Artículo en PDF
How to cite
Complete issue
More information about this article
Journal's homepage in redalyc.org
Sistema de Información Científica
Red de Revistas Científicas de América Latina y el Caribe, España y Portugal
© Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Section
A
RT I C L E S
Influence of the economic cycle on the determinants
of nascent entrepreneurial activity. An empirical
analysis of the Spanish case
Jesús Martínez Mateo
*, Ignacio Mira Solves
*, José M.ª Gómez Gras
*
ABSTRACT:
This paper explores the contribution of a selection of elements rep-
resentative of human capital and perception as determinants of entrepreneurship in
different stages of the economic cycle. The results confirm the significance of self-
efficacy, the perception of opportunities, and the fear of failure, and highlight the
importance of personal knowledge of entrepreneurs. They remain influential in dif-
ferent economic times in which their analyses have been replicated, although some
differences are felt that point to, in contraction periods, a loss of influence of the
confidence in one’s own abilities, compared to an increase in the case of judgment
on the existence of opportunities in the environment, and in the case of the pres-
ence of entrepreneurs in the surrounding context. In contrast, the behavior of the
fear of failure, as a barrier to entrepreneurship, remains unchanged in an adverse
context with respect to a positive context due to reduced opportunity costs.
JEL Classification:
E32; G01; L26; M13.
Keywords:
GEM; determinants; entrepreneurship; nascent entrepreneurs; percep-
tions; environment; economic cycle.
Influencia del ciclo económico sobre los determinantes de la actividad
emprendedora naciente. Un análisis empírico del caso español
RESUMEN:
Este trabajo explora la contribución de una selección de elemen-
tos representativos de capital humano y de percepción como determinantes de la
creación de empresas ante distintas etapas del ciclo económico. Los resultados
confirman la significación de la autoeficacia, la percepción de oportunidades y el
miedo al fracaso, y resaltan la importancia del conocimiento de emprendedores.
Su influencia se mantiene en los distintos momentos económicos en los que se han
19
Received: 15 may 2012 / Accepted: 25 june 2013.
* Miguel Hernández University of Elche. GEM Team Valencian Community.
jmateo@umh.es
The authors are grateful to the Miguel Hernández University of Elche for support given in the course
of this research.
The authors are grateful to the Spanish Ministry of Science and Innovation for the support given in
the course of the research undertaken within the framework of Research Project 21276 ECO2010.
20
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
replicado los análisis, si bien se intuyen algunas diferencias que apuntan, en etapas
contractivas, a una pérdida de influencia de la confianza en las propias habilidades,
frente a un incremento en el caso del juicio sobre la existencia de oportunidades en
el entorno y en el caso de la presencia de emprendedores en el contexto cercano.
En cambio, el comportamiento del miedo a fracasar, como barrera para emprender,
se mantiene invariable en un contexto adverso respecto a uno positivo debido a la
reducción de costes de oportunidad.
Clasificación JEL:
E32; G01; L26; M13.
Palabras clave:
GEM; determinantes; creación de empresas; emprendedores na-
cientes; percepciones; entorno; ciclo económico.
1.
Introduction
The GEM Project in Spain has been compiling entrepreneurial data in that coun-
try for more than ten years. This has provided for a rich database of information about
variables related to entrepreneurial activity during different economic times.
This potential of GEM facilitates, among others, the analysis over time of the
ability of different elements to influence that are assumed in the literature to be de-
terminants in the creation and start up of businesses. In particular, if we consider
the evolution of the economy in recent years, which has progressively shifted from
a period of growth to one of contraction and crisis, we feel it is important to inquire
about possible differences in the influence that recognized determinants of entrepre-
neurship may wield in different economic environments.
In this sense, this paper’s purpose is mainly empirical, focused on analyzing the
capacity of some elements of human capital and perception to influence nascent en-
trepreneurial activity, in addition to their evolution throughout the last seven years,
identifying possible differences at different times of the economic cycle.
Its objective, therefore, is to analyze whether different growth and recessionary
environments condition the behavior of the determinants of entrepreneurial activity.
This way, this paper’s contribution derives from the use of an extensive temporal
comparison of influential elements in the individual decision to start a business, with
the conviction that studying entrepreneurial activity within a territory, under objec-
tively different stages of the economic cycle, may contribute to improve the under-
standing of the determinants of entrepreneurship.
2.
Entrepreneurial activity and the economic environment
There is broad consensus on the positive role that entrepreneurship plays for ter-
ritorial development (Acs & Audretsch, 2003; Wagner & Sternberg, 2004; Reynolds
et al.
, 2005; Mueller, 2006; Minniti & Lévesque, 2008), and several studies have
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
21
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
shown its positive effects in terms of job creation, economic growth, and innovation
(
e. g.,
van Praag & Versloot, 2007; Acs
et al.
, 2008).
This relationship does not only move in one direction, as the set of conditions
that form the setting, and particularly, those that lead to the economic environment,
in turn have a considerable influence on the rate of entrepreneurial activity in the
territory (Reynolds
et al.
, 1994; Carree & Thurik, 2003; Bergmann & Sternberg,
2007).
Thus, rates of entrepreneurial activity may differ considerably between different
territories and between different periods, due to the peculiarities of their environ-
ments (Verheul
et al.
, 2002), and the dynamics of entrepreneurship may be very dif-
ferent depending upon the institutional context and level of development (Acs
et al.
,
2008).
Several empirical studies show that these different entrepreneurship rates be-
tween regions are affected by economic, cultural, and institutional components, while
inter-temporal differences within the same territory are dominated by influences from
within their own economic environment (Wennekers
et al.
, 2002; Freytag & Thurik,
2007). That is, between different regions, different entry rates of the entrepreneurial
process may largely be explained by their structural characteristics (Naudé
et al.
,
2008), while from an evolutionary or temporal point of view, within a given territory,
the context shaping the economic environment would be that primarily influential on
the dynamics of entrepreneurial activity (Acs
et al.
, 2008).
The current crisis is bringing change to environmental conditions, which not
only affects existing businesses, but additionally the possibilities of new business
creation and entrepreneurship (Naudé & MacGee, 2009; Gries & Naudé, 2010). In
particular, the last seven years of evolution in the Spanish economy (table 1) have
been characterized by a first stage with some growth until reaching, at the start of
2008, a turning point caused by the international crisis and the peculiarities of the
national situation. After 2008, a series of periods characterized by stagnating and
declining GDP and sustained unemployment rate increases followed one after the
other (figure 1). All this portrays two different stages in the economic cycle: one
expansive stage or that of growth until the end of 2007, with maximum peaks, then
followed by a second recessionary or contracting stage, one that we find ourselves
in yet today.
A progressively worsening situation like that shown by the GDP data is a reflec-
tion of a decline in economic activity, which, regarding entrepreneurship, directly
translates into a reduced need or demand for new businesses, in addition to indirectly
acting by affecting people in their confidence in the expectations when evaluating or
considering putting a business initiative into motion. Naudé & MacGee (2009) argue,
in this sense, that the recession and slowing growth in developed economies reduce
opportunities, causing businesses to fail and fewer new initiatives to be launched, but
the full effect on self-employment may be ambiguous due to reduced opportunity
costs and reduced competition, which, on the other hand, can also facilitate access to
business activity.
22
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Regarding unemployment, the rates shown also clearly indicate the change in the
cycle. Its turning point also occurred in 2008, and the unemployment rate in 2011
was almost triple that of four years earlier. The effects of unemployment upon entre-
preneurship can also be contradictory, from both an individual as well as a territorial
perspective (Bergmann & Sternberg, 2007). From the first point of view, the pressure
of self-employment may be greater in those out of work than in those employed, but
Table 1.
