The self-employment sector is often seen as a driver of innovation and economic growth where entrepreneurs incubate and develop new ideas into lucrative businesses. Indeed, virtually every company begins life through an entrepreneurial process and many are born within a framework of self-employment. However, not every self-employed business grows to become an economic powerhouse. In fact, most new companies fail within a few years. Most that continue longer remain small, providing no more than a livelihood for their owners, many of whom are also the main, or only, source of labor, and whose retirement from work signifies the end of the business.

Lucas and others have asserted that self-employment rates fall with economic growth as successful firms achieve economies of scale, allowing those firms to raise wages and hire workers out of the self-employment sector. In line with this, the Korean self-employment rate has fallen from nearly 70% in the 1960s to close to 25% today. However, 25% is still exceptionally high compared with other countries at a similar level of development. For example, the self-employment rate is well below 10% in the US and only slightly higher than 10% in Japan. Only three OECD countries (Turkey, Greece and Mexico) have higher self-employment rates than Korea.*

In Western countries, many express a desire to become self-employed and those in self-employment report higher life satisfaction than wage earners. However, Korean self-employed are characterized not by an awareness of the opportunities they see in their businesses, but in a sense of despair at the lack of alternatives. The self-employed in Korea are concentrated in a few service businesses (e.g. food service, personal and goods transportation, personal services, retail sales), and compared to wage earners, the self-employed suffer from higher household debt and work longer hours. On average, the self-employed are also older and have lower levels of education than wage earners.

A current academic and policy topic is whether the Korean self-employment rate is too high, and if so, why and what can be done to reduce it. Some research has concluded that the Korean self-employment rate should be somewhat lower and has linked high self-employment rates to lower economic growth. Various reasons have been suggested for the high Korean self-employment rate. These include jobless economic growth across the economy as a whole, particularly with a shortage of options for older workers; lack of a social safety net for the unemployed and retired; the combination of early retirement age for most career workers and the large cash severance packages they receive upon retirement and the proliferation of franchising in recent years.

Much self-employment-focused public policy in Korea (what there is of it) is directed at supporting so-called “small merchants” through business consulting, market protection and financial support. The Korean term for small merchants (소상공인) is even defined in laws and regulations supporting them, with this definition of small merchants largely, but not fully, overlapping with the “weak self-employment” definition I propose in my research (which is described below).

Another relevant issue is lagging development of the Korean service sector as a whole. The service sector encompasses most economic activity not part of the manufacturing, agriculture and public sectors. It includes everything from restaurants and retail outlets, to medical, legal and transportation services. The service sector in Korea makes up nearly 70% of total Korean employment and over 80% of self-employed work in the service sector. However, productivity in the Korean service sector is strikingly low. Average service-sector per capita productivity in OECD countries is around 92% that of manufacturing, but in Korea, per capita productivity in services barely exceeds 40% of the level in manufacturing. In certain service industries (especially those with high self-employment rates), productivity levels languish below 25% of the manufacturing average. Furthermore, even from this low base, productivity increases in key service businesses are not keeping up with productivity growth in manufacturing.

I am investigating several research questions related to the Korean service and self-employment sectors. If self-employment is linked to entrepreneurship as a a driver of economic development, why are the high levels of self-employment in Korea not being celebrated? What are some links between high self-employment rates and low service sector productivity. If Korean self-employment levels are in fact too high, is public policy supporting self-employment contributing to the national economy overall or is it undermining development? What meaningful measures can be taken to improve the Korean service sector in the context of self-employment, as well as the lives of Koreans working in the self-employment sector.

A first step in answering these questions is understanding the heterogeneous reality within the self-employment sector. I identify three types of self-employed, each of which exhibits unique characteristics, entrepreneurial motivations and economic functions. The first type is the traditional entrepreneur, someone who invests and innovates to build a business that provides returns to its investors and promotes economic growth through higher productivity and employment. A second type of self-employed is someone I refer to as a “professional-type”. A professional self-employed is a person with in-demand, high personal-capital skills offering a professional service who could find a good wage job in his or her specialty but instead has chosen to work independently to earn more, enjoy better working conditions (such as working from home or while travelling), or to just have more control over his or her work schedule and processes. The professional self-employed fills critical needs in the market and is rewarded well for the value he or she creates. Finally, the third type of entrepreneur is someone without unique skills who becomes self-employed as a way to earn a living, in many cases, due to being unable to find a job. The so-called “weak-type” self-employed does not innovate or successfully grow the company but uses the business mainly as a means of subsistence. Motivation for self-employment is commonly explained by the “push-pull” hypothesis. Under this hypothesis, some become self-employed by being “pushed” into it due to lack of alternatives (the weak self-employed), and others join the self-employment sector by being “pulled” by the opportunities they see (the entrepreneur and professional self-employed).

My research focuses on these weak self-employed in the Korean service sector. I am not so much concerned with the dynamics of their plight, but rather with the negative impact their presence may have on development of the Korean service sector. To focus my analysis effectively, I have defined a new concept, which I refer to as “self-employment congestion” and a new metric called the “self-employment congestion rate”. In contrast to the self-employment rate, which measures the proportion of non-wage earners in the labor force (and thus covers all three types of self-employed), the self-employment congestion rate attempts to capture just the proportion of weak self-employed in the labor force. Weak self-employed are defined (in currently updated form; we used a slightly different definition in 스티븐 밤멜ㆍ서환주 (2017)) as non-wage earners with reported earnings less than the average reported earnings of all workers across the entire economy (both wage earners and non-wage earners) in the respective year and who do not report having any paid employees. This definition includes self-employed working with unpaid family workers, as well as the unpaid family workers themselves.

