Zoltan Acs, George Mason University
Erkko Autio, Imperial College Business School
It is widely accepted that entrepreneurs play a central role in economic growth. However, while everyone agrees on this broad notion, the details are much murkier. As a result, there is little real understanding of policies that actually support the creation of the kind of high-growth, high-impact firms that drive economic dynamism.
In policy discussions, the notions of high-growth entrepreneurship, new ventures and small busi-ness are often confused. A frequent misconception is to equate increasing the number of new businesses with the enhancement of entrepreneurship. This approach confuses outputs and the processes that drive those outputs. Large numbers of new businesses do not automatically trans-late into rapid growth.
In fact, job creation does not flow from the creation of numerous tiny businesses. The bulk of new jobs by entrepreneurs are instead the result of a small number of extraordinary high-growth entre-preneurial ventures, or “gazelles”. While there are many societal benefits to small and new firms, the real economic impact is generated by “gazelles”.
In his classic text, Rostow (1960) suggested that countries go through five stages of economic growth. Michael Porter (2002) has provided a modern rendition of Rostow’s typology by identifying three stages of development (as opposed to growth): a factor-driven stage, an efficiency-driven stage, and an innovation-driven stage, and he adds two transitions. While Rostow focused on the age of high mass consumption, Porter’s model encompasses recent developments in the economics of knowledge and innovation. Historically, an elite entrepreneurial class appears to have played a leading role in innovation and economic development.
The factor-driven stage is marked by high rates of agricultural self-employment. Countries in this stage compete through low-cost efficiencies in the production of commodities or low value-added products. Sole proprietorships—i.e., the self-employed—probably account for most small manufacturing firms and service firms. Almost all economies experience this stage of economic development. These countries neither create knowledge for innovation nor use knowledge for exporting.
To compete in the efficiency-driven stage, countries must have efficient productive practices in large markets, which allow companies to exploit economies of scale. Industries in this stage are manufacturers that provide basic services. The efficiency-driven stage is marked by decreasing rates of self-employment. When capital and labour are substitutes, an increase in the capital stock increases returns from working and lowers returns from managing.
The innovation-driven stage is marked by an increase in knowledge-intensive activities (Romer 1990). In the innovation-driven stage knowledge provides the key input. In this stage the focus shifts from firms to agents in possession of new knowledge (Acs et al 2009). The agent decides to start a new firm based on expected net returns from a new product. The innovation-driven stage is biased towards high value added industries in which entrepreneurial activity is important.
According to Sala-I-Martin et al (2007), the first two stages of development are dominated by insti-tutions. In fact, innovation accounts for only about 5 percent of economic activity in factor-driven economies and rises to 10 percent in the efficiency driven stage. However, in the innovation-driven stage when opportunities for productivity gains from factors and efficiency have been exhausted, innovation accounts for 30 percent of economic activity. Many of today’s developing countries are experiencing all three stages of development; this places addition pressure on policy makers, business leaders, and social leaders making decisions in such dynamic socio-economic systems.
We see an S-shaped relationship between entrepreneurship and economic development because in the first transition stage entrepreneurship plays a role but it increases at a decreasing rate as the efficiency stage takes over. However, as a country moves from efficiency-driven stage to innova-tion driven stage (the knowledge-driven stage), entrepreneurship plays a more important role in-creasing at an increasing rate and latter at a decreasing rate (Figure 1).
Figure 1: Entrepreneurship and stages of economic development
Note that Figure 1 highlights the potential impact that entrepreneurship can generate in the econ-omy. If we were to look at raw numbers of new businesses, the s-curve would be turned upside down, because low-income economies are characterised by large numbers of subsistence and self-employment businesses.
The Global Entrepreneurship and Development Index (GEDI)
For a long time, the level of entrepreneurship has been evaluated by some quantitative output measure, such as the self-employment rate, business ownership rate, or business start-ups  . Over the last decade, the Global Entrepreneurship Monitor’s Total Early-stage Entrepreneurial Activity (TEA)  ratio has become a widely used measure of entrepreneurship at the country level.
While these output indicators or ratios have undergone some modification and change to incorpo-rate qualitative measures, like education and high-growth firms, they are basically limited to meas-uring the quantity of existing or nascent businesses  . There are five major shortcomings with these attempts at measuring entrepreneurship:
- While all the definitions emphasise the multifaceted nature of entrepreneurship— including innovation, risk taking, opportunity recognition, high-growth opportunity motivations, and unusual “judgmental” decision-making, they measure only one, and perhaps not even the most important, aspect of entrepreneurship.
