CAIRN-INT.INFO : International Edition


  • Start-up support programmes have no impact on a country’s entrepreneurship rate.
  • Start-up support programmes contribute to raising the necessity entrepreneurship rate but do not affect the opportunity entrepreneurship rate.
  • The level of institutional protection offered to the unemployed reduces the positive effect of business start-up support programmes.
  • Public financial resources dedicated to entrepreneurship start-up programmes do not reduce unemployment.

1For several decades now entrepreneurship has been on the political agenda due to the belief that it constitutes a key tool in fighting and reducing unemployment. As a result, and with the help of state funding, many business start-up structures and innumerable programmes have gradually sprung up all around the world, including in OECD countries and France.

2These start-up support programmes continue to develop, but one question remains unanswered: what real effects do they have on entrepreneurship activity and unemployment levels? A number of researchers have attempted to answer this question but it remains open to debate given the lack of convergence of the studies carried out to date. On the one hand, studies have shown the positive effects of start-up programmes from a microeconomic point of view [1]. In general, these studies reveal that support programmes lead to higher levels of new businesses, with higher than average longevity. On the other hand, a number of researchers have not found a significant effect and argue that public money could be made better use of. [2] These researchers criticise the microeconomic approach to the question and highlight the fact that most entrepreneurs do not create additional jobs during the first few years of a company’s life. What’s more, newly created businesses don’t necessarily innovate, and around half of them will fail within the first five years. [3] Faced with this controversial issue, we think that a macroeconomic approach would be useful, both in order to clarify the issue and to better inform public policy relating to entrepreneurship.

3With this goal in mind, we looked at the Global Entrepreneurship Monitor (GEM) data for the period 2002-2013. Every year, GEM runs a survey on entrepreneurial activity in just under 100 countries, for which they interview at least 2000 people in each country in order to establish representative samples [4]. In the study results presented here, we concentrated on the OECD countries and added macroeconomic data from the OECD and World Bank [5] to the GEM data. We carried out several econometric analyses [6] going back 10 years for 25 OECD countries. Below, we present the results of these analyses on the effects of start-up support programmes and the moderating role of the institutional framework, [7] in terms of quantity and quality, on entrepreneurial activity within a country, and on unemployment reduction. These results probably appear counter-intuitive, but as well as being based on solid methodology with a longitudinal design, they do make sense and they lead to very clear practical implications that we think could be usefully taken into account in future consideration of public policy on entrepreneurship and unemployment reduction.

No impact on a country’s entrepreneurship rate

4This might be seen as bad news and a controversial result, so we need to be very clear. The entrepreneurial activity rate measured by GEM is defined as the percentage of the adult population of a country that is involved with a developing business (i.e. a recently-created business that has not yet paid its owner a salary three months after creation) or in a young business (older than a new start-up, where the owner is regularly paid a salary and which has been running for no more than 42 months). Our results show that the entrepreneurship rate is not affected in a significant way by the amount of public expenditure allocated to business start-up support programmes. [8]

5What do we mean by the expression “is not affected in a significant way”? Specifically, using a Bayesian method of estimation, we found that the amount of public spending allocated to business start-up support (as a percentage of GDP) only appears to be a key variable in estimating the rate of entrepreneurship in less than 30% of all explanatory models. According to our analyses, the main factor that plays a significant role in determining a country’s rate of entrepreneurship is its level of economic growth as measured by the GDP growth rate, [9] present in 93% of all explanatory models.

6This result must be qualified in three ways. First, even if it is only present in less than 30% of the explanatory models, of the variables that we analysed, the amount allocated to business start-up support programmes remains the third most important factor for predicting the entrepreneurial activity of a country (right behind the amount allocated to research and development, [10] which represents, as a percentage of GDP, 31% of all models). Second, this result partly depends on the method used. We observed first-hand a significant result in a fixed-effects model and a non-significant result in a multilevel model. [11] However, the Bayesien approach remains the most relevant for dealing with the question because it does not impose a previous model on the data. [12] Finally, and most interestingly, our results suggest that the amount allocated to business start-up support programmes has a different impact on distinct entrepreneurial modes.

Impact on the necessity entrepreneurship rate but not on the opportunity entrepreneurship rate

7GEM distinguishes between two types of motivation for entrepreneurship. Necessity entrepreneurship is motivated by a lack of economic alternatives, particularly in terms of job opportunities. In this case, the individual becomes an entrepreneur to survive, to create a job for him or herself because there are no other solutions. Conversely, opportunity entrepreneurship is motivated by the pursuit of a business opportunity perceived as interesting, promising and feasible. In this case, the individual is not obliged to choose entrepreneurship, but pursues an opportunity for value-creation. [13] The academic literature recognises these two types of entrepreneurship and a number of researchers recommend concentrating on public policy that favours opportunity entrepreneurship.

