AdobeStock_235261731.jpeg
VoxEU Column Productivity and Innovation

Firm age, productivity, and intangible capital

How do firms grow as they age after establishment? What drives high growth rates for young firms? Using a large dataset from Japan for the period from 1995 to 2015, this column argues that the accumulation of intangible capital plays a significant role in the growth of physical productivity, which, in turn, accounts for a major part of sales growth as firms age. Of the three types of intangible capital – organisational capital, software, and R&D stocks – organisational capital explains a large part of the sales growth.

How do firms grow as they age after establishment? Start-ups are by nature small in terms of sales and employment, but how do they evolve with age? These questions have attracted both policymakers’ and economists’ attention as aggregate productivity growth and business dynamism have been stagnant for the last two decades. Many studies have shown that the size of manufacturing plants and firms grows in terms of sales and employment as they age, and that younger plants and firms have higher growth rates in the US (e.g. Davis et al. 1996). In our recent study (Hosono et al. 2020), we find that Japanese firms show such an age-size relation as well. Figure 1 depicts the log difference in sales and production factors of Japanese firms in our sample. The growth rates in sales and all production factors are higher for younger firms. Growth rates decline with age and reach zero around 30 to 40 years after establishment. In addition, the growth rate of sales is higher than the growth rates of production factors, suggesting that productivity increases with age to the extent that productivity is measured as the ratio of sales to the weighted average of production factors.

Figure 1 Age and growth rates in sales and inputs

Source: Hosono et al. (2020)

What is less known is the mechanism that drives such age-size and age-productivity relationships. One branch of the literature stresses the selection mechanism through which less productive firms exit and more productive firms survive (e.g. Jovanovic 1982). While empirical studies on the selection mechanism are relatively abundant, Cabral and Mata (2003) showed, using a comprehensive dataset of Portuguese manufacturing firms, that selection accounts for little of the change in the firm size distribution. Another branch of the literature examines the role of organisational capital that plants and firms accumulate as they age (Atkeson and Kehoe 2005, Hsieh and Klenow 2014). Atkeson and Kehoe (2005) build a growth model for the life cycle of plants that incorporates the accumulation of plant-specific knowledge, which they call organisational capital. In their model, firms accumulate organisational capital through the learning process as they age. Using manufacturing plant-level data from Mexico, India, and the US, Hsieh and Klenow (2014) find that plant size in terms of labour and productivity (quantity-based total factor productivity: TFPQ) grows less as plants age in Mexico and India as compared to the US. This finding indicates that firms accumulate less organisational capital in India and Mexico in comparison to those in the US. They further find that the revenue-based total factor productivity (TFPR) rises much more steeply with the TFPQ in India and Mexico than in the US. This finding indicates that distortions in taxes, factor costs, financial frictions, and transportation and trade costs become larger with age in India and Mexico. Both Atkeson and Kehoe (2005) and Hsieh and Klenow (2014) show that plant-specific investment in organisational capital plays a key role in plant growth in terms of size and productivity, although they do not directly measure organisational capital.

Organisational capital is one type of intangible capital. Since the IT revolution in the US during the 1990s, many studies have focused on measuring intangible capital and analysing its role in firm activities (e.g. Bresnahan et al. 2002). Recently, some researchers have attributed the decline in business dynamism to the concentration of intangible capital. Akcigit and Ates (2019a, 2019b) stress the decline in knowledge spillover across firms as a reason for the declining business dynamism in the US. They mention tacit knowledge and proprietary big data as one of the potential sources for the decline in knowledge spillover. Using data on French firms and US publicly listed firms, De Ridder (2019) focuses mainly on software and finds that the increasing use of intangible capital such as software explains the slowdown of productivity growth, the decline in business dynamism, and the rise of market power. Crouzet and Eberly (2019) also show that intangible capital is associated with market power gains, productivity gains, and consequently, market concentration.

Intangible capital and firm growth with age

In our recent study (Hosono et al. 2020), we examine the roles of intangible capital including organisational capital in firms’ growth in sales and productivity over time. We measure firm-level intangible capital to examine whether it plays a significant role in the growth over the firms’ life cycles. 

We first look at how young firms accumulate intangible capital. While many studies have examined the determinants of investment in intangible capital, none have focused on its relationship to firm age. Figure 2 shows the mean log difference of intangible capital by age for our sample of Japanese firms for the period from 1995 to 2015. Firms are divided into five age groups: age 2-9, age 10-29, age 30-49, age 50-69, and age 70 and over. In the figure, the mean of the youngest firms is the highest in every year of the observation period. The second youngest firms have the second highest investment rate and the other three groups are almost equal to each other. While some studies point out that the accumulation of intangible capital is slow and negative in some years in Japan (e.g. Fukao et al. 2009 and Chun et al. 2012), the figure shows that young firms actively undertake investment in intangible capital in all years including during severe recessions. This figure demonstrates the importance of focusing on the role of age in the accumulation of intangible capital.

Figure 2 Investment rate of intangible capital by firm age groups

Source: Hosono et al. (2020)

We construct a model to show the relationship between sales and three parameters: physical productivity (TFPQ), markup, and distortion in the factor prices. While Hsieh and Klenow (2009) show that the log of sales is proportional to the difference between the logs of TFPQ and TFPR under the assumption of a constant markup across firms, we decompose TFPR into the markup and the distortions in factor prices, both of which vary across firms and over time. This decomposition is important given the potential effect of intangible capital on market power.