- The indexes fail to incorporate the businesses’ differing impacts. A traditional agricultural business established in Uganda or Peru is given equal importance as an Internet-related venture in Silicon Valley.
- The most entrepreneurial nations are defined as those having the largest number of busi-nesses. These are generally the developing countries of Africa or South America.
- These measures do not take into account differences in environmental factors. In fact, the efficiency and sophistication of the institutional setting could have a major influence on the quality of entrepreneurship.
- Since self-employment and the business ownership ratio decline as a country develops, indexes that rely on them appear to show that higher levels of development are associated with decreasing levels of entrepreneurship. This phenomenon is inconsistent with main-stream economic theories which posit a direct connection between entrepreneurship and development.
An output-based index would not nece ssarily give policymakers good guidance, as it can prompt a focus on i ncreasing the quantity of entrepreneurship, when quality is of greater import.
Recent efforts of the O ECD and European Union have aimed to provide a sophisticated measure of entrepreneurship encompassing thre e broad areas: the determinants of entrepreneurship (regu-lation, R&D, entrepreneurial capabilities, culture, access to finance and market conditions); entre-preneurial performance (firms, employment, and wealth); and the impact of entrepreneurship. While the first two publications of the OECD’s Entrepreneurship Indicator Programme contain many entrepreneurship-related data and indicators, a more highly evolved measure of entrepreneurship is still missing.
The shortcomings of previous entrep reneurship indicators and the need to clarify the role of entre-preneu rship in economic development ar e the two major reasons underlying the creation of the Global Entrepreneurship and Development Index (GEDI) GEDI this is the only index to fulfil the three major requirements of entrepreneurship index building, namely,
- Sufficient complexity to capture the multidimensional nature of entrepreneurship;
- Inclusion of indicators encompassing quality-related differences, in addition to quantitative or level-related measures; and
- Inclusion of individual-level as well as institutional variables.
Unlike other entrepreneurship indexes, the relationship between the GEDI and economic develop-ment appears to be mildly S-shaped, implying a positive relationship between entrepreneurship and economic development  . Therefore, the GEDI is a potentially useful tool to provide policy suggestions to increase economic development via entrepreneurship. However, it is acknowledged that since economic growth is ultimately the result of many factors in addition to entrepreneurship, the GEDI can explain only a part of short-term economic growth.
Table 1: The GEDI rankings
Understanding GEDI and its sub-indexes
The Global Entrepreneurship and Development Index (GEDI) is comprised of three sub-indexes that capture the contextual features of entrepreneurship across individual and institutional variable.
For the first sub-index, entrepreneurial attitudes are defined as the general disposition of a coun-try’s population toward entrepreneurs, entrepreneurship, and business start-ups. The index in-volves measures for the population’s opportunity perception potential, the perceived start-up skills, feel of fear of failure, networking prospects, and cultural respect for the entrepreneur.
For the second sub-index, entrepreneurial activity is defined as the start-up activity in the medium- or high- technology sector initiated by educated entrepreneurs in response to business opportunities in a somewhat competitive environment. The choice of indicators used to build this sub-index reflects the belief that opportunity entrepreneurs are better prepared, possess superior skills, and earn more than necessity entrepreneurs (Bhola, Verheul, Thurik, & Grilo, 2006; Block & Wagner, 2006). Operating in the technology sector is important, as high rates of start-ups in most factor-driven countries are mainly in the traditional sectors and do not represent high potential (Acs & Varga, 2005). The entrepreneur’s level of education is another important feature of a venture with high growth potential (Bates, 1990). And cut-throat competition may hinder business existence and growth, so a lower number of competitors improve chances of survival, as well as future develop-ment prospects (Baumol et al., 2007).
The third sub-index, entrepreneurial aspiration, is defined as the efforts of the early-stage entrepre-neur to introduce new products and services, develop new production processes, penetrate foreign markets, substantially increase the number of firm employees, and finance the business with either formal or informal venture capital, or both. Product and process innovation, internationalisation, as well as high growth are included in the measure. The capability to produce or sell products that customers consider to be new is one of Schumpeter’s forms of creating ‘new combinations’ (Schumpeter, 1934). Applying or creating new technology and production processes are another important feature of businesses with high growth potential (Acs & Varga, 2005). The role of ‘gazelles’ or high-growth businesses is also vital to entrepreneurship as are internationalisation of trade and the availability of risk capital.
