1,700 patents to Intel. It signed multiple licensing agreements since then, further monetizing this value. Finally, the creation of the Signal Trust Fund signifies management’s efforts to monetize these assets. Two, if new 5G technology/IoT tech/sensor technologies take off, IDCC will be there to reap the awards.
We are seeing a proliferation of self-driving cars, and many tech companies are using IoT tech to improve cloud-based, multiple device software, so a future of IoT is not a far-off assumption. Three, market coverage, industry presentations (such as the presentation at the Mobile World Congress) will put the spotlight on IDCC, allowing for a market reevaluation.
Disclosure: I am/we are long IDCC. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.