I received inspiration for the concept of self-employment congestion from recent research published by the OECD (McGowan, Müge Adalet and Dan Andrews & Valentine Millet (2017), “The Walking Dead? Zombie Firms and Productivity Performance in OECD Countries,” Economics Department Working Papers, No. 1372, OECD). This paper investigates the negative effects of so-called zombie firms on the economic performance of non-zombie firms. Zombie firms are defined as non-competitive companies kept alive by bank forbearance and other support measures provided to avoid the unemployment that would result from closing such zombie firms. McGowan, et al. defines “zombie congestion” as the proportion of total capital tied up in zombie firms in a respective industry.

Applying a similar logic, my research asserts that weak self-employed remain in self-employment even at low income levels due to lack of alternatives. This is in spite of the fact that, from an overall economic standpoint, it would be better if they were in wage positions or out of the employment market altogether. In 스티븐 밤멜ㆍ서환주 (2017), we attempt to demonstrate that high self-employment congestion in a particular service business has a negative effect both on the earnings of the weak self-employed themselves and on other participants in the same business. Under McGowan, et al., zombie congestion assumes an unnatural supply of funding, which drives up wages and thus, maintains the demand for labor at an unnatural level. On the other hand, under the concept of self-employment congestion, an unnatural surplus of labor pushes down the return on labor, leading to an unnatural degree of competition, thus reducing overall ROI in the market, and as a result, reduces innovation and drags down economic development in the Korean service sector.

In 스티븐 밤멜ㆍ서환주 (2017), we demonstrated that self-employment congestion has a negative impact on the earnings of self-employed people working in service businesses. This would seem to be a self-evident and unremarkable finding. However, we also demonstrated that self-employment congestion has a negative impact on the earnings of wage earners working in service businesses with high levels of self-employment congestion (though not in businesses with low levels of self-employment congestion). In addition, we demonstrated that the negative effect of self-employment congestion in low self-employment congestion businesses is focused on lower earning self-employed persons, but that the negative effects of self-employment congestion in high self-employment congestion businesses weigh more heavily on higher earning self-employed businesses, and especially on the very highest earning group.

This last conclusion implies a lack of differentiation (and by extension, ability to innovate) in high self-employment congestion service businesses because high levels of self-employment congestion should not otherwise have such a negative effect under effective quality differentiation. The reasoning being that if providers were able to differentiate effectively at higher quality levels, then regardless of the congestion at the bottom of the market, buyers would simply buy from higher quality providers. However, without differentiation, increased congestion results in higher market encroachment on those with more market share (i.e. the more successful ones). This implies that price competition is the main operator in these service businesses (mainly businesses with low barriers to entry and relatively little opportunity for innovation), preventing even successful market participants from breaking free of the self-employment congestion headwinds and achieving economies of scale.

Our analysis indicates that the effects of self-employment congestion are not just limited to the weak self-employment sector itself and that self-employment congestion has broader negative effects, causing difficulties to both wage earners and more successful self-employed (entrepreneur and professional types), leaving open the possibility that high levels of weak self-employment may be a factor holding back development of the Korean service sector as a whole. This conclusion calls into question the wisdom of government efforts to promote self-employment in Korea, suggesting that stronger efforts to guide weak self-employed into wage earning, unemployment or retirement could free up resources to positively contribute to development of the Korean service sector.

Notably, running the same set of analyses using the self-employment rate (rather than the self-employment congestion rate) does not produce significant results. Therefore, the fact that these results were achieved using self-employment congestion rates, but that the results were not replicated with self-employment rates, strongly supports the concept of important heterogeneity in the Korean self-employment sector, and that efforts to study self-employment should take into account these different types of self-employed (entrepreneurial, professional and weak) in order to achieve more meaningful conclusions about the sector as a whole.

At this point, I see several potential pathways for further research. I would like to investigate the channels through which self-employment congestion negatively impacts the economic results and activities of non-weak self-employed and wage earners. I would also like to further reinforce the concepts presented here by finding other impact channels and comparing results of self-employment congestion with self-employment rate-based analyses using a wider range of data, variables and analytical methods, as well as data from other countries. It would also be interesting to look at the factors promoting higher self-employment congestion, potentially including franchising and employment market dysfunction. These conclusions could then be applied to policy recommendations that productively inform government policy toward the self-employed in Korea and elsewhere.

* This overview is largely based on 스티븐 밤멜ㆍ서환주 (2017), “자영업 혼잡의 경제적 영향에 대한 분석: 서비스부분 자영업자와 임금근로자의 소득에 미치는 영향을 중심으로,” 산업혁신연구, 제33권, 제4호, pp. 145-174. See 스티븐 밤멜ㆍ서환주 (2017) for detailed citations. While the underlying research effort is mainly mine, the paper was a joint effort with Hwan-Joo Seo, my advisor in the Ph.D. program at Hanyang University, ERICA Campus in Ansan, Korea.