Economic situation and environmental confidence indicators
Indicator
1
2005
2006
2007
2008
2009
2010
2011
GDP variation
2
Quarterly
0.9%
1.0%
0.8%
0%
–1.1%
0.3%
0.2%
Annual
3.7%
4%
3.7%
1.9%
–4.4%
–0.0%
0.7%
Unemployment rate
9.33%
8.53%
7.95%
10.44%
17.92%
20.09%
20.89%
Nascent
entrepreneurial
activity
Registered
rate
2.4%
3.0%
3.5%
3.3%
2.3%
2.2%
3.3%
Annual
variation
+14.3% +25.0% +16.7%
–5.7%
–30.3%
–4.4%
+50.0%
Consumer Confidence Index
(CCI)
Values from 0 to 200. Neutral
value: 100
91.2
84.9
93.4
57.3
64.0
65.9
74.9
Business Confidence Index
(BCI)
Values from –100 to +100
+7.2
+9.5
+9.0
–12.6
–19.0
–14.8
–9.2
INDSUP
(Individual
perception to
entrepreneurship
index)
Values from 0 to 3
Average
1.07
1.09
1.04
0.87
0.97
0.83
Mode
1
1
1
0
1
0
0/3
33.8%
32.5%
33.5%
39.9%
34.6%
40.8%
1/3
33.8%
34.8%
36.1%
37.3%
38.4%
38.4%
2/3
23.8%
23.8%
23.5%
19.0%
21.8%
18.0%
3/3
8.6%
9.0%
6.9%
3.8%
5.2%
2.9%
CULSUP
(Cultural
support for
entrepreneurship
index)
Values from 0 to 3
Average
1.72
1.77
1.79
1.71
1.52
1.69
1.81
Mode
2
2
2
2
2
2
2
0/3
11.0%
12.9%
10.5%
11.9%
17.5%
12.9%
9.7%
1/3
13.6%
24.3%
25.9%
28.3%
31.7%
27.8%
26.9%
2/3
19.0%
36.3%
37.8%
36.9%
32.3%
36.1%
35.8%
3/3
11.0%
26.6%
25.8%
22.9%
18.6%
23.2%
27.5%
1
The indicators on GDP, unemployment, CCI, and BCI are those registered in the second quarter of the years indicated
so that they coincided in time with the dates the GEM APS survey was taken.
2
Gross domestic product (GDP). Chained volume with the year 2000 as reference. Data corrected for seasonal and
calendar effects. Units: rates.
Sources
: GDP and unemployment rate, INE; CCI, Instituto de Crédito Oficial; BCI, Cámaras de Comercio – Servicio
de Estudios; INDSUP, CULSUP, nascent entrepreneurial activity, GEM – Adult Population Survey (APS) Spain, 2005
to 2011.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
23
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
often they do not possess the necessary resources and skills. On a macro level, higher
unemployment leads more to utilize self-employment as a way out; but then there is
also less purchasing power on behalf of the population, and therefore, less demand,
which in aggregate have a negative effect on the number of start-ups.
Within a framework like that described, the creation of businesses as an inte-
gral part of the economic reality has not been immune to this situation. Successive
drops in the numbers of start-ups can be noted equally beginning in 2008 (GEM
measures this nascent activity) until 2011, a year that despite bad economic data,
such activity increased, basically due to the reduction in opportunity costs prompt-
ed by the deteriorating starting situation for many new entrepreneurs (table 1 and
figures 2a & 2b).
In particular, the current economic crisis is an extreme situation, which like
other extreme events related to natural disasters or manmade conflicts whether civil,
military, or economic (Naude, 2010), eventually affect growth, development, and
levels of uncertainty in the environment, which can influence people psychologi-
cally, affecting their cognitive processes of forming expectations and perceptions.
In fact, these intangible psychological effects may become more important than the
direct consequences that are visible or material (Brück
et al.
, 2010). In this sense,
research like that by Marcu
et al.
(2012) presents an interesting process about how
the influence of psychological factors on entrepreneurial tendencies can be seen af-
fected in crisis environments, specifically using the internal locus of control as an
example.
In this regard, specifically in Spain, various indicators (table 1) show how con-
sumer and business confidence have suffered significantly with the changing cycle.
Figure 1.
Annual growth of GDP and unemployment rate, 2005-2011
Unemployment rate
Annual growth rate of GDP
25
22.5
20
17.5
15
12.5
10
7.5
5
2.5
0
–2.5
–5
–7.5
2005
2006
2007
2008
2009
2010
2011
Years
(own elaboration)
Source:
Own elaboration.
24
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Thus, the Consumer Confidence Index (CCI) shows that a situation close to neutral
has shifted to an environment in which the public’s perception regarding economic
activity has deteriorated significantly. Additionally, the Business Confidence Index
Figures 2a and 2b.
Nascent activity registered in Spain and annual variation,
2005-2011
Nascent activity
5
4.5
4
3.5
3
2.5
2
1.5
1
0,5
0
2005
2006
2007
2008
2009
2010
2011
Years
Registered nascent activity (%)
Annual growth rate in
nascent activity
60
50
40
30
20
10
0
–10
–20
–30
–40
2005
2006
2007
2008
2009
2010
2011
Years
(own elaboration)
Annual growth rate in nascent activity (%)
Source:
Own elaboration.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
25
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
(BCI) illustrates a scenario in which the perception of entrepreneurs concerning the
situation has shifted from positive to negative, and this may influence their business
behavior and the development of new projects.
Table 1 also contains two indices developed by GEM from the Adult Population
Survey (APS) that try to bring together some cultural aspects used in studies that
link culture with entrepreneurial behavior. In terms of the synthesis by Freytag &
Thurik (2007), the INDSUP (Individual perception to entrepreneurship index) would
be related to a series of added individual psychological features, in such a way that a
higher proportion of persons possessing entrepreneurial values could lead to a higher
proportion of entrepreneurs within a society. In turn, the CULSUP (Cultural support
for entrepreneurship index) would be related to the degree of legitimization or moral
approval —social norms— of entrepreneurship within a culture, in the sense of great-
er respect for the tasks of entrepreneurs, presence in the media and educational sys-
tems, etc., which could lead to an increased supply and demand of entrepreneurs. The
comparison shows some reduction in the mean values of the individual component,
modifying the distribution in the percentages of responses by the population towards
0 and 1 between the expansionary phase and first years of the contracting phase. This
is in line with that stated in preceding paragraphs with respect to the psychological
factors of individuals. The cultural component related most to social norms does not
vary, which may be consubstantial to the fact that the comparison is made within the
same territory of reference. Therefore, it can be assumed that no significant change
has occurred either in the components of the socio-cultural environment or in the
institutional framework, but rather, the influences on entrepreneurship deriving from
the change in the environment would obey their economic component (Wennekers
et al.
, 2002; Freytag & Thurik, 2007).
In short, the cited indicators bring us to the investigated matter and reflect two
distinct environments in the economic cycle, and the data suggest that the judgment
about the economic climate made by individuals may have deteriorated, thereby also
affecting the formation of perceptions related to entrepreneurial activity and their ef-
fects on business involvement
1
. Similarly, other aspects related to human and social
capital may have seen their influence wane with the considerable change in the eco-
nomic scenario of recent years.
3.
Nascent entrepreneurs and determinants of business
involvement
Generally, much research on entrepreneurship
is carried out retrospectively, only
including
business survivors
years after their creation. This carries the risk of intro-
ducing bias, like capturing characteristics and influences related more with business
survival than with the decision to start a business, or incorporating mistakes in the
1
Within the context of this paper, we understand the term
entrepreneurial involvement
as referring
to the start-up and development of nascent entrepreneurial activity.
26
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
information due to memory loss or reinterpretations of facts due to the passage of
time and transpiring events (Delmar & Davidsson, 2000; Davidsson & Honig, 2003).
Furthermore, not incorporating information about individuals who failed in the proc-
ess causes the loss of valuable information about the characteristics, attitudes, and
circumstances that led them to try (Delmar & Davidsson, 2000; Gartner
et al.
, 2004;
Johnson
et al.
, 2006).
All this recommends directing the research about the determinants towards what
are called the early stages of the entrepreneurial process. In this regard, research
focusing on these initial phases usually revolves around models of entrepreneurial
intentions and nascent entrepreneurs (Autio
et al.
, 2001; Davidsson & Honig, 2003).
However, these authors, along with Delmar & Davidsson (2000) and Krueger (2003),
warn that the use of intentions exclusively is not without risk either, due to the danger
of not distinguishing between
dreamers and doers
.
