Why we need national systems of entrepreneurship

Promoting entrepreneurialism is increasingly seen as a central plank of national economic strategy, as evidenced by a proliferation of entrepreneurship policies.

But what does the notion of an ‘entrepreneurial country’ mean? By now, it is widely accepted that an ‘entrepreneurial’ country does not simply mean that there are more entrepreneurs. After all, Uganda has the highest self-employment rate on the planet, followed closely by countries such as Peru. Even if these two countries have many merits, they are hardly leading examples of economic productivity and dynamism.

Historically, there have been two major approaches to understanding what makes countries ‘entrepreneurial’. In the 1990s, there was much enthusiasm around the notion of National Systems of Innovation (NSIs). This theory was all about structure: how institutions drive knowledge production and application. Because of its focus on structure, it overlooked individual agency: the individuals were nowhere to be seen. As a result, the theory became static, and its popularity has gradually waned.

Trade-offs and opportunity costs

In contrast, entrepreneurship researchers have focused myopically on the individual and tended to ignore the regulating effect of context on individual action. Yet, we know that the majority of the trade-offs and opportunity costs faced by entrepreneurs are regulated by context – for example national policies, resource distribution mechanisms, market access, social norms, and so on.

Because entrepreneurship researchers have focused on the individual and ignored the context, they have missed three important points:

(1) that it is the context that regulates who decides to start a new firm;

(2) it is the context that regulates what kind of firm they will start; 

(3) context also decides how aggressively the firm will pursue growth and with what outcomes.

In other words, the NSI literature has ignored the individual, and entrepreneurship researchers have ignored the context. Yet, we know that entrepreneurship is embedded action: both the individual and the context matter. What is needed, therefore, is a systemic understanding of the entrepreneurial process. The National Systems of Entrepreneurship (NSE) approach seeks to address this gap.

Resource allocation

In the NSE perspective, the fundamental aspect of entrepreneurship is that it drives productive resource allocation in countries. This is a trial-and-error process, driven by individuals who allocate resources under uncertainty. Over time, unsuccessful attempts are selected out, whereas successful attempts are retained and result in firm-level profitability and growth. At the macro level, this dynamic resource reallocation drives total factor productivity, and therefore, economic growth.

The notion of entrepreneurship as a systemic process has important implications. By definition, systems are made up of sub-systems and components which interact to produce system-level performance. In consequence, system-level performance may be held back by bottleneck factors.

Therefore, the main emphasis of policy should be on identifying system-level bottlenecks and alleviating them. A corollary of the above is that we may be able to improve system-level performance by smart (re)allocation of policy resources. However, to do this successfully, one needs to understand the relative strengths and weaknesses of the multiple factors that make up the system.

The fourteen pillars

The Global Entrepreneurship and Development Index (GEDI: www.thegedi.org) seeks to achieve this outcome. It draws on the Global Entrepreneurship Monitor (GEM) data and the NSE philosophy to compile a multi-item index for profiling NSEs in different countries. The index is made up of fourteen ‘pillars’, which reflect ‘attitudes’, ‘activities’, and ‘aspirations’, respectively.

Each pillar combines an individual-level aggregate from the GEM data (for example, national percentage of individuals who perceive skills for entrepreneurship) with a matching measure of a national descriptor (for example, gross enrollment in tertiary education). The product of the two contextualises the measure – in this case, start-up skill perceptions are likely to be more impactful if the population is highly educated, as this will increase the resulting quality of new entrepreneurs.

The GEDI index also applies a Penalty of Bottleneck algorithm: if there are poorly performing pillars, other pillar values are ‘penalised’ to reflect the notion that the poorly performing pillar may constitute a bottleneck for system performance. With this Penalty for Bottleneck algorithm, it is possible to start ‘optimising’ policy portfolios – that is determine how policy resources should be allocated if policies should always seek to address the most weakly performing pillars first.

What, then, does the GEDI index tell us? If we look at overall scores, the US ranked first in 2010 data, with a value of 0.60 (on a scale from 0 to 1), followed by Australia, Sweden, Canada, and Switzerland. In other words, even the US NSE operates at 60% efficiency relative to its theoretical maximum.

The value of the GEDI index is in contextualisation, systemic approach, and its ability to allow pairwise comparisons between countries. But GEDI is not prescriptive – it is only a starting point. The GEDI analysis needs to be combined with a detailed analysis of a country’s policies and framework conditions to provide the necessary context for the GEDI portrayal.

But, combined with policy portfolio simulations, it could provide an interesting platform for debating policy priorities - and a given country’s strengths and weaknesses. Combined with deeper analysis of country-specific factors, GEDI could help policy-makers look beyond their own policy silos and start considering priorities and trade-offs between alternative policy scenarios. This way, GEDI could help pave the way towards a more systemic approach to entrepreneurship policy.

Erkko Autio is professor in technology venturing and entrepreneurship at Imperial College Business School, London