8Yet, our study shows that the levels of public spending allocated to start-up support programmes is not really a determining factor for the opportunity entrepreneurship rate. It is only the fourth factor on the list, behind GDP growth, the intensity of the country’s international commercial trade, and the amount invested in R&D. Public funds allocated to start-up incentives are only present in less than 14% of all models for the opportunity entrepreneurship rate in the 25 countries of the OECD that we analysed.

9However, this amount is the second most influential factor for the necessity entrepreneurship rate in these countries, being present in around 70% of all explanatory models. The first factor remains the unemployment rate.

10This result makes sense, because most public start-up support policies target unemployment as a priority. The main objective (sometimes not stated) is to incite the unemployed to become entrepreneurs in order to reduce the unemployment rate. But even though motivation for choosing entrepreneurship varies amongst the unemployed and some find real business start-up opportunities, most choose entrepreneurship by necessity.

The effect of the protection offered to the unemployed

The level of protection reduces the positive effect of business start-up programmes

11We studied the moderating effect that the institutional environment can have on the effectiveness of public policy relating to entrepreneurship. Specifically, we examined the moderating effect of unemployment benefits on the relationship between public spending on support programmes and the different rates of entrepreneurship. The results show that there is a negative moderating effect on the rate of necessity entrepreneurship. This means that, the more generous the assistance offered to unemployed people, the more the positive effect of start-up support programmes on necessity entrepreneurship falls. This effect is negligible in relation to opportunity entrepreneurship, which leads to the same negligible result on the overall rate of entrepreneurship in a country.

12Once again, these results make sense. Unemployment benefits and the different allocations for the unemployed reduce the need to create an income-generating activity immediately and allow unemployed people to spend more time looking for suitable employment. A welfare state thus allows its citizens to better manage career accidents without pushing individuals towards necessity entrepreneurship (at least in the short-term). However, the question of overall coherence arises where a state spends large sums both on start-up support for the unemployed and social protection for those same unemployed. Value judgements aside, from a macroeconomic point of view, states operating such policies are acting as both an accelerator and a brake on necessity entrepreneurship.

Does public spending through entrepreneurship start-up programmes reduce unemployment?

13The answer to this question is, unfortunately, no. Public spending on business start-up programmes does not bring the unemployment rate down, either directly or indirectly. We found that the amount spent on start-up support programmes had no direct effect on unemployment in the countries studied. Neither did we find an indirect effect through necessity entrepreneurship.

14Our results show that the main macroeconomic source of unemployment reduction is the GDP growth rate. This is present in 53% of all explanatory models for unemployment measured two years after measuring the opportunity entrepreneurship rate. Opportunity entrepreneurship is also the second factor contributing to a fall in unemployment one year after, but to a lesser extent (25% of all models). This indicates that the beneficial effect of opportunity entrepreneurship is built over time and becomes visible gradually as businesses created to pursue real opportunities start to recruit in order to expand.

15As for necessity entrepreneurship, there is no real effect on unemployment. It features as the eighth factor on the list for unemployment after one year and the tenth factor after two years, present in 12% of all explanatory models but with coefficients very close to zero.

16In conclusion, despite the more or less explicit aim of unemployment reduction, public spending allocated to business start-up support programmes is inefficient since it does not encourage the type of entrepreneurship that contributes to lowering unemployment. The funds spent on trying to incite and support unemployed people to become entrepreneurs only increase the necessity entrepreneurship rate, which creates very few jobs. The type of entrepreneurship most likely to create jobs is opportunity entrepreneurship, whose rate is not affected by public subsidies (at least not according to our study).

Limitations and practical implications

17The results that we present here come from a macroeconomic analysis of 25 OECD countries monitored between 2002 and 2013. One of the key limitations of this study, in common with all macroeconomic analyses, is that it is not possible to distinguish the different types of start-up support programmes in different countries. Our results are therefore generalised. Of course, it’s possible that certain support programmes boost opportunity entrepreneurship, as certain support structures (such as incubators and accelerators) claim to. Indeed, support programmes targeting the unemployed can have a beneficial effect on a microeconomic level, for example by making an unemployed person the head of a company. However, our results suggest that such cases remain anecdotal, since the overall entrepreneurship rate as well as the rate of opportunity entrepreneurship do not change significantly from one country to another and from one year to another according to the level of investment in start-up support programmes. As for the necessity entrepreneurship rate, it is affected by these programmes but does not affect the unemployment rate.