By keeping these aspects in mind, we consider it appropriate to focus this paper
on nascent entrepreneurs, individuals who are taking steps to found businesses of
their own, but who have yet to successfully finish this step of the process (Carter
et al.
, 1996), dealing with subjects who «start to commit time and resources to found-
ing a new firm» (Reynolds & White, 1997; Reynolds, 2000).
About these individuals, several studies have analyzed the influence of elements
of human and social capital and individual perceptions:
3.1.
Influences of elements of human and social capital
The elements of human and social capital refer to the resources of individuals.
They come in the form of educational baggage, experiences, and accumulated skills,
in addition to networks of contacts, family history and, in general, close role models
upon whom to focus, who exert their influence and provide vicarious experience.
As for educational levels, several authors point to an uncertain relationship in
general (Greene, 2000; Blanchflower, 2004) due to their value by affording improve-
ments in the capacity for self-employment, but also for employment by others (Crosa
et al.
, 2002). On the other hand, Shane (2003) provides a varied relationship of jobs
where the educational level correlates positively with business involvement, justify-
ing this relationship on the basis that the educational component increases the stock
of skills and information that are influential in the exploitation of opportunities, and
subtracts uncertainly in the assessment of the expected returns from the entrepre-
neurial activity. Likewise, within the GEM research context, various studies have
also found positive effects on the probability of being a nascent entrepreneur (
e. g.,
Wagner & Sternberg, 2004; Arenius & De Clercq, 2005; Mueller, 2006).
On the other hand, personal knowledge of other entrepreneurs within the inner
circle is the object of study in relation to entrepreneurial activity, mainly in terms of
the social capital component (relationships or networks of entrepreneurs) and their
positive influence as role models. In this sense, it provides a human capital compo-
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
27
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
nent, generating vicarious learning about exploiting opportunities through observing
the behavior of others (Storey, 1994; Reynolds, 1997; Shane, 2003). Many studies
have found positive effects on nascent entrepreneurship that derive from the presence
of entrepreneurs within the family. Examples of these include Delmar & Davidsson,
2000; Davidsson & Honig, 2003; De Clercq & Arenius, 2003; Wagner, 2004; Wagner
& Sternberg, 2004; Arenius & Minniti, 2005; Mueller, 2006; and Tamásy, 2006.
3.2.
Influence of perceptual elements
The importance of perceptions for the nascent entrepreneur has been demonstrat-
ed fundamentally in the paper by Arenius & Minniti (2005), who understand them as
subjective perceptual variables, occasionally partial, coming from the psychological
and sociological literature, with importance in the decision, and that do not neces-
sarily reflect objective circumstances. These types of variables have been dealt with
in different models related to entrepreneurial activity, fundamentally in the literature
related to intentions (Shapero & Sokol, 1982; Krueger & Carsrud, 1993; Krueger &
Brazeal, 1994; Krueger, 2000 & 2003).
These models consider the perception of desirability as the degree to which the in-
dividual is attracted to a given behavior, and they tend to agree that it depends upon the
expected results of the behavior (Degeorge & Fayolle, 2005; Brännback
et al.
, 2006).
In this sense, individuals do not only perceive their own desirability towards business
behavior, but they could also consider their fear of failure, and underestimate it. With
respect to nascent activity, GEM research has analyzed this perception based on it being
able to pose a barrier, and generally, a negative influence of this fear on the propensity
to start a business was found (De Clercq & Arenius, 2003; Wagner & Sternberg, 2004;
Arenius & Minniti, 2005; Lee
et al.
, 2005; Köllinger
et al.
, 2005; and Tamásy, 2006).
Concerning
entrepreneurial
opportunities,
contributions
by
Venkataraman
(1997), Shane & Venkataraman (2000), and Eckhardt & Shane (2003) have given a
prominent role to their existence, detection, and exploitation. Similarly, models by
Gnyawali & Fogel (1994), Verheul
et al.
(2002), and GEM (Reynolds
et al.
, 2005)
have demonstrated the importance of the existence of surrounding opportunities, and
their perception by the individual, for subsequent entrepreneurial initiatives. Regard-
ing the analysis of nascent activity, this element has been frequently incorporated. In
this manner, Alsos
et al.
, 2003; De Clercq & Arenius, 2003; Arenius & Minniti, 2005;
Lee
et al.
, 2005; Köllinger
et al.
, 2005; Köllinger & Minniti, 2006; and Tamásy, 2006
find that the perception of future opportunities has a positive and significant effect on
the decision to start a business.
Perceived self-efficacy, an element highlighted by Shane (2003) as a psycho-
logical factor with influence on the aptitude for exploiting opportunities, is a variable
centered on the individual that refers to the perception of one’s capacity to execute
and perform, and has been shown to be an element with positive influence on the
generation of entrepreneurial intentions. Specifically, with regards to nascent activ-
ity, studies suggest a strong impact of self-efficacy on entrepreneurial propensity
28
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
(Diochon
et al.
, 2002; Alsos
et al.
, 2003; De Clercq & Arenius, 2003; Wagner, 2004;
Arenius & Minniti, 2005; Köllinger
et al.
, 2005; Lee
et al.
, 2005; Köllinger & Min-
niti, 2006; Tamásy, 2006), with the perception variable usually highlighted most.
Socio-cultural elements, and in particular the beliefs and attitudes of the members
of society in relation to the social desirability of entrepreneurial activities, are consid-
ered by Shane (2003) to be part of the institutional context. Within the scope of the
principal theoretical models of entrepreneurial intentions, these aspects would form
part of the so-called subjective (Ajzen, 1991) or social norms (Krueger & Carsrud,
1993; Krueger & Brazeal, 1994; Krueger, 2000; 2003) regarding the detected social
pressure with respect to behavior, with influence on the development of the inten-
tion and subsequent entrepreneurial conduct. Within the GEM context, these ques-
tions have been introduced as subjective norms (Bruyneel
et al.
, 2006), socio-cultural
norms of the institutional environment (Driga
et al.
, 2005), or approximations of so-
cial acceptance of entrepreneurial conduct and social legitimization of the employer
(Tominc & Rebernik, 2007), without finding a clear significant relationship.
Based on that previously mentioned, this paper focuses on comparing the influ-
ence of the educational level, contact with entrepreneurs, social desirability, fear of
failure, perception of opportunities, and perceived self-efficacy on nascent entrepre-
neurial activity. All of this is done within a broad timeframe that contemplates the
changing phases of the economic cycle, testing the impact capacity of these determi-
nants (figure 3).
Figure 3.
Research approach
SITUATION OF THE ECONOMIC CYCLE
ENTREPRENEURIAL
INVOLVEMENT
EDUCATIONAL LEVEL
ENTREPRENEURIAL CONTACT
(SOCIAL) DESIRABILITY
FEAR OF FAILURE
PERCEPTION OF OPPORTUNITIES
PERCEIVED SELF-EFFICACY
Source:
Own elaboration.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
29
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
4.
Methodology
Data from Adult Population Surveys (APS) conducted in Spain between 2005
and 2011 were used for the empirical work under the consideration that they provide
an appropriate reflection of two different economic climates marked by two different
stages in the economic cycle. To do this, 2008 was taken as the year of inflection,
with the three years immediately preceding it and the three following it examples of
the two different directions of the cycle.
For the set of the seven analyzed years, 154,419 sample observations were used,
whose detail per year is in table 2. The sample size for each year permits, working at
a 95% confidence level and accepting as an assumption the hypothesis of maximum
indetermination and infinite population, reaching some sampling errors for simple
estimations that all vary between
±
0.61 and
±
0.82%.
The research focused on the study of nascent entrepreneurial activity
2
, which acts
as a variable to explain. Excluded from the sample were those individuals involved
in any stage of the GEM entrepreneurial process different from this phase. The other
variables selected are indicative of the baggage of human and social capital (educa-
tional level and knowledge of or contact with entrepreneurs) and perceptual variables
(social desirability, fear of failure, opportunities, and self-efficacy). Also considered
were the sociodemographic elements of age and sex as control variables. The Annex
contains the questions, values, and classifications carried out on the population to
operationalize all the variables.