A schematic diagram of GEDI and its sub-indexes is presented below in Figure 2.
Figure 2: The structure of the GEDI
All Sub-Indexes are Not Equal, and Some Can Even Hurt
It must be noted that the three entrepreneurial sub-indexes are not of equal importance.
The attitude sub-index measures society’s basic attitudes toward entrepreneurship through educa-tion and social stability. The activity sub-index measures what individuals are actually doing to improve the quality of hu-man resources and technological efficiency. The aspiration sub-index measures how much of the entrepreneurial activity is being directed to-ward innovation, high-impact entrepreneurship, and globalisation.
It must also be noted, that because entrepreneurship is a complex social phenomena, the attrib-utes used in the sub-indexes, are more powerful collectively than they are individually. This is what makes true entrepreneurship, the killer application for sustainable economic development.
Therefore GEDI is built using configuration theory, which lowers sub-index scores if there is a shortage or low level score on its components. The low scores can act as ‘bottlenecks’ to entrepreneurship and can therefore have negative results on entrepreneurial activity. The wider the range between low and high scores, the more severe the ‘bottleneck’ penalty applied. The most entrepreneurial economies are both broad and deep across most of the components of the 3 GEDI sub-indexes: Attitudes, Activity, and Aspirations.
Additionally, the sequence of these sub-indexes in development is also important. Attitudes are an essential prerequisite for either activity or aspirations. This is in part cultural, as certain societies (e.g., communism and feudalism) outlawed entrepreneurship.
Attitude is followed by activity, and after activity aspirations become important. In some sense, this process is cumulative over time; however it has large overlaps as well. Figure 3 depicts the sub-index that corresponds to each stage of economic development.
Figure 3 Mapping the Sub-Indexes onto Stages of Development
Using GEDI for Policy Design
Entrepreneurship, the high-impact type that creates jobs, societal wealth, and improvements in standards of living, can be found at a variety of le vels that may differ according to the level of development in a give n country. In a factor-driven (agricultural economy) the focus needs to be on entrepreneurial attitudes in the p opulation. In an efficiency-driven economy (manufacturing), indi-vidual entrepreneurs need to be encouraged to be entrepreneurs and start businesses. In an inno-vation-driven economy (knowledge-based economy) some people need to create very large and successful businesses.
A third important aspect of development is the roles of institutional and individual variables. While institutional improvement is vital for factor-driven countries to advance to the next level of development, the enhancement of individual characteristics is increasingly critical for innovation-driven economies.
The recent economic turmoil has highlighted shortcomings in current economic models and prompted a search for new sources of economic growth. Governments in developed countries face the challenge of scaling down the public sector and replacing lost jobs by encouraging private-sector developments, but without sacrificing productivity. This challenge has prompted renewed interest in entrepreneurship. However, past experiences in entrepreneurship support have shown that not enough is known about the sources of the kind of entrepreneurial activity that truly drives economic dynamism.
To date, the tools for understanding entrepreneurship have been limited, often counting firms, and focusing on either individual or institutional variables. The GEDI index, with its contextual approach and applicability to the three stages of development, offers a more complete view by shifting the focus from outputs to the institutions and processes that drive those outputs. With its focus on processes rather than outputs, the GEDI index offers promise of generating more policy-relevant insights than would be possible by relying on output indicators alone.
 The self-employment rate measures the proportion of the adult population who are self-employed and not employees (Blanchflower 2000; Blanchflower et al. 2001). The business ownership rate is the proportion of the population at some stage of business ownership, excluding public firms and mutual funds. (Caree et al 2003) Business density is defined as the number of firms per 1,000 persons (Lowrey 2004).
 The Total Early-stage Entrepreneurial Activity (TEA) index measures the percentage of a country’s working-age population who are actively trying to start a new business (nascent entrepreneurs) and those who at least partially own and manage a business less than 3.5 years old (a baby business) (Reynolds et al 2005), (Bosma, et al 2008, 2009).
 For more details see Iversen et al. (2008).
 Wennekers et al. (1999).
 See Shane (2009) for a critique of this.
 Understanding Entrepreneurship (OECD 2006); Measuring Entrepreneurship (OECD 2009).
 The previous version of the index can be found in Acs and Szerb (2009).
 Other indexes yield a U-shape (TEA) or L-shape (business ownership, self employment) relationship. Recent research seems to support the upward trend of self-employment, the “U” shape phenomenon (Wennekers et al. 2010)