18These results call into question the role of the state in encouraging entrepreneurship and show the need for real overall coherence in establishing public policy relating to jobs and entrepreneurship. Specifically, they suggest two practical implications that should be explored further. Firstly, we note that most countries that have a generous policy of social protection invest heavily in start-up support programmes, whereas countries with a flexible employment market allocate relatively low levels of public resources to this type of programme. For example, in our study, France and Spain are the countries with the highest levels of assistance for business start-ups and the most generous policies in terms of social and employment protection. By contrast, England and the United States have the least protective institutions in terms of jobs and the lowest levels of public investment in the promotion of entrepreneurship. On the basis of our results, a first practical implication is to suggest that the countries with a flexible employment market (like the United States) invest more in promoting entrepreneurship, particularly in order to re-start the economy in post-crisis periods, because it is in these countries that a greater effect in terms of overall entrepreneurship rates might be seen. By contrast, countries with a rigid institutional environment in terms of the employment market (such as France) should either rethink the configuration of the assistance offered to the unemployed in order to incentivise getting back into work more rapidly, or rethink the allocation of resources to start-up support programmes, since the current system is simply ineffective.

19Our second recommendation relates to rethinking the allocation of resources to start-up support programmes. Our results show that these programmes only augment the rate of necessity entrepreneurship and do not bring down the unemployment rate. Indeed, other studies have shown to what extent support workers were ill-equipped, poorly-trained and often working in conditions that are not conducive to supporting necessity entrepreneurs. [14] Given this observation, there is a real temptation to suggest a reallocation of public resources towards the development of opportunity entrepreneurship, which does lead to job creation. But alas, the equation is not so easy to solve. Firstly, detecting opportunity entrepreneurs is largely done through self-reporting and does not guarantee the success of a new business. In fact, the success of a new business is very difficult to predict (even the performance of risk management professionals look more dependent on luck than skill). In addition, offering support can itself be a source of bias and diversions [15], which can increase the risk of supporting a bad project as well as refusing a project with strong potential. Finally, inciting entrepreneurs motivated by the pursuit of an opportunity to create a company might simply be unnecessary, since these individuals are naturally inclined to do so even without assistance.

20Given all these observations, our fellow citizens might quite rightly ask if public resources allocated to start-up support programmes wouldn’t be better spent in hospitals and schools, in order to ensure access to healthcare and education for everyone. Not that entrepreneurship is just a gimmick, quite the opposite! But it’s quite obvious that a well-educated and healthy population will be more likely to choose entrepreneurship and with greater likelihood of success. Indeed, the desire to create and pursue opportunities might be instilled more easily and effectively through education, particularly education that rewards effort and initiative rather than just punishing mistakes. And a good health system can always help to maintain the health of entrepreneurs, which constitutes a major factor in the longevity and development of small and medium-sized businesses.

21The good news is that the pursuit of opportunities and job creation can be encouraged by certain measures that are less costly and certainly more efficient than most start-up support programmes. Again, we would recommend a focus on education and health as part of a more holistic approach. In terms of more specific ideas, we would suggest not encouraging the promotion of new start-ups but encouraging instead the development and growth of new businesses, as well as small structures by, for example, lowering taxes on micro-entrepreneurs (the self-employed and freelancers who are the most vulnerable in terms of education and health) and facilitating hiring in small and medium-sized businesses.