Binomial logistic regression analysis, a generalization of the classic linear re-
gression model applied to the case of categorical dichotomous variables, was selected
as multivariate technique for the analysis.
In order to compare the periods under consideration, seven regressions with an
identical approach and incorporation of variables, one for each year, were replicated.
The method for the comparison was the Wald test
3
on the significance of the dif-
ferences between the corresponding coefficients found in the different regressions,
although for illustrative purposes and simplification, the same information was col-
lected under a comparative graph of the odds ratios and their confidence intervals.
2
At GEM, individuals are classified as nascent entrepreneurs if they are carrying out activities that
lead to starting a business, of which they will be the owner, at least in part, and furthermore, no wages have
been paid for more than three months.
3
The Wald chi-square statistic has one degree of freedom. Its formula is the following:
()
()
()
ββ
ββ
12
2
1
2
2
2
+
se
se
In it, the betas are logit coefficients estimated for each particular variable in two different years,
taking the square of their difference as the numerator and the sum of the squares of the standard errors
as the denominator. The results it provides are equivalent to those that would be obtained traditionally by
the incorporation of a dummy variable of interaction that reflects the years being compared. Likewise,
the graphic comparison of the overlaps between the ends of the confidence intervals provides the same
information as to the existence of significant differences.
30
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
The data were subjected to a preliminary analysis in order to compare the con-
ditions for using logistic regression, and they were properly verified. In each case,
the sample size was superior to 10 (
k
+
1), with
k
being the number of explanatory
variables, including all the dummy variables created. There were no zero frequen-
cies in the contingency table compartments that cross the explanatory variables with
the dependent variable
4
or collinearity recorded between variables. Moreover, and
given that we are in a working scenario of «infrequent events» (King & Zeng, 2001a,
2001b; Weiss
et al.
, 2007) caused by the low appearance frequency of nascent activ-
ity in the samples used, in order to solve classification problems and avoid underes-
timating probabilities with respect to the positive state in the event of interest, the
default cutoff point was modified by collecting and analyzing the ROC curves in the
seven initial regressions, after which seven definitive regressions were reestimated.
5.
Results
5.1.
Descriptive analysis
Table 2 records, for the set of samples used, a decrease in nascent activity starting
in 2008, until the upturn that occurred in 2011.
With respect to the educational level, as an objective descriptor of individual
baggage, it registers a lower percentage of individuals at the middle level in all cases,
while the weight change between the extremes responds better to the different way of
computing this specific variable at GEM those years.
As for the remaining variables of interest, they show movements that responded
to the different economic context between the years of the expansive phase in the
cycle and those of the contractive phase, with the greatest brunt of these adjustments
occurring in 2008 and 2009.
Thus, the perception of social desirability of the activity decreased slightly in
2008, and then with greater intensity when the individuals were surveyed in 2009, the
year after the crisis was recognized. Particularly serious is the case of the perception
of good opportunities for entrepreneurship in the environment,
i. e.,
the optimism with
which the feasibility of developing an initiative is contemplated in terms of the pos-
sibilities of finding good opportunities. It began its descent in 2008, and by 2011, it
hardly represented 40% of what it had in 2007. The fear of failure as a barrier, for its
part, grew above the psychological threshold of 50% beginning in 2008. The presence
of entrepreneurs who were personally known and who had started businesses up to
two years prior starting decreasing in 2009, which is logical because the very number
of people starting businesses also started falling that year. Only the recognition of self-
efficacy remained at similar levels at all times, regardless of the phase of the cycle.
4
In 2006, a frequency of 0.1% was registered in the cell that intersects the dependent value at its val-
ue of 1 (nascent entrepreneur) with the perception of self-efficacy at its value of 0 (lack of self-efficacy),
which causes the estimation of an extraordinarily high coefficient in the logit, and the anomalous value
that we find in its odds ratio.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
31
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Overall, the indices of listed nascent entrepreneurial activity, as well as the per-
centages for the variables related to entrepreneurial activity, clearly show the worsen-
ing situation.
Table 2.
Frequencies of nascent activity and variables considered in the paper
in the starting sample
No. (count)
16,102
25,518
25,004
25,540
25,165
22,829
14,261
Variables used
2005
2006
2007
2008
2009
2010
2011
Nascent activity
1
Yes
2.7%
3.0%
3.8%
3.7%
2.1%
2.1%
3.5%
No
97.3%
97.0%
96.2%
96.3%
97.9%
97.9%
96.5%
Sex
Men
45.6%
47.8%
48.6%
48.8%
47.5%
48.4%
46.7%
Women
54.4%
52.2%
51.4%
51.2%
52.5%
51.6%
53.3%
Age
Mean
43.4
42.0
41.8
41.6
43.8
44.1
41.5
SD
12.652
12.866
12.526
12.449
12.423
12.387
12.677
Educational level
Low
57.0%
57.1%
35.8%
34.1%
42.3%
41.0%
36.4%
Middle
14.9%
16.9%
23.4%
21.5%
15.6%
14.2%
13.4%
Higher
28.1%
26.0%
40.8%
44.0%
42.1%
44.8%
49.8%
Entrepreneurial
contact
Yes
26.8%
32.4%
32.8%
36.0%
27.2%
27.1%
25.1%
No
73.2%
67.6%
67.2%
64.0%
72.8%
72.9%
74.9%
Social desirability
Yes
71.4%
70.1%
71.0%
68.0%
61.1%
65.5%
66.8%
No
28.6%
29.9%
29.0%
32.0%
38.9%
34.5%
33.2%
Fear of failure
Yes
49.7%
47.7%
49.6%
52.5%
54.1%
46.7%
53.7%
No
50.3%
52.3%
50.4%
47.5%
45.9%
53.3%
46.3%
Opportunities
Yes
35.8%
32.1%
33.1%
24.7%
15.2%
16.7%
13.6%
No
64.2%
67.9%
66.9%
75.3%
84.8%
83.3%
86.4%
Self-efficacy
Yes
41.1%
44.8%
44.3%
43.5%
43.2%
43.2%
43.1%
No
58.9%
55.2%
55.7%
56.5%
56.8%
56.8%
56.9%
1
The percentages differ from those shown in Table 1 because, in order to suitably capture the influence of the
determinants arising in the regression, excluded from the sample was any individual involved in any phase of the
entrepreneurial process different from that of nascent.
Source
: APS Spain, 2005 to 2011, nascent entrepreneurs and individuals without any entrepreneurial activity.
5.2.
Logistic regression analysis
Table 3 shows the final seven models of estimated logistic regression in order to
observe the influence of the proposed explanatory variables on the entrepreneurial
involvement throughout the analyzed period with
ceteris paribus
consideration
.
This
32
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Table 3.
