Can patent indicators predict company valuation? A review of recent findings
One of the newest and most promising areas in econometrics research focuses on the valuation of companies based on their patent stock. This work, initiated by Hall’s observations regarding the correlation between patent citations and a company stock value, has blossomed in the past 5 years. Here is a review of recent findings.
Cockburn and Wagner studied the link between the survival rate of companies first listed on the Nasdaq during the internet bubble at the end of the 90s and the valuation of their patent stock. In order to measure the patent quality, the authors counted the number of patents awarded by USPTO, EPO and the Japanese Patent Office, as well as the family size (number of international extensions) and the number of citations by patent and by claim for the awarded patents. In practice, less than half of these companies had patents, and 2/3 were delisted from Nasdaq before 2005; the authors show that the companies listed on the Nasdaq that had filed at least one technological patent (with the exclusion of the business model patents) had a survival rate 34% higher than the others. On the other hand, the quality of those patents, quantitatively measured using the number of forward citations in particular, only marginally contributes to the company survival rate; it is more often associated with an exit from Nasdaq via a merger or acquisition, underlying the importance of formalizing intangible assets through quality patents in anticipation of M&A/exit operations.
Helmers and Rogers measured the survival rate over 5 years of the 162,000 limited liability companies founded in Great Britain in 2001 as a function of their innovation rate, measured through their intellectual property filings activity: patents, which characterize invention, i.e. the beginning of innovation formalization, but also the trademarks, which are more specifically linked to commercialization, hence to the passage from innovation to exploitation. Of the 162,000 firms founded in 2001, less than 2% protected their intellectual property in either form, and 70% of the total survived to 2006. As in the Nasdaq case, and even in this large number of companies of all shapes, sizes and sectors, the authors show that the newly founded British companies having filed for patents have a higher survival rate than the others (16% higher), and the correlation is even higher in the case of a patent filed at the European rather than just United Kingdom level. It is no surprise that the impact of the patents on the survival rate is more significant in the technological sectors, in manufacturing, information technology, R&D and services, while is it marginal in the financial, real estate and retail industries. On the other hand, the trademarks, by nature more general, correlate with a higher survival rate in most sectors. It is to be noted that this study also shows a positive correlation between geographical proximity with a university and company survival rate.
Hypothesizing that a company’s value is increased by its ability to differentiate itself, especially on the technological level, Czarnitzki and Kraft set out to measure in German companies the correlation between patent stocks and profitability, measured as the ratio between their profits and their expenses. They showed that in their sample, a patent stock increases the company’s profit margin by 0.7% on an average 4% basis compared to the patent-less companies.
In the same way, and more recently, Neuhäusler, Frietsch, Schubert and Blind directly measured the correlation between the patent indicators, used as approximation of R&D performance, and the financial performance of a company. To estimate the latter, the authors used not only stock information, but also a return on investment measure (ROI – profit before interest and taxes, divided by the sum of assets) and Tobin’s q factor, indicating respectively the immediate and anticipated profitability of a company, of particular interest when looking to invest for value and dividends rather than speculation (stock market value). The authors used in their analysis a sample of 479 companies actively investing in R&D, listed between 1990 and 1997 on the English index DTI-Scoreboard. In this sample, the gross quantitative measure of the patent number is too noisy to be statistically significant regarding the gross company performance in the market, but nevertheless gives a positive correlation with the return on investment. Furthermore, a refined analysis using complementary qualitative indicators such as the international family size, the citations number and the proportion of oppositions, the value of a company on the market seems directly correlated to the quality of its patents.
Those studies, fairly recent and little known outside of academia yet, suggest that it should be possible to predict a technological company’s financial performance using refined patent indicators. As outlined by Neuhäusler, it is particularly true in the chemical, pharmaceutical, information technologies and telecommunications sectors, and to some extent the electronics and mechanics sectors. Such measures are likely to be developed in the coming years, in particular to support analyses aiming at mid- to long-term investments geared toward profitability (dividend distribution) rather than speculation (stock price increase) of a company.