  • [1]
    For example, the following studies: Baumgartner, H., and M. Caliendo. 2008. “Turning Unemployment into Self-Employment: Effectiveness of Two Start-Up Programmes.” Oxford Bulletin of Economics and Statistics 70 (3): 347-373. Caliendo, M., and A. Kritikos. 2010. “Start-ups by the unemployed: characteristics, survival and direct employment effects.” Small Business Economics 35 (1): 71-92 (July). Røed, K., and J. F Skogstrøm. 2014. “Unemployment Insurance and Entrepreneurship.” LABOUR 28 (4): 430-448 (December).
  • [2]
    See, for example: Acs, Z. J., Åstebro, T., Audretsch, D. B., & Robinson, D. T. (2016). Public policy to promote entrepreneurship: a call to arms. Small Business Economics, 47(1), 35-51. Åstebro, T. (2017). The private financial gains to entrepreneurship: Is it a good use of public money to encourage individuals to become entrepreneurs? Small Business Economics, 48(2), 323-329. Parker, S. C. 2007. “Policymakers Beware!” In D. B. Audretsch, I. Grilo, & A. R. Thurik (Eds.), Handbook of research on entrepreneurship policy (pp. 54-63). Cheltenham: Edward Elgar. Shane, S. A. 2009. “Why encouraging more people to become entrepreneurs is bad public policy.” Small business economics 33 (2): 141-149.
  • [3]
    The statistics diverge according to the country and the study. In the studies we looked at, the failure rate varies between 30% and 70% in the first five years.
  • [4]
    GEM reports are available on the site In France, the GEM survey is coordinated by the Entrepreneurship Research Center at emlyon business school.
  • [5]
    The complete study can be found online: Laffineur, C., Dubard Barbosa, S., Fayolle, A., & Nziali, E. (2017). Active labor market programs’ effects on entrepreneurship and unemployment. Small Business Economics, doi:10.1007/s11187-017-9857-7.
  • [6]
    Specifically, we used the Bayesian model averaging estimator in a country fixed effect linear regression context in order to control for model uncertainty.
  • [7]
    More precisely, we took into account the level of social protection through unemployment benefits, social assistance, tax reductions and various types of help available to the unemployed.
  • [8]
    In order to measure the amount of public expenditure allocated to support start-ups we used a variable provided by the OECD which specifically measures active labor market programs targeted at entrepreneurship. Precisely, it is a measure of national expenditure on start-up incentives as a percentage of GDP, i.e., programs encouraging unemployed workers and targeted groups to start their own businesses or to become self-employed.
  • [9]
    Annual GDP growth rate from the World Bank Indicators.
  • [10]
    As a proxy of technological progress, we used a variable measuring countries’ research and development expenditure as a percentage of GDP, derived from the World Bank.
  • [11]
    We run a fixed effect model without controlling for model uncertainty in order to check for consistency with previous research.
  • [12]
    Therefore, BMA controls for model uncertainty, whereas other commonly used methods do not.
  • [13]
    The distinction between opportunity and necessity motives to start a business is identified in GEM with the question “Are you involved in this start-up to take advantage of a business opportunity or because you have no better choices for work?”.
  • [14]
    See, for example: Nakara, W. A., & Fayolle, A. (2012). Les “bad” pratiques d’accompagnement à la création d’entreprise : Le cas des entrepreneurs par nécessité. Revue Française de Gestion, 228–229, 231–251.
  • [15]
    See, for example: Dubard Barbosa, S. & Duquenne, L. (2016). Les dérives des systèmes d’accompagnement sur la prise de décision et de risque dans la création d’entreprise : réflexions pour la recherche et pour la pratique. Revue Internationale PME, 29(3-4): 193-239.

In this article, based on the Global Entrepreneurship Monitor data, we carried out several econometric analyses going back ten years for 25 OECD countries. We present the results of these analyses on the effects of start-up support programmes and the moderating role of the institutional framework, in terms of quantity and quality, on entrepreneurial activity within a country, and on unemployment reduction. Our results probably appear counter-intuitive, but as well as being based on solid methodology with a longitudinal design, they do make sense and they lead to very clear practical implications that we think could be usefully taken into account in future consideration of public policy on entrepreneurship and unemployment reduction.

Saulo Dubard Barbosa
Saulo Dubard Barbosa is Associate Professor at EMLYON Business School, where he currently teaches new venture creation and social entrepreneurship at different levels. His main research interests are on decision making under uncertainty, venture creation, and social change.
Catherine Laffineur
Catherine Laffineur is Assistant Professor at the Department of Economics at the University of Nice Sophia-Antipolis and a research fellow at EMLYON Business School. Her main research interests include labor market dynamics and the economics of entrepreneurship.
Alain Fayolle
Alain Fayolle is Professor of Entrepreneurship and Director of the Entrepreneurship Research Center at EMLYON Business School. His main research interests include the entrepreneurial process of new venture creation, organizational entrepreneurship, and entrepreneurship education.
This is the latest publication of the author on cairn.
Uploaded on on 15/03/2018
Distribution électronique pour De Boeck Supérieur © De Boeck Supérieur. Tous droits réservés pour tous pays. Il est interdit, sauf accord préalable et écrit de l’éditeur, de reproduire (notamment par photocopie) partiellement ou totalement le présent article, de le stocker dans une banque de données ou de le communiquer au public sous quelque forme et de quelque manière que ce soit.
Loading... Please wait