The macroeconomic footprint of the Elons

Sónia Félix, Sudipto Karmakar, Petr Sedláček 05 November 2021

a

A

Elon Musk is the (co-)founder of Tesla, SpaceX, Neuralink and The Boring Company and was previously involved in firms such as Zip2, PayPal or OpenAI. His current businesses have created an estimated 110,000 jobs and, at the time of writing, Forbes ranked him as the richest person on the planet. Serial entrepreneurs are not limited to the well-known examples such as Elon Musk, Sir Richard Branson or Oprah Winfrey; serial entrepreneurship is widespread.

Business dynamism has long been recognised as a key driver of aggregate outcomes, with startups and young firms playing a particularly important role (e.g. Haltiwanger 2012). Seemingly small changes to the numbers and growth potential of startups may leave large and persistent footprints on the aggregate economy (see e.g. Clementi and Palazzo 2016, Sedláček and Sterk 2020). Therefore, understanding the sources of firm performance is of macroeconomic importance and can help inform policy.

In a recent paper (Félix et al. 2021), we argue that entrepreneurs themselves are key determinants of firm performance. Our primary source of information is a unique administrative dataset from Portugal, the Quadros de Pessoal. The rare advantage of the Quadros de Pessoal is that it allows us to link individual business owners to their firms and follow both individual owners and the performance of their firms over time. Using this dataset, we document how firms of serial entrepreneurs compare to all other, ‘regular’ businesses and how important they are for aggregate outcomes, including top income inequality.

The serial entrepreneur premia

We begin by defining serial entrepreneurship as the simultaneous ownership of multiple businesses. Using this categorisation, we show that about 18% of all businesses are owned by serial entrepreneurs. Moreover, the sectoral composition of serial entrepreneurship suggests that it occurs throughout the entire economy and is not an obscure feature of only a few industries.

Next, we estimate ‘serial entrepreneur premia’ – the average difference in a particular characteristic (e.g. growth or size) between serial entrepreneurs and regular firms. We find that firms of serial entrepreneurs are about 60% larger, roughly 25% less likely to shut down, grow about 35% faster, and are 34% more productive compared to regular firms.

Moreover, these premia are present throughout firms’ life cycles. Figure 1 provides a glimpse at some of these differences. In particular, it plots average firm size and exit rates by firm age, distinguishing between serial entrepreneurs and regular businesses. The figure clearly shows the superior performance of serial entrepreneurs’ firms over their entire life cycles.

Figure 1 Firm size and exit rates by age: Serial entrepreneur and regular firms

Source: Félix et al. (2021).

Finally, we also document that this group of businesses alone accounts for more than one-third of aggregate job creation and productivity growth. Given that they represent less than one-fifth of all businesses, these values further underscore their superior performance and importance for the aggregate economy.

Selection versus learning

But what lies behind the superior performance of serial-entrepreneur firms? One way to think about these differences between firms is that serial entrepreneurs are ‘better’ at running businesses – for example, due to innate personality traits or favourable family backgrounds (e.g. Lindquist et al. 2016, Dalton et al. 2019). An alternative view is that business owners gradually learn (e.g. Guiso et al. 2016) or that their businesses are ‘lucky’ during the course of their firms’ operation (e.g. Hvide and Melling 2019). 

The distinction between such ex-ante and ex-post heterogeneity is important not only for understanding aggregate outcomes (e.g. Pugsley et al. 2018) but also for policy design (e.g. Lindquist et al. 2016).

To address the above, we explicitly distinguish between ‘first’ and ‘subsequent’ businesses of serial entrepreneurs. First businesses are the very first firms serial entrepreneurs own, i.e. before owning multiple businesses and – effectively – before being classified as a serial entrepreneur. 

Comparing the performance of the first businesses of serial entrepreneurs, their subsequent businesses, and regular firms allows us to gauge the extent to which the superior performance of serial-entrepreneur firms relative to regular businesses is present from the onset (i.e. also for first firms of serial entrepreneurs) or develops only gradually over time (i.e. only for subsequent firms of serial entrepreneurs).

In our data, we observe that the first and subsequent businesses of serial entrepreneurs have very similar dynamics, substantially outperforming regular businesses (Figure 2). Our results, therefore, point to (selection on) ex-ante heterogeneity as a key source of serial entrepreneurs’ success, rather than learning or favourable ex-post shocks.