Figure 3 shows the relationship between the estimated parameters and age. TFPQ increases with age up to about age 30 and has a similar path to sales, suggesting that young firms improve their technology rapidly. Markup also increases with age, but its growth rate is smaller than that of sales and the TFPQ when firms are young. The markup path means that newly established firms are forced to set low prices and suffer from low markups for an extended period. Factor price distortion is close to zero at any age. It decreases slightly after establishment, and then increases steadily. The initial decline in the distortion is consistent with the reputation hypothesis in the credit market that postulates that firms can reduce borrowing costs over time (Sakai et al. 2010), although the magnitude is small. 

Figure 3 Age and firm-level parameters of TFPQ, markup, and distortion  

Source: Hosono et al. (2020)

Next, we analyse the role of intangible capital in the age-size relationships through the three parameters. Using firm-level panel data of intangible capital that consists of organisational capital, software, and R&D stocks, we regress the changes in sales, TFPQ, markup, and distortion on age and the changes in intangible capital. The estimation results show that intangible capital has significant effects on firm growth through the TFPQ. Of the three types of intangible capital, organisational capital accounts for the largest portion of the growth in sales, but software and R&D stocks also have some effects on markup and distortion of the factor prices.

Summary and policy implications

Many studies have shown that younger firms grow faster in terms of size and productivity. While the accumulation of organisational capital is an important candidate for  explaining such age-size and age-productivity relationships, empirical studies are relatively scarce. In our recent paper, we explore the role of intangible capital in the growth of firms over time through the lens of a model with firm-level TFPQ, markup, and factor price distortion. 

Using firm-level panel data from Japan, we found that the accumulation of intangible capital plays a significant role in the growth of TFPQ, which, in turn, accounts for a major part of sales growth as firms age. Among the three types of intangible capital that we examine (organisational capital, software, and R&D stocks), organisational capital accounts for the largest portion of the sales growth.

The results of our study imply that the accumulation of intangible capital by young firms is a key driver of aggregate economic growth. On the other hand, the low markups and high factor prices facing young firms are not significant barriers to firm growth. Policies that help young firms accumulate intangible capital and enhance productivity can be more effective than those that promote market concentration or remove factor market distortions such as financial constraints.

Editor’s note: The main research on which this column is based (Hosono et al. 2020) first appeared as a Discussion Paper of the Research Institute of Economy, Trade and Industry (RIETI) of Japan.

References

Akcigit, U and S T Ates (2019a), “Ten Facts on Declining Business Dynamism and Lessons from Endogenous Growth Theory”, NBER Working Paper No. 25755.

Akcigit, U and S T Ates (2019b), “What Happened to US Business Dynamism?” NBER Working Paper No. 25756.

Atkeson, A and P J Kehoe (2005), “Modeling and Measuring Organization Capital”, Journal of Political Economy 113(5): 1026-1053.

Bresnahan, T F, E Brynjolfsson and L M Hitt (2002), “Information Technology, Workplace Organization, and the Demand for Skilled Labor: Firm-Level Evidence”, Quarterly Journal of Economics 117(1): 339-376.

Cabral, L and J Mata (2003), “On the Evolution of the Firm Size Distribution: Facts and Theory”, American Economic Review 93(4): 1075-1090.

Chun, H, K Fukao, S Hisa and T Miyagawa (2012), “Measurement of Intangible Investments by Industry and Its Role in Productivity Improvement Utilizing Comparative Studies between Japan and Korea”, RIETI Discussion Paper 12-E-037.

Crouzet, N and J Eberly (2019), “Understanding Weak Capital Investment: The Role of Market Concentration and Intangibles”,  Jackson Hole Economic Policy Symposium 2018: 87-149.

Davis, S J, J C Haltiwanger and S Schuh (1996), Job Creation and Destruction, MIT Press.

De Ridder, M (2019), “Market Power and Innovation in the Intangible Economy”, mimeo.

Fukao, K, T Miyagawa, K Mukai, Y Shinoda and K Tonogi (2009), “Intangible Investment in Japan: Measurement and Contribution to Economic Growth”, Review of Income and Wealth 55(3): 717-736.

Hosono, K, M Takizawa and K Yamanouchi (2020), “Firm Age, Productivity, and Intangible Capital”, RIETI Discussion Paper 20-E-001. 

Hsieh, C and P J Klenow (2009), “Misallocation and Manufacturing TFP in China and India”, Quarterly Journal of Economics 124 (4): 1403–48.

Hsieh, C and P J Klenow (2014), “The Life Cycle of Plants in India and Mexico”, Quarterly Journal of Economics 129(3): 1035–1084.

Jovanovic, B (1982), “Selection and the Evolution of Industry”, Econometrica 50: 649–670.

Sakai, K, I Uesugi and T Watanabe, “Firm Age and the Evolution of Borrowing Costs: Evidence from Japanese Small Firms”, Journal of Banking and Finance 34(8):1970–1981.

1,470 Reads