Who works for startups? The relation between firm age, employee age, and growth

Young firms are often associated with an up or out dynamic. Young firms have high failure rates, but, conditional on surviving, young firms exhibit higher average growth rates as compared to older firms. These differential dynamics between young and old firms raise important questions. What is it about young firms that makes them unique? Do firm characteristics which differentiate between young and older firms also distinguish young firms that will grow rapidly from those that will not?

In this paper, we focus on one specific firm characteristic: the employee workforce. Labor and human capital are important components to production, especially in high-tech and innovative industries where startup activity abounds. This paper documents a number of new facts showing how employees differ at young firms, yielding insight into the characteristics and joint dynamics of young firms and young workers.

Using over a decade׳s worth of firm-level data from the US Census Bureau, we find that young firms employ relatively more young workers. Around 27% of employees in firms aged one to five years are between 25 and 34 years old, and over 70% are under the age of 45. In contrast, in established firms that have been in existence for 20 years or more, fewer than 18% of employees are between the ages of 25 and 34, while almost half are over the age of 45. We find similar results when we control in a regression framework for firm size, industry, geography, and time. Furthermore, we document results consistent with causality. Following a plausibly exogenous increase in young workers, we observe an increase in new firm creation.

Young employees may be more likely to match with young firms for several reasons. Given younger employees will, on average, have had more recent education, they may possess more current technical skills. Building a workforce with such characteristics can be especially critical to young firms, especially those developing new products or establishing new methods of production. In addition, young employees may be a better fit for young firms due to the fact that younger employees are likely to be relatively more risk tolerant. Greater risk tolerance can make young employees more willing to bear the labor income and human capital risk of working for a young firm or to take on riskier projects within the firm.

In addition, young firms will employ more workers who have recently completed a job search by nature of being new. Young workers are likely to switch jobs early on in their careers as they acquire job- and task-specific skills and learn about their own skills and productivity (e.g., Johnson, 1978, Topel and Ward, 1992). Thus, to the extent young workers make up a higher percentage of workers who recently completed job searches, they will be more likely to work at young firms, strictly as a function of the joint dynamics of young firms and young workers. Finally, young firms may disproportionately employ young workers due to assortative matching of workers and firms based on productivity or quality of workers and firms. To the extent that younger firms are, on average, less productive than older firms and younger workers are, on average, less productive than older workers, assortative matching would imply that less productive (young) workers would match to less productive (young) firms.

Each of these four mechanisms may explain part of the positive relation between firm age and employee age in the data. The intent of our analysis is not to identify one unique driver of the relation or even to provide a complete accounting of all the potential underlying mechanisms. Instead, we show how the evidence is consistent with three non-mutually exclusive mechanisms related to the unique skills, greater risk tolerance, and joint dynamics of young firms and young workers.

We find that the young firm, young employee relation is moderated but still significant when we limit the sample to new hires, a set of workers who have all recently undergone a recent job search. The reduction in the magnitude of the effect is consistent with the joint dynamics mechanism. Moreover, the fact that the relation still holds with new hires suggests multiple mechanisms are responsible for the employee age-firm age relation.

Considering wages of young firms and young employees, we find that young firms pay lower wages, on average, a fact consistent with the existing literature. However, interesting patterns emerge when we investigate by employee age. Young employees in young firms earn higher wages than young employees in older firms. Assuming wages proxy for human capital, this result is consistent with the unique skills mechanism and does not directly support the assortative matching mechanism.

Using firm outcomes, we document several more interesting correlations involving employee age. First, firms that are created with a larger share of young employees are more likely to subsequently raise venture capital (VC) financing. Given findings in Hellmann and Puri (2000) that VC investors select more innovative firms, this result is consistent with young employees (those with recently acquired skills especially valuable for innovation) matching to more innovative startups. Second, young firms which employ relatively more young workers subsequently experience higher growth rates (conditional on survival) and, with some qualifications, higher failure rates. These results are consistent with the unique skills and greater risk tolerance of young workers mechanisms.

