Enterprise and Small Business Principles
The ‘n’-shaped puzzle of the US
The previous section provided some accounts for why for many OECD countries have seen business ownership rates increase. Such interpretations can also be stretched to include ‘decreasers’ such as France and Japan if it is believed that these countries are dominated by larger enterprises and a burdensome regulatory environment. What is not clear is why the US has followed an ‘n’-shaped business ownership pattern or why self-employment rates have declined in the US. These questions are important because the US is often seen as a model of technological change, rapid innovation and a country that has been responsible for driving much of the global economic growth in the 1990s. The US is also seen as a country that strongly promotes entrepreneurship and has very few administrative burdens to inhibit individuals setting up or growing a business (Global Entrepreneurship Monitor (GEM), 2004).
One obvious explanation for the ‘n’-shaped distribution of business ownership in the US is that the COMPENDIA dataset is a partial measure of business ownership in the US. Because COMPENDIA is designed to measure international business ownership rates, the particular measure it has chosen (the number of unincorporated and incorporated non-agricultural self-employed as a share of the labour force) may be thought to poorly reflect the actual nature of entrepreneurship in the US. Earlier on, Table 2.5 showed, for example, that the US has significantly more larger enterprises than either Europe or Japan. The COMPENDIA data may then underestimate the contribution of larger enterprises to entrepreneurship in the US as they measure the rate of business ownership rather than the contribution of particular businesses. Indeed, given that the US is a large, well-integrated market with a flexible labour force, ENSR (2004) have suggested that larger enterprises predominate in the US because it allows for the development of economies of scale and scope. The implication, therefore, is that the economic growth of the US is due to the economic activities of larger enterprises.
Another potential explanation for the US is that the ‘n’-shaped pattern is appropriate for the US economy. Carree et al. (2002) argue that there is an equilibrium rate between economic development and business ownership. Using the same COMPENDIA data, they find that countries that deviate from an equilibrium rate suffer: ‘by and large, a five percent point deviation implies a growth loss of three percent over a period of four years’ (p. 285). Given that Carree et al. (2002) find that the US tends to follow the equilibrium rate, this then may give some insights into the economic performance of the US.
These arguments are essentially static arguments: they point to what is happening rather than why it is happening. For example, the US economy may be dominated by large multinational businesses such as Microsoft or Dell, but this ignores the fact that these very enterprises are relatively new. Jovanovic (2001) has shown, for the period 1926-96, that smaller enterprises significantly outperformed larger, more established, enterprises.
A more dynamic explanation for the US is that what matters is not the rate of business ownership but the degree of ‘churn’ (rate of entry and exit) in the economy. Because the US places relatively fewer regulatory burdens on business start-up, entry barriers are lower and this may, in turn, encourage greater numbers of new enterprises. These new entrants stimulate competition in the sector so as to ensure that enterprises that are inefficient exit the sector. The overall effect, therefore, is to drive up efficiency and innovation in the sector.
There is some evidence to support the idea of the dynamic role of new enterprises. GEM (2004), for example, looks at the numbers of individuals in a given economy that are seeking to set up a new venture or who have a new venture. GEM (2004) calls this rate total entrepreneurial activity (TEA). The US, like Australia and New Zealand, had TEA rates that were among the highest in the OECD in 2004. It goes on to argue that there is a correlation between TEA and GDP growth. Similarly, for the UK, Disney et al. (2003) have argued that the dynamic nature of enterprise birth and death can have a large impact on productivity: ‘Between 1980 and 1982, single establishment firms (25% of manufacturing employment) experienced no productivity growth among survivors; all productivity gains for this group came from entry and exit’ (p. 691). Moreover, it may be more than coincidental that the VAT registration and deregistration rates for the UK mirror regional inequalities. Prosperous regions such as London and the South East (e. g. Sussex and Buckinghamshire) had VAT registration rates of 36,600 and 30,300, respectively, in 2003. Their deregistration rates were also high: 34,600 (London) and 27,800 (South East). This compares with registration/deregistration rates of 4,600/4,000 for less prosperous regions such as the North East of England (e. g.
Tyne and Wear) or 3,800/4,000 for Northern Ireland. Armington and Acs (2002) have also shown that there are similar regional disparities in the US. In 1994, the average annual firm birth rate in the US was 3.85 per 1,000 of the labour force. Well above this average were the South and West of the US whilst the Northeast and Midwest of the US were well below this average.
This evidence points to the view that the ‘n’-shaped pattern of business ownership in the US is the outcome of the dynamic nature of entry and exit. If so, it may be anticipated that the US would have higher rates of entry and exit compared with other economies. Evidence from 10 OECD economies suggests (Figure 2.4) that this is not the case.
Figure 2.4 shows that the US tends to have higher rates of manufacturing and service entry and exit than most of the nine other economies but that these differences tend to be relatively small. Moreover, simply encouraging yet larger numbers of new entrants into particular sectors does not necessarily improve efficiency or innovation. Many of these individuals may have skills that are unsuited to running an enterprise. If they enter, they may not have any discernable impact upon efficiency or innovation because they do not have the skills to compete effectively with existing enterprises in the sector (Greene et al., 2004).
An alternative but equally dynamic explanation for business ownership rates in the US is the role played by fast-growth enterprises. Again, as with enterprise churn, what matters here is not the rate of business ownership but the patterns of enterprise activity. Fast-growth enterprises have long been seen as important. Storey (1985) showed that over a ten-year period (with an enterprise exit rate of 60%), 4% of enterprises will
Denmark Spain Netherlands UK Finland Portugal Belgium Sweden Italy
Country
□ Manufacturing entry □ Services entry a Manufacturing a Services exit
contribute 50% of employment. Such a result is fairly typical: Birch et al. (1997), for example, showed that ‘gazelles’ - around 3% of all US enterprises - were responsible for about 70% of gross job growth.
Therefore, the argument is that the US should have higher rates of growth among its enterprise population. This is a view supported by Scarpetta et al. (2002) who suggest that the US, unlike the EU, has lower regulation and stricter employment legislation. This allows individual entrepreneurs to enter a market at a small scale so that they can test their ideas out. Subsequently, those that find that their ideas are novel or efficient are able to expand. The impact of this is shown in Figure 2.5 which shows the employment gains of surviving firms over a two-year period (as a percentage of initial employment). Figure 2.5 clearly shows that US enterprises have rates of employment growth that are more than double the nearest EU country (Finland) and more than four times that of economies such as the UK or Denmark.
This evidence suggests that what matters is fast-growth enterprises. However, it should not be assumed that entry and exit rates are unimportant. Scarpetta et al. (2002) also indicate that entry and exit accounts for from 20-40% of total productivity growth.
This section has shown that there are a variety of explanations for the ‘n’-shaped pattern to business ownership rates in the US. Some of it may be due to measurement issues, the relative size and integration of the US economy or because the US is better able to balance its business ownership rates with economic development. This section has also shown that a more dynamic explanation is the influence of enterprise churn and fast-growth enterprises. These factors, rather than more static interpretations, are likely to drive economic growth in the US and in other OECD countries.