Logistic regression for nascent entrepreneurial activity (2005-2011)
2005
2006
2007
2008
2009
2010
2011
Exp (
b
)
(SE)
Exp (
b
)
(SE)
Exp (
b
)
(SE)
Exp (
b
)
(SE)
Exp (
b
)
(SE)
Exp (
b
)
(SE)
Exp (
b
)
(SE)
AGE
**1.069
(0.034)
***1.086
(0.026)
*1.045
(0.023)
**1.047
(0.022)
***1.120
(0.032)
***1.183
(0.034)
***1.155
(0.034)
AGE SQUARED
**0.999
(0.000)
***0.999
(0.000)
**0.999
(0.000)
**0.999
(0.000)
***0.998
(0.000)
***0.998
(0.000)
***0.998
(0.000)
GENDER (MALE)
**1.306
(0.115)
***1.385
(0.091)
***1.439
(0.085)
^1.103
(0.079)
***1.479
(0.105)
***1.441
(0.108)
**1.239
(0.107)
EDUCATION
^
*
^
^
^
^
*
Education (low/higher)
^0.834
(0.127)
^1.052
(0.101)
^1.072
(0.101)
**0.816
(0.098)
**0.781
(0.123)
^0.879
(0.121)
^0.851
(0.132)
Education (middle/higher)
^0.884
(0.168)
*0.803
(0.127)
^0.967
(0.103)
^0.901
(0.104)
^0.969
(0.137)
^0.908
(0.153)
^1.265
(0.153)
ENTREPRENEURIAL
CONTACT
***1.950
(0.117)
***1.532
(0.090)
***1.761
(0.085)
***1.754
(0.082)
***2.265
(0.103)
***2.344
(0.106)
***2.726
(0.107)
SOCIAL DESIRABILITY
**0.766
(0.121)
**0.825
(0.094)
***0.763
(0.086)
**0.822
(0.084)
*0.832
(0.102)
*0.819
(0.105)
^0.980
(0.110)
FEAR OF FAILURE
***0.497
(0.130)
***0.591
(0.096)
***0.523
(0.089)
***0.520
(0.085)
***0.412
(0.111)
***0.538
(0.113)
***0.541
(0.111)
OPPORTUNITIES
***1.921
(0.117)
***1.596
(0.089)
***1.679
(0.083)
***2.296
(0.081)
***2.331
(0.105)
***2.127
(0.108)
***2.724
(0.116)
SELF-EFFICACY
***6.288
(0.170)
***51.871
(0.321)
***11.382
(0.150)
***9.620
(0.132)
***11.222
(0.193)
***8.887
(0.178)
***6.059
(0.151)
CONSTANT
***0.003
(0.707)
***0.000
(0.592)
***0.004
(0.484)
***0.004
(0.465)
***0.001
(0.649)
***0.000
(0.698)
***0.001
(0.658)
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
33
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Hosmer-Lemeshow
goodness of fit test
X
2
5.559
(Sig. 0.697)
X
2
4.807
(Sig. 0.778)
X
2
8.301
(Sig. 0.405)
X
2
2.417
(Sig. 0.966)
X
2
9.315
(Sig. 0.316)
X
2
11.992
(Sig. 0.152)
X
2
10.171
(Sig. 0.052)
Concordance: Area under
ROC curve (C-statistic)
0.810
0.833
0.811
0.813
0.850
0.834
0.826
Concordant pairs
80.3%
82.1%
80.6%
80.8%
84.0%
82.6%
82.1%
Discordant pairs
18.3%
16.5%
18.4%
18.1%
14.3%
15.8%
16.8%
Tied pairs
1.4%
1.4%
1.0%
1.1%
1.7%
1.6%
1.1%
Percentage oF hits
75.6%
65.8%
62.2%
63.8%
73.6%
70.0%
73.5%
Speci±city (TNR)
75.7%
64.8%
60.8%
62.7%
73.4%
69.7%
73.3%
Sensitivity (TPR)
72.0%
90.5%
89.5%
86.3%
82.8%
83.1%
78.0%
Prevalence (nascent %)
3.32%
3.69%
4.99%%
4.63%
2.30%
2.29%
3.86%
Optimal cutoFF point
4.60%
4.34%
4.23%
3.89%
2.63%
2.25%
4.03%
Exp (
b
) must be interpreted as how more or less likely it is to occur the event oF interest expressed in the dependent variable by the Fact oF having the characteristic that contains
the independent variable (1) versus not having it (0), whichever is greater or less than 1, respectively. IF it is exactly equal to 1, the independent variable in question does not exert
any infuence on the dependent variable
.
In the Exp(
b
): ***sig<0.01; **sig<0.05; *sig<0.10; ^sig>0.10.
Hosmer-Lemeshow test signi±cance > 0.05 implies good calibration oF the model.
C-statistic: 0.50 discrimination capacity null; 0.70-0.79 acceptable; 0.80-0.89 excellent;
0.90 exceptional.
The prevalence reFers to the distribution oF the total observations incorporated in the analysis in response to the event oF interest. In this case, it indicates the percentage oF
individuals with nascent entrepreneurial activity From the total oF individuals who were included in the analysis For each period considered.
The optimal cutoFF point is obtained as the one that maximizes the speci±city and sensitivity sum.
34
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
means they analyze the impact of each of the proposed variables on the likelihood of
developing nascent entrepreneurial activity, but keeping the effect from the remain-
ing variables controlled.
The regressions show the odds ratios associated to the estimated coefficients, as
well as the significance linked to the respective Wald statistics for each coefficient,
and the standard error.
As for the validity of the estimated models, these show a good degree of cali-
bration with the data based on the result of the Hosmer-Lemeshow goodness of fit
test. Furthermore, the estimated areas under the ROC curve (all above 80%) indi-
cate very good discrimination ability, with a high degree of concordance for all the
possible mixed pairs of cases
5
. Additionally, the percentages of hits offer, for the
optimal cutoff point in each case, a high predictive power for the nascent activity
event
6
.
With regard to the variables of interest in the present work, in all the regression
models the significance attached to the Wald statistics for each coefficient indicates
that the fear of failure, perception of opportunities, perceived self-efficacy, and
knowledge of entrepreneurs were significant for the target level of 5%, with the
first influencing entrepreneurial involvement negatively, and the last three posi-
tively, especially the perception of self-efficacy. The perceived social desirability
had lower levels of significance, with negative influence compared to what is com-
monly expected, and in the last year analyzed it ceased to be significant. As for the
educational level, it was not significant for the 0.05 level in any of the estimated
models.
For their part, the control variables are significant in all the estimated models
with the exception of gender in 2008 (a year in which male nascent activity was seen
especially affected by that female), showing typical results that, in the case of age,
suggest an inverted
U
shape, and for gender, greater male entrepreneurial propensity
than the female variety.
6.
Comments on the results
Starting with the variables proposed in relation to human and social capital, these
registered different types of behavior. The educational level fails to be significant
for a level of 5%, so the fact of possessing different baggage does not seem to have
5
The graphic representation of all the possible cutoff points on two axes (sensitivity and 1-specific-
ity) defines the ROC curve. The area under the curve (AUC) indicates, for all the possible combinations of
pairs of individuals in which one shows the event but the other does not, the probability of being assigned
a higher probability of the event to which, effectively, indeed shows. This means that it approximates the
probability of correctly classifying a pair of individuals (one 1 and one 0) chosen at random. It is the best
instrument indicative of the discrimination ability of a pattern, given that, moreover, is not affected by
modification of the cutoff point.
6
The optimal cutoff or diagnostic point is defined as that offering a better sensitivity/specificity
pair.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
35
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
a clear relationship with entrepreneurial involvement. In any case, this result is not
surprising given the variety of registries gathered in previous research, which point to
a generally uncertain relationship (Blanchflower, 2004).
Knowledge of recent entrepreneurs, for its part, shows how this is usually a
positive influence at all times, which appears to increase with the change in the
stage of the cycle, and as this stage is more negative. Thus, we observed how the
estimated odds ratio for this variable progressively increased from 2009 to 2011,
as compared to 2008 and years previous, until reaching a central value of 2.72 in
2011, indicating that the presence of role models can make entrepreneurial propen-
sity almost triple. These data are of interest in that they suggest that the presence of
recent entrepreneurs within nearby surroundings appears to become progressively
more influential in entrepreneurial involvement when facing ever more adverse
economic contexts.
Figure 4.
Odds ratio for knowledge of entrepreneurs, 2005-2011
Exp(B)
4
3.5
3
2.5
2
1.5
1
0.5
0
2005
2006
2007
2008
2009
2010
2011
Source:
Own elaboration.
Concerning the behavior of the variables of perception, both their significance
and their direction of influence remain unchanged over time (with the exception of
the perception of social desirability), however showing interesting nuances that are
discussed next.
The evolution of the odds ratio for recognizing business opportunities indi-
cates that this perception increases its influence on entrepreneurial decision-mak-
ing during negative phases of the economic cycle, although significant differences
can only be spoken of properly between 2006-2007 and 2011. In this manner, it
could be felt that the identification of opportunities acts more strongly on entre-
preneurial involvement during adverse stages than stages with better economic
conditions.
36
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Figure 5.
Odds ratio for perception of opportunities, 2005-2011
Exp(B)
4
3.5
3
2.5
2
1.5
1
0.5
0
2005
2006
2007
2008
2009
2010
2011
Source:
Own elaboration.