Finally, we ask whether the rate of new firm creation is affected by the supply of young workers. If young employees are important for young firm growth due to their unique skills or attributes, we should expect that when more young employees are available, entrepreneurs find it easier to start and grow young companies, especially in more innovative industries. Using historical demographic information on the relative ratio of youth in a state as a predictor for the ratio of younger to older workers ten or 20 years later, we argue that a causal relationship exists between the supply of young workers and the rate of new firm creation, especially in high-tech industries, where innovation is greater. These results suggest that the supply of young workers, in addition to the supply of financial capital, is an important factor in the creation and growth of new firms.

Our study contributes to the literature on what drives new firm creation and growth. A large focus of this literature has been on understanding the role of financial market development and structure.¹ We explore the role of labor markets and how the relative supply of young workers can impact firm creation and growth. Our study is related to recent work by Lazear (2005), who examines the human capital traits of entrepreneurs and Doms, Lewis, and Robb (2010), who show that there is more new firm creation in regions where the local labor force is more educated.²

Our results also contribute to the labor and organizational economics literatures in documenting the strong positive association between employee age and firm age and the narrower wage spread between older and younger workers at young firms. Previous studies have explored the relation between firm size and wages (e.g., Brown and Medoff, 1989) and between firm age and wages (e.g., Brown and Medoff, 2003), but none have examined the role of employee age in explaining the relation between firm size and age with wages. Moreover, our results are applicable to the organization economics literature which suggests that firm hierarchies might be flatter in young, entrepreneurial firms (e.g., Rajan and Zingales, 2001).

The remainder of the paper proceeds as follows. Section 2 reviews the relevant literature and considers four mechanisms that could underlie the positive relation between firm age and employee age and their associated empirical predictions. Section 3 describes the data. Section 4 examines the relation between firm age and employee age. Section 5 considers evidence on the mechanisms underlying the relation. Section 6 discusses sample selection issues. Section 7 provides evidence of a causal relation between the supply of young workers and new firm creation. Section 8 concludes.

Possible mechanisms underlying the relation between firm age and employee age

In this section, we briefly review the related literature on firm age, employee age, and firm and career dynamics. We then consider four mechanisms suggested by this body of work that may underlie the positive relation between firm age and employee age observed in the data.

Data

We use four primary data sources in the analysis. We use data from the US Census Bureau׳s Longitudinal Employer-Household Dynamics (LEHD) program to obtain information on the ages and wages of employees. We use data from the US Census Bureau׳s Longitudinal Business Database (LBD) to obtain information on the industry, age, and geography of the firms for which the employees in the LEHD work. We use Compustat to obtain additional information on the publicly traded firms in the sample. Finally, we

The relation between firm age and employee age

We first present the main fact in our paper—that younger firms employ more young workers—before turning to an examination of the mechanisms that may drive this relation. We examine the relation between firm age and employee age for our entire sample, which is dominated by smaller private firms, as well as for a subsample of public firms.

Evidence on mechanisms

In the previous section we show a positive relation between employee age and firm age. In this section, we seek to understand the drivers of this relation by exploring related empirical evidence. We first look at the relation between firm age and age of new hires. Second, we explore average wages for young and older workers at both young and older firms. Next, we look at the relation between firms׳ potential for innovation and employee age. Finally, the correlation between firm outcomes and

Sample selection

Throughout the analysis, we make use of different subsamples. We limit the sample to just public firms (e.g., Table 1, Panel B), firms which are hiring new workers (e.g., Table 3), the cross-section of new firms (columns 1 and 4 of Table 6), new firms which survive for five years (columns 3 and 6 of Table 6), and firms just in high-tech industries (columns 4–6 of Table 6). Repeating similar tests across various samples, such as comparing the firm age-employee age relation at all firms, then at

Does the supply of young workers affect new firm creation?

In the previous sections, we showed that young firms disproportionately employ young workers and presented evidence that the unique skills of young employees, greater risk tolerance of young employees, and the joint dynamics of young firms and employees can partially explain this relation. If young employees are a critical component for young firms, then we should expect to find that exogenous changes to the supply of young workers affect new firm creation. In this section, we test this

Conclusion

We present large-scale evidence that young employees are an important component in the creation and growth of young firms. We first show that young firms disproportionately employ younger workers. Around 27% of employees in firms aged one to five years are between 25 and 34 years old, and over 70% are under the age of 45. In contrast, in established firms that have been in existence for over 20 years, fewer than 18% of employees are between the ages of 25 and 34, while nearly half are over the