Intangible assets have become key factors in economics and have received much attention of late. Recent studies focus on the single aspect of intangible-related information, whereas aggregate proxies of intangible assets are scarce. In this post, we introduce our recent working paper in which we construct a comprehensive intangible assets-related measure, I-SCORE, and study the predictive power of I-SCORE on cross-sectional stock returns in the U.S. We show that firms with a higher I-SCORE generate higher stock returns, higher profitability, and higher cash flow. Furthermore, we find that the I-SCORE premium is stronger among firms with higher valuation uncertainty, higher limits to arbitrage, and higher cash flow volatility.
Why are intangible assets so important?
Intangible assets are assets that lack physical substance. Innovation, brand value, human capital, and financing capacity are all intangible assets. Technology firms like Apple, Microsoft, and Alphabet tend to have a higher proportion of intangible assets. In U.S., the growth rate of intangible asset investment has overtaken investments in tangible assets, and intangible assets generate greater returns. In the 1970s, the ten biggest firms in the U.S. were primarily manufacturing and retail companies. Currently, the largest and most profitable U.S. companies are mainly high-tech and service-oriented companies. This transformation implies that intangible assets increase firm performance, earnings, and stock market value. In our paper, we use a theoretical model and empirical studies to illustrate that intangible assets positively affect firms’ future earnings and expected stock returns.
Why do we aggregate information?
Information aggregation method is the method that collects information from several sources. In empirical asset pricing, information aggregation is to extract the informative signals about stock returns from multiple different accounting variables. We use the information aggregation method for two reasons. First, intangible assets contain multiple pieces of information and are hard to value. Intangible assets include innovation, human capital, brand value, and financing capacity. Intangible assets are not reflected on firm balance sheets. The recent asset pricing literature usually focuses on the signals of intangible assets as measurements, for example, R&D, innovation capacity, or hiring rate, and carries only one aspect of intangible-related information. In our paper, we consider the different categories of intangible-related characteristics, aggregate the effective information from them, and construct a comprehensive intangible-related predictor. Unlike the well documented measurements of tangible assets, the measures of intangible assets, especially the aggregate proxies, are still ambiguous.
The second reason is that researchers have discovered hundreds of anomalies in empirical asset pricing. The existing factors composite a “factor zoo.” Some factors may contain the same information and researchers may have been rediscovering the same anomalies. Extracting the relevant information from a large number of characteristics is critical in both academic research and industry practice. Thus, we employ a partial least squares approach to aggregate information from several intangible assets-related firm characteristics and construct our composite intangible-related characteristic I-SCORE.
Why do we use partial least squares approach?
Applying information aggregation methods reduces dimension and aggregates information from the diversified characteristics. Researchers have been using many information aggregation techniques in asset pricing studies. Compared with several different methods, the partial least squares approach has its strengths. Literature has shown that the partial least squares method can screen out both the common and individual information within the characteristics that are unrelated to future stock returns. The partial least squares approach outperforms several alternative information aggregation methods in dissecting cross-sectional stock returns. For example, the principal component analysis may pick up information that is unrelated to expected stock returns. The Fama-MacBeth regression may suffer from the multilinearity problem, especially when firm-level variables are highly correlated. The forecast combination approach is the simple average of the return predictability of each factor, which may lose the information that is related to expected returns and suffer from the positive and negative counteraction. Overall, the partial least squares-based measurement, I-SCORE, is more informative about future stock returns than the alternative information aggregation approaches-based predictors.
Our findings
Our goal is to measure U.S. corporations’ intangible assets and dissect the cross-sectional stock returns in terms of I-SCORE. We find that I-SCORE has significant positive predictive power on cross-sectional stock returns in the U.S. stock market. Firms with higher I-SCORE generate subsequent higher stock returns. The return predictability of I-SCORE is significantly larger than the individual intangible assets-related characteristics. The long-short strategy provides significant abnormal returns as well. The I-SCORE premium remains significant when adding the R&D-related factors and the innovation-based factors, suggesting that I-SCORE contains the aggregate information which cannot be explained by single signals. I-SCORE also leads to higher future profitability and higher cash flow growth. Investment in total assets and intangible assets increases the future I-SCORE. Furthermore, we investigate the economic explanations of the I-SCORE premium. The results show that the I-SCORE premium is stronger among firms with higher mispricing and higher cash flow risk.
Implications
We find that measuring intangible assets from different dimensions and dissecting its return predictability provides valuable information for firms and investors. The positive relationship between: I-SCORE and future stock returns, I-SCORE and future profitability, and I-SCORE and future cash flow, suggests that firms should care more about intangible assets. Overall, intangible assets are important in firm value creation and investors should pay attention to firms with higher I-SCOREs.
Cunfei Liao is a PhD candidate in finance at College of Finance and Statistics of Hunan University.
Fuwei Jiang is the dean and professor of School of Finance of Central University of Finance and Economics.
Fujing Jin is an assistant professor at School of Economics and Management of Beijing Jiaotong University.
Guohao Tang is an assistant professor at College of Finance and Statistics of Hunan University.
This post is adapted from their paper, “Intangibles to Tangible: In Search of Firm Value Creation”, available on SSRN.