Perceived self-efficacy is by far the most notable factor in all the cases; how-
ever, it is felt that its impact is reduced with the change of stage of the cycle (from
11.38 in 2007 to the 6.05 registered in 2011, a difference that between these two
years becomes significant). Therefore, the deepening of the negative phase of the
economic cycle and the prolongation of the crisis may ultimately undermine part of
this self-confidence with respect to its influence on entrepreneurship.
Figure 6.
Odds ratio for perceived self-efficacy, 2005-2011
Exp(B)
18
16
14
12
10
8
6
4
2
0
2005
2006
2007
2008
2009
2010
2011
Source:
Own elaboration.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
37
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
The perception of the fear of failure as a barrier maintains its negative influ-
ence, without significant differences, in such a way that entrepreneurial propensi-
ty can be reduced by approximately one-half regardless of whether the economic
context is positive or negative in nature. This result could attract attention, as a
more negative influence in adverse contexts might be expected. However, the reg-
istry obtained seems to indicate that the progressive deterioration in the starting
point for new entrepreneurs reduces opportunity costs of business involvement,
which could have a hand in the influence remaining without significant differ-
ences.
Figure 7.
Odds ratio for fear of failure as a barrier, 2005-2011
Exp(B)
1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
2005
2006
2007
2008
2009
2010
2011
Source:
Own elaboration.
Finally, the perception of social desirability, with a negative influence until
2008, ceased being significant for the 0.05 level from 2009 onwards. The results
for this item certainly seem contradictory given that in the literature it is often
linked positively with the development of entrepreneurial intentions. In this sense,
significant results with these variables were not found in any studies revised within
the GEM context, and in particular, Tominc & Rebernik (2007) point out some
thoughts and concerns about the wording of these questions in the APS survey.
In fact, the response rates obtained for this question in the analyzed sample of-
fer significant differences for the
yes
response, favorable to individuals who do
not engage entrepreneurially, which explains an apparent negative influence of the
perception of social desirability until 2011. Beginning that year, the response rate
differences become diluted as nascent entrepreneurs gain greater recognition as
the population considers entrepreneurial activity desirable, perhaps due to the its
greater merits in times of crisis.
38
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Figure 8.
Odds ratio for perception of social desirability, 2005-2011
Exp(B)
1.4
1.2
1
0.8
0.6
0.4
0.2
0
2005
2006
2007
2008
2009
2010
2011
Source:
Own elaboration.
7.
Conclusions
This work is based on an extensive temporal comparison of GEM data of ele-
ments that are influential in the individual decision to start a business, with the con-
viction that studying entrepreneurial activity in a territory, under decidedly different
stages of the economic cycle, may contribute to improve the understanding of the
determinants of entrepreneurial activity.
Used for this was data belonging to the GEM research consortium in Spain that
resulted from APS surveys carried out between 2005 and 2011. By focusing the study
on nascent entrepreneurs, bias linked to retrospection was minimized (Davidsson,
2006).
The results reinforce the importance of the role of the variables of perception
and role models as determinants of entrepreneurial activity, regardless of the
state
of the environment
effect. This is in line with several studies, among which Arenius
& Minniti (2005), Köllinger
et al.
(2005), and Minniti & Nardone (2007) are found.
These authors mention the agreement of a growing number of investigators who clas-
sify the cited elements among the most important inducers of entrepreneurial behav-
ior, describing their influence in the decision as universal. In this sense, this paper
contributes important reinforcement to the evidence on this matter, while the above
variables have collected results with the same significance and sign of influence, hav-
ing replicated the analysis on seven occasions, with seven different samples, which
moreover were collected at objectively different moments of the economic cycle.
In particular, the influence of perceived self-efficacy is shown as a key factor,
which is related to that raised by the generality of models of intentions and other re-
search on nascent activity (McGee
et al.
, 2009). In this sense, we agree with Minniti
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
39
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
& Nardone (2007: 236) when they affirm, «the perception of having sufficient skills
is a dominant variable that seems to have an effect regardless of institutional settings,
culture and overall level of entrepreneurial activity». In any case, although our results
indicate that judging one’s own capacity positively is the factor with a greater associ-
ated influence coefficient regardless of the context, it seems that a certain reduction
of this influence is glimpsed in a context of economic difficulties.
The results also emphasize the importance of detecting opportunities. Consider-
ing the environment a source of opportunities increases entrepreneurial propensity
in general, especially in adverse contexts. In fact, during the contracting phase of
the cycle, this factor registers influences superior to those found during the growth
phase, showing significant differences. This can be related to the fact that potential
entrepreneurs are more likely to decide to exploit a business opportunity when the
gap
between the expected return of this option and other alternative uses of their time
is greater (Shane, 2003), so that when an opportunity is recognized, individuals with
lower opportunity costs (unemployment, lower household income) will be more in-
clined to exploit it (Amit
et al.
, 1993). The crisis and worsening of the negative phase
of the economic cycle have deteriorated the average economic and labor situation of
the population in aggregate terms, so it is expected that the average opportunity cost
is less and the recognition of opportunities increases its influence.
This means that in hostile economic environments, like the present, the avail-
ability of mechanisms necessary for helping individuals, containing both information
about potential business within their environment as well as tools to identify and
judge the feasibility of such opportunities, becomes even more important.
In this regard, a notable element is called vicarious learning. Its importance in en-
trepreneurial propensity is clear, in that contact with other entrepreneurs can almost
triple it. Furthermore, its importance is even greater in the sense that it also influences
indirectly, as knowledge of recent entrepreneurs and the influence these can exert on
those who have yet to become them (either by facilitating contacts and networks,
learning from the experience of others, imitation, or the
iF
somebody else has done
it, so can I
) have often been highlighted as a source of self-efficacy in several stud-
ies (Bandura, 1986). Similarly, it is also related to the perception of opportunities
(Shane, 2003; Ramos-Rodríguez
et al.
, 2010).
In this element, the scenario analysis performed also shows that its importance is
especially patent in the negative phase of the economic cycle, when it is noticed that
the influence of the knowledge of entrepreneurs on entrepreneurial propensity is pro-
gressively greater, to the point of registering data significantly different from those
collected in the positive phase. Individuals find greater support in networks of con-
tacts and nearby role models. Thus, if the promotion of policies supporting the entre-
preneur and networking among businesses and entrepreneurs comes to be practiced
by many governments, the evidence provided indicates that this policy is especially
relevant in economic environments of crisis and recession like the current one.
Another element traditionally linked to entrepreneurial involvement, but in the
negative sense, is the risk of doing business, of which GEM has obtained an approxi-
40
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
mation with the fear of failure as a possible barrier. This paper adds to the general-
ity of those who have obtained empirical support in this sense by showing it as a
deterrent to entrepreneurship. Moreover, in the two analyzed contexts, its influence
remained unchanged. In a crisis environment, an expected higher barrier could have
been offset by the fact that further deterioration in the entrepreneur’s starting position
would reduce the opportunity costs of the entrepreneurial decision, which would also
reduce the barriers caused by fearing the consequences of a hypothetical failure.
Overall, the results show, on an exploratory basis, the interest in studying in
depth the behavior of these influencing factors in objectively different economic con-
texts. In this regard, future research could confirm the different intensities detected,
and at the intensities that some factors affect decision-making in each scenario, by
incorporating more extensive temporal samples into the research, using data from
upcoming years, as well as their possible replication in other territories.
Nevertheless, this paper provides empirical evidence that supports the impor-
tance of establishing policies that encourage the development of actions to raise self-
efficacy within the population and facilitate the recognition of opportunities and ac-
cess to them, as these elements have shown significant influence regardless of the
environment we find ourselves in. In this sense, strengthening social networks and
promoting knowledge of and contact with entrepreneurs also become essential ob-
jectives, not only because of their direct influence, but also because of their indirect
effects. The nuances found in relation to the different economic climates in which the
analysis was replicated reinforce this idea, and demonstrate the importance of adapt-
ing promotional actions to the situation at all times.
8.
References
Acs, Z. J., and Audretsch, D. B. (2003): «Editor’s introduction», in Z. J. Acs and D. B. Audret-
sch (eds.),
Handbook of Entrepreneurship Research. An Interdisciplinary Survey and Intro-
duction
, Kluwer Academic Publishers, Great Britain, 3-20.
