With these questions in mind, the occupational choice between wage work and entrepreneurship when people are heterogeneous in their ability as workers and start-ups differ in productivity. This decision has implications for the firm productivity distribution and, through this, for aggregate productivity. A substantial number of people now a days, choose to become entrepreneurs. In the U.S., for instance, the ratio of entrepreneurs to entrepreneurs plus wage and salary workers is 12.8%, using CPS data for 2003 cited in Hipple (2004). This rate is even higher in most other industrialized economies (see Blanchflower 2000). Of course, who becomes an entrepreneur may matter a lot for the firm productivity distribution, aggregate productivity, and welfare.
Before modelling the occupational choice, I review some relevant facts about entrepreneurship from the empirical literature. First, do the most or the least able become entrepreneurs? A priori, this is not clear, and depends on what type of firm one thinks of. Lazear (2005), for instance, puts it this way: It is tempting to argue that- the most talented people become entrepreneurs because they have the skills required to engage in creative activity. Perhaps so, but this flies in the face of some facts. The man who opens up a small dry-cleaning shop with two employees might be termed an entrepreneur, whereas the half-million-dollar-per-year executive whose suit he cleans is someone else’s employee. It is unlikely that the shop owner is more able than the typical executive. The reverse might be true. As necessity is the mother of invention, perhaps entrepreneurs are created when a worker has no alternatives. Rather than coming from the top of the ability distribution, they are what is left over. This argument also flies in the face of some facts. Any ability measure that classifies John D. Rockefeller, Andrew Carnegie, or, more recently, Bill Gates near the bottom of the distribution needs to be questioned.
Some facts on entrepreneurship
- The relationship between entrepreneurship and education is U-shaped, i.e. people with low or high levels of education are more likely to be entrepreneurs than people with intermediate levels of education.
- There is a substantial fraction of people who become entrepreneurs “out of necessity”, and not to pursue an opportunity.
- Most firms are small. Most of these firms remain small and, conditional on age, are not much more likely to exit than their larger counterparts.
- Returns to entrepreneurship have a much higher cross-sectional variance than wages.
Lucas (1978) built a model where agents, depending on their entrepreneurial abilities, choose between either being entrepreneurs or workers. On a different type of model, developed by Kihlstrom and Laffont (1979), agents differ on their level of risk aversion: agents with low risk aversion will chose entrepreneurial activities. Lucas (1978) and Kihlstrom and Laffont (1979) models are the cornerstones of modern economic theories of entrepreneurship. But numerous gaps in our knowledge remain. With regard to the famous Lucas (1978) model, for example, we still know relatively little about where entrepreneurial ability comes from, and to what extent it can be shaped by schooling and enterprise education. This relates to the age-old question of whether entrepreneurs are “born or made”.
In the classical Lucas (1978)/Rosen (1982) model of occupational choice, individuals with greater managerial skills - defined as the ability to extract more output from a given combination of capital and labor - will sort into entrepreneurship because the return from managing a firm exceeds the wage they can earn working as employees. Entrepreneurial skills can be interpreted more broadly to include, for instance, the ability to manage (and stand) risk, and the capacity to identify and assess the economic potential of a new product or process. While it remains unclear how people obtain these skills, it is important to understand whether such skills are innate characteristics or are instead acquired through learning - and if so, how.
Distinguishing between these two sources of entrepreneurial skills (i.e., innate or learned) is of practical importance. If entrepreneurial ability is innate, then its distribution should not differ substantially across populations and much of the observed differences in entrepreneurship across countries or regions within countries should be traced back to factors that facilitate or discourage people with entrepreneurial abilities to set up a firm - such as the availability of capital. Fostering entrepreneurship then requires removing these obstacles. If instead managerial abilities can be acquired through learning, differences in entrepreneurship can partly reflect differences in learning opportunities across countries or regions and the constraints to entrepreneurship are induced by learning frictions. Fostering entrepreneurship requires improving the learning process.
Entrepreneurial income is:
π(x) = xg(k, l) − rk − wl − c
where x is entrepreneurial ability, k capital, l labor, r the rental price of capital, w the wage and c a fixed entry cost.
Lucas (1978), the model implies that individuals with ability above a given threshold will become entrepreneurs, as entrepreneurial profits monotonically increase with ability. Given a threshold z, the probability of becoming an entrepreneur is prob(x > z) = 1 − f(z, λ).
The parameter λ is a shifter of the distribution of talent. It represents the learning opportunities that characterize each location. Hence, individuals who grow up in a high λ region have higher entrepreneurial talent.
The seminal paper of occupational choice and firm size distribution of an economy is Lucas (1978). Individuals have heterogenous one-dimensional abilities as entrepreneurs and choose between entrepreneurship and paid employment. The most talented individuals become entrepreneurs and the less talented ones become workers. The ability differentials across entrepreneurs give rise to different spans of control (firm sizes).
Two main predictions of Lucas’ model are that the mean returns to entrepreneurship are greater than average wages and that the return distributions of entrepreneurs and workers have non-overlapping supports.
These two predictions stand in contrast to empirical evidence on the returns to entrepreneurship. First, the returns to entrepreneurship are found, on average, not to be higher than wages. For example, Hamilton (2000) finds that after 10 years in business the median entrepreneurial earnings are 35 percent less than those on a paid job of the same duration. Similarly, Moskovitz and Vissing-Jorgensen (2002) find that the returns to entrepreneurship are, on average, not different from the return on a diversified publicly traded portfolio—the private equity puzzle. Second, the returns to entrepreneurship are found to be highly variable, more than wages, and more than the returns on public equity (Borjas and Bronars (1989), Hamilton (2000), and Moskovitz and Vissing-Jorgensen (2002)). Hence, the empirical return distributions of entrepreneurs and workers have overlapping supports.
However, these representations of the occupational choices of agents are probably adequate for developed countries, where most agents are either entrepreneurs or workers. Nonetheless, these models are an inadequate way of representing the labor force distribution from less developed countries, where an important proportion of the population are self-employed workers. Any research work whose main purpose is to analyze and comprehend the main economic and social problems faced by the poorest people in developing countries must include study self-employment formation
Banerjee and Newman (1993), built a model where self-employment is an occupational choice and the decisions are based on an initial wealth distribution. Because of the existence of a warranty, rich individuals can receive a loan in order become high scale entrepreneurs, while agents located in the middle of the initial wealth distribution receive smaller loans that allows them to enter to self-employment with a low scale production process. On the other hand, agents in the lower end of the wealth distribution, without high enough collateral, can only join wage-employment.
However, it is important to notice that, in developing countries, an important proportion of agents that have self-employment as their occupational choice, live on economic activities that provide only a subsistence level of income and are poorer than wage earners. Furthermore, some empirical studies show that self-employment, besides been an important and growing sector in some developing countries, it can be found mostly in the lower end of the income distribution. Therefore, it seems that the model developed by Banerjee and Newman (1993) does not accommodate these stylized facts for developing countries.