Figure 2 Firm size and exit rates by age: Regular firms and first and subsequent firms of serial entrepreneurs

Source: Félix et al. (2021).

As a next step, we make use of the fact that the Quadros de Pessoal dataset includes several observable characteristics of business owners. Estimating the contributions of individual owner characteristics reveals that age and especially education (e.g. Choi et al. 2021) can account for over 20% of the estimated premia. However, this still leaves a large chunk of the estimated differences between serial-entrepreneur and regular firms unexplained, paving an avenue for future research.

Serial entrepreneurship and top income inequality

Finally, we investigate the role serial entrepreneurs play for top income inequality (defined as the share of income accounted for by the top 10% earners). We borrow a simple analytical framework of entrepreneurship and inequality from Jones and Kim (2018) and extend it for the presence of serial entrepreneurs. Within this framework, we show with pencil and paper that the prevalence of serial entrepreneurs increases income inequality. 

The intuition is simple – the ownership of multiple businesses serves, among other things, as a diversification of business risk. Therefore, serial entrepreneurs enjoy, on average, longer periods during which their (multiple) businesses remain in operation. This, in turn, provides them with an opportunity to grow their income for longer.

We generalise this framework further to allow for serial entrepreneur premia in firm size, growth and exit and estimate this model by matching a range of moments in the Quadros de Pessoal dataset. Our results suggest that, although serial entrepreneurs represent fewer than 3% of all business owners, they account for 10–20% of top income inequality.

Concluding remarks

A large body of research strives to understand the sources, and aggregate consequences, of firm-level growth. Our findings contribute to these studies by pointing to the role of entrepreneurs themselves as key drivers of firms’ success. 

Our results serve two distinct purposes. First, individually our results can be used to assess, develop, and parametrise (macroeconomic) models with heterogeneous firms which have gained in importance in recent years. Second, our results illustrate that studying serial entrepreneurship is not an obscure research niche but can help further our understanding of key economic questions.

Moreover, we believe that our results open up several avenues for future research. Should current state-of-the-art macroeconomic models with heterogeneous firms incorporate the possibility of serial entrepreneurship and, if so, how? Do firms of serial entrepreneurs respond to shocks and policies in the same way as all other businesses? Does serial entrepreneurship open the door to novel, as of yet unexplored, policy measures?

Therefore, although Elon Musk alone has over 61 million followers on Twitter, we think that serial entrepreneurs deserve more attention still.

References

Choi, J, N Goldschlag, J Haltiwanger and D Kim (2021), “Founding teams and startup performance”, NBER Working Paper 28417.

Clementi, G L, and D Palazzo (2016), “Entry, exit, firm dynamics, and aggregate fluctuations”, American Economic Journal: Macroeconomics 8(3): 1–41.

Dalton, M, S Pekkala Kerr and W Kerr (2019), “Entrepreneurial personalities”, VoxEU.org, 6 September.

Félix, S, S Karmakar and P Sedláček (2021), “Serial entrepreneurs, the macroeconomy and top income inequality”, CEPR Discussion Paper 16449.

Guiso, L, L Pistaferri and F Schivardi (2016), “Learning entrepreneurship from other entrepreneurs: Evidence from Italy”, VoxEU.org, 3 April.

Haltiwanger, J (2012), “Job creation and firm dynamics in the US”, Innovation Policy and the Economy, NBER, 17–38.

Hvide, H, and T Meling (2019), “Startups and the long-run importance of luck: New evidence”, VoxEU.org, 16 December.

Lindquist, M, J Sol, M van Praag and T Vladasel (2016), “The importance of family background and neighbourhood effects as determinants of entrepreneurship”, VoxEU.org, 2 December.

Pugsley, B, P Sedláček and V Sterk (2018), “Disappearing gazelles: New evidence from administrative data”, VoxEU.org, 11 May.

Sedláček, P, and V Sterk (2020), “Startups and employment following the COVID-19 pandemic: A calculator”, VoxEU.org, 25 April.

a

A

Topics:  Productivity and Innovation

Tags:  entrepreneurs, entrepreneurship, firm growth, personality traits, Portugal

Economist, Bank of Portugal; Research Fellow, Nova School of Business and Economics

Senior Research Economist, Bank of England; Research Fellow, King’s College London

Professor of Economics, University of Oxford; CEPR Research Affiliate

Events

CEPR Policy Research