Acs, Z. J.; Desai, S., and Hessels, J. (2008): «Entrepreneurship, economic development and
institutions»,
Small Business Economics
, 31, 219-234.
Ajzen, I. (1991): «The theory of planned behaviour»,
Organizational Behaviour and Human
Decision
, 50, 179-211.
Alsos, G. A.; Ljunggren, E., and Pettersen, L. T. (2003): «Farm-based entrepreneurs: what
triggers the start-up of new business activities?»,
Journal of Small Business and Enterprise
Development
, 10 (4), 435-443.
Amit, R.; Glosten, L., and Muller, E. (1993): «Challenges to theory development in entrepre-
neurship research»,
Journal of Management Studies
, 30, 815-834.
Arenius, P., and de Clercq, D. (2005): «A network-based approach on opportunity»,
Small
Business Economics
, 24 (3), 249-265.
Arenius, P., and Minniti, M. (2005): «Perceptual variables and nascent entrepreneurship»,
Small Business Economics
, 24 (4), 233-247.
Autio, E.; Keeley, R.; Klofsten, M.; Parker, G., and Hay, M. (2001): «Entrepreneurial intent
among students in Scandinavia and in the USA»,
Enterprise and Innovation Management
Studies
, 2 (2), 145-60.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
41
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Bandura, A. (1986):
Social Foundations oF thought and action. A social cognitive theory,
Engle-
wood Cliffs, NJ: Prentice Hall (Translation to Spanish:
Pensamiento y acción. ±undamen-
tos sociales
, Barcelona: Martínez Roca, 1987).
Bergmann, H., and Sternberg, R. (2007): «The changing face of entrepreneurship in Germany»,
Small Business Economics
, 28, 205-221.
Blanchflower, D. G. (2004):
SelF-employment: more may not be better
(NBER Working Paper
no. 10286), National Bureau of Economic Research, Cambridge, MA.
Brück, T.; Llusá, F., and Tavares, J. (2010): «Perceptions, expectations and entrepreneurship:
the role of extreme events»,
Discussion Paper no. 5351, IZA Discussion Papers.
Bruyneel, S.; Carree, M., and Peeters, L. (2006): «Employment status and the business found-
ing process, the case of Norway»,
Paper presented at the 14
th
Nordic ConFerence on Small
Business Research (May 11-13, 2006, Stockholm, Sweden).
Carree, M. A., and Thurik, A. R. (2003): «The Impact of entrepreneurship on economic
growth», in Z. J. Acs and D. B. Audretsch (eds.),
Handbook oF Entrepreneurship Research.
An Interdisciplinary Survey and Introduction
, Kluwer Academic Publishers, Great Britain,
437-471.
Carter, N. M.; Gartner, W. B., and Reynolds, P. D. (1996): «Exploring start-up event sequenc-
es»,
Journal oF Business Venturing
, 11, 151-166.
Crosa, B.; Aldrich, H. A., and Keister, L. A. (2002): «Is there a wealth affect? Financial and
human capital determinants of business start-ups», in W. D. Bygrave
et al.
(eds.),
±rontiers
oF Entrepreneurship Research 2002
, Wellesley, MA: Babson College.
Davidsson, P. (2006): «Nascent entrepreneurship: empirical studies and developments»,
±oun-
dations and Trends in Entrepreneurship,
2 (1): 1-76.
Davidsson, P., and Honig, B. (2003): «The role of social and human capital among nascent
entrepreneurs»,
Journal oF Business Venturing,
18 (3), 301-331.
De Clercq, D., and Arenius, P. (2003): «Effects of human capital and social capital on entrepre-
neurial activity»,
Working Paper Steunpunt OOI
(in progress), June 2003.
Delmar, F., and Davidsson, P. (2000): «Where do they come from? Prevalence and characteris-
tics of nascent entrepreneurs»,
Entrepreneurship and Regional Development
, 12, 1-23.
Diochon, M.; Gasse, Y.; Menzies, T., and Garand, D. (2002): «Attitudes and entrepreneurial
action: exploring the link»,
Proceedings oF the Annual ConFerence oF the Administrative
Sciences Association oF Canada, Entrepreneurship and ±amily Business Division
, 1-10
(best paper award). ASAC 2002, Winnipeg, Manitoba. Another version of this paper exists,
signed by Gasse and Menzies (2002).
Driga, O.; Lafuente, E., and Vaillant, Y. (2005): «Gender differences in entrepreneurial activity:
an analysis of informal institutional factors»,
Paper presented at the 50
th
World ConFerence
oF ICSB 2005: Golden Opportunities For Entrepreneurship
, Washington, DC. June 15-18,
2005.
Eckhardt, J. T., and Shane, S. A. (2003): «Opportunities and Entrepreneurship»,
Journal oF
Management
, 29 (3), 333-349.
Freytag, A., and Thurik, R. (2007): «Entrepreneurship and its determinants in a cross-country
setting»,
Journal oF Evolutionary Economics
, 17 (2), 117-131.
Gartner, W. B.; Shaver, K. G.; Carter, N. M., and Reynolds, P. D. (2004): «Foreword», in W. B.
Gartner, K. G. Shaver, N. M. Carter and P. D. Reynolds (eds.),
Handbook oF Entrepreneur-
ial Dynamics: The Process oF Business Creation,
Sage, Thousand Oakes, ix-xxiii.
Gnyawali, D. R., and Fogel, D. S. (1994): «Environments for entrepreneurship development:
key dimensions and research implications»,
Entrepreneurship: Theory and Practice
, Sum-
mer 1994, 43-62.
Greene, P. G. (2000): «Self-Employment as an economic behavior: an analysis of self-
employed women’s human and social capital»,
National Journal oF Sociology
, 12 (1),
1-55.
42
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Gries, T., and Naudé, W. (2010): «Entrepreneurship, structural change and a global economic
crisis», UNU-WIDER. United Nations University. World Institute for Development Eco-
nomics Research, Working Paper no. 2010/57.
Johnson, P. S.; Parker, S. C., and Wijbenga, F. (2006): «Nascent entrepreneurship research:
achievements and opportunities»,
Small Business Economics
, 27 (1), 1-4.
King, G., and Zeng, L. (2001a): «Logistic regression in rare events data»,
Political Analysis
,
9 (2), 137-163.
— (2001b): «Explaining Rare Events in International Relations»,
International Organization
,
55 (3), 693-715.
Köllinger, P., and Minniti, M. (2006): «Not for lack of trying: American entrepreneurship in
black and white»,
Small Business Economics
, 27 (1), 59-79.
Köllinger, P.; Minniti, M., and Schade, C. (2005): «I think I can, I think I can: overconfidence
and entrepreneurial behavior»,
DIW Berlin. German Institute for Economic Research,
Dis-
cussion Papers, 501.
Krueger, N. F. (2000): «The cognitive infrastructure of opportunity emergence»,
Entrepreneur-
ship: Theory and Practice
, 24 (3), 5-23.
— (2003): «The cognitive psychology of entrepreneurship», in Z. J. Acs and D. B. Audretsch
(eds.),
Handbook of Entrepreneurship Research
, 105-140.
Krueger, N. F., and Brazeal (1994): «Entrepreneurial potential and potential entrepreneurs»,
Entrepreneurship: Theory and Practice
, 18 (3), 91-104.
Krueger, N. F., and Carsrud, A. (1993): «Entrepreneurial intentions: applying the theory of
planned behavior»,
Entrepreneurship and Regional Development
, 5, 315-330.
Lee, L.; Wong, P. K.; Chen, J., and Chua, B. L. (2005): «Antecedents for entrepreneurial pro-
pensity: findings from Singapore, Hong Kong and Taiwan»,
NUS Entrepreneurship Centre
,
Working Papers, reference no. WP2005-06.
Marcu, G.; Oirdanescu, C., and Iordanescu, E. (2012): «The influence of the psychological
factors upon the entrepreneurial tendency in crisis environments»,
Procedia. Social and
Behavioral Sciences
, 33, 473-477.
McGee, J. E.; Peterson, M.; Mueller, S. L., and Sequeira, J. M. (2009): «Entrepreneurial self-ef-
ficacy: refining the measure»,
Entrepreneurship: Theory and Practice
, July 2009, 965-988.
Minniti, M., and Lévesque, M. (2008): «Recent developments in the economics of entrepre-
neurship»,
Journal of Business Venturing
, 23, 603-612.
Minniti, M., and Nardone, C. (2007): «Being in someone else’s shoes: the role of gender in
nascent entrepreneurship»,
Small Business Economics
, 28 (2/3), 223-238.
Mueller, P. (2006): «Entrepreneurship in the region: breeding ground for nascent entrepre-
neurs?»,
Small Business Economics
, 27 (1), 41-58.
Naudé, W. A. (2010): Global crises and entrepreneurial incentives», Conference presented at
UNU Worldwide Series, New York. June 24, 2010. Available at
events-forums/new/WWNY/2010/global-crises-and-entrepreneur.php and http://www.us-
tream.tv/recorded/7865573
.
Naudé, W. A., and MacGee, J. C. (2009): «Wealth Distribution, the Financial Crisis and Entre-
preneurship», WIDER Angle Newsletter, March 2009.
Naudé, W. A.; Fries, T.; Wood, E., and Meintjies, A. (2008): «Regional determinants of entre-
preneurial start-ups in a developing country»,
Entrepreneurship and Regional Develop-
ment: An International Journal
, 20 (2), 111-124.
Ramos-Rodríguez, A. R.; Medina-Garrido, J. A.; Lorenzo-Gómez, J. D., and Ruiz-Navarro,
J. (2010): «What you know or who you know? The role of intellectual and social capital
in opportunity recognition»,
International Small Business Journal
, December 2010, 28,
566-582.
Reynolds, P. D. (1997): «Who start new firms? Preliminary explorations of firms-in-gestation»,
Small Business Economics
, 9 (5), 449-462.
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
43
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Reynolds, P. D., and White, S. B. (1997):
The entrepreneurial process: economic growth, men,
women and minorities
, Westport, CT: Quorum Books.
Reynolds, P. D.; Bosma, N.; Autio, E.; Hunt, S.; De Bono, N.; Servais, I.; López-García, P., and
Chin, N. (2005): «Global entrepreneurship monitor: data collection design and implemen-
tation 1998-2003»,
Small Business Economics
, 24, 205-231.
Shane, S. (2003):
A general theory oF entrepreneurship. The individual-opportunity nexus
,
Cheltenham: Edward Elgar.
Shane, S., and Venkataraman, S. (2000): «The promise of entrepreneurship as a field of re-
search»,
Academy oF Management Review
, 25 (1), 217-226.
Shapero, A., and Sokol, L. (1982): «The social dimension of entrepreneurship», in Kent, C. A.;
Sexton, D. L., and Vesper, E. (eds.),
Encyclopedia oF entrepreneurship
, 72-90. Prentice-
Hall, Englewood Cliffs, NJ.
Storey, D. (1994):
Understanding the Small Business Sector
, London: Rouledge.
Tamásy, C. (2006): «Determinants of regional entrepreneurship dynamics in contemporary
Germany: a conceptual and empirical analysis»,
Regional Studies
, 40 (4), 365-384.
Tominc, P., and Rebernik, M. (2007): «Growth aspirations and cultural support for entrepre-
neurship. A comparison of post-socialist countries»,
Small Business Economics
, 28 (2-3),
239-255.
Van Praag, C. M., and Versloot, P. H. (2007): «What is the value of entrepreneurship? A review
of recent research»,
Small Business Economics
, 29, 351-382.
Venkataraman, S. (1997): «The distinctive domain of entrepreneurship research: an editor’s
perspective», in J. Katz and J. Brockhaus (eds.),
Advances in Entrepreneurship, ±irm
Emergence, and Growth. Greenwich
, CT: JAI Press, 3, 119-138.
Verheul, I.; Wennekers, D.; Audretsch, D. B., and Thurik, R. (2002): «An eclectic theory of
entrepreneurship: policies, institutions and culture», in D. B. Audretsch, R. Thurik and I.
Verheul (2002),
Entrepreneurship: determinants and policy in a European-US comparison,
Boston: Kluwer Academic Publishers.
Wagner, J. (2004): «Nascent entrepreneurs»,
IZA DP,
no. 1293. Bonn, Germany: Forschungsin-
stitut zur Zukunft der Arbeit.
Wagner, J., and Sternberg, R. (2004): «Start-up activity, industrial characteristics and the re-
gional milieu: lessons for entrepreneurship support policy from German microdata»,
An-
nals oF Regional Science
, 38, 219-240.
Weiss, G. M.; McCarthy, K., and Zabar, B. (2007): «Cost-sensitive learning vs. sampling:
Which is best for handling unbalanced classes with unequal error costs?»,
Proceedings oF
the 2007 International ConFerence on Data Mining
, CSREA Press.
Wennekers, S; Uhlaner, L. M., and Thurik, R. (2002): «Entrepreneurship and its conditions: a
macro perspective»,
International Journal oF Entrepreneurship Education
, 1 (1), 25-65.
44
Martínez Mateo, J., Mira Solves, I. and Gómez Gras, J. M.ª
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Annex. Operationalization of variables
Nascent activity
In order to be identified, all individuals are asked: (1)
«Are you, alone or with
others, currently trying to start a new business, including any self-employment or
selling any goods and/or services to others?».
Those answering affirmatively are
inquired about: (2)
«Over the past 12 months, have you done anything to help start
this new business, such as looking for equipment or a location, organizing a start-
up plan, working on a business plan, beginning to save money, or any other activity
that would help launch a business?»
;
(3)
«Will you personally own part of this busi-
ness?»
;
(4)
«Has the new business paid any salaries, wages, or payments in kind,
including your own, for more than three months?».
For all questions, the individuals
have the option of responding one of four ways:
Yes
,
No
,
Don’t know
, or by not an-
swering/refusing. In the subsequent classification of variables, a person is classified
as a
nascent entrepreneur
if, in addition to question (1), he/she answers
Yes
for items
(2) and (3), and
No
for (4) (SUBOANW variable
=
1).
Explanatory variables
Table 4.
Explanatory variables used: questions, values, and classifications
(GEM APS - Spain 2007 and 2009)
Variables of interest
Corresponding question
in the APS survey
Values
and classiFcations
Knowledge
of entrepreneurs
(KNOWENT)
Do you know someone personally who
started a new business in the past two
years?
— Yes (1)
— No (0)
Level of education
(EDUC)
What is the highest level of education that
you have completed?
Recoded by the surveying body from the
original response obtained.
— None or primary (1)
— Lower secondary (2)
— (Upper) secondary (3)
Social desirability
(NBGOODC)
In your country, most people consider
starting a new business a desirable ca-
reer choice.
— Yes (1)
— No (0)
Fear of failure
(FEARFAIL)
Would fear of failure prevent you from
starting a business?
— Yes (1)
— No (0)
Perception of
opportunities (OPPORT)
In the next six months, will there be good
opportunities for starting a business in
the area where you live?
— Yes (1)
— No (0)
Perceived self-efficacy
(SUSKILL)
Do you have the knowledge, skill and
experience required to start a busi-
ness?
— Yes (1)
— No (0)
Infuence oF the economic cycle on the determinants oF nascent entrepreneurial activity
45
Investigaciones Regionales, 26 (2013) – Pages 19 to 45
Table 4.
(continue)
Control variables
Corresponding question
in the APS survey
Values
and classi±cations
Gender
Sex oF the person being interviewed
— Male (1)
— Female (0)
Age
What is your current age in years?
— Years
The questions are formulated for the entire sample. In addition to the response options listed in the table, the individuals
could have answered
Don’t know
or refused to answer, options that were considered missing values in all the
questions.
As for age, its value squared was also used to identify nonlinear relationships between it and nascent entrepreneurial
activity.
In the case of education, (1) indicates that they have none or at most have completed part of secondary education;
(2) corresponds to a secondary degree; and (3) indicates education beyond secondary and higher education.
logo_pie_uaemex.mx