Introduction

In recent years, studies of entrepreneurship has increased in Mexico’s academic and business scope. ^{Gúzman and Trujillo (2008)} illustrate two types of entrepreneurship: social-oriented and business-oriented. The primary objective of social-oriented entrepreneurship is the creation of social sustainable value. Business-oriented entrepreneurship refers to the innovating activities that create value to the entrepreneur. This business focused entrepreneurial activity in Mexico contributes to economic growth and the creation of jobs.

The ^{INEGI (2012)} stated that there were approximately 5 million enterprise units in Mexico. Of these, 99.8% are micro, small or medium enterprises that generate 52% of Mexico’s Gross Domestic Product (GDP) and employ 72% of the people in the country. Consequently, micro, small, and medium enterprises constitute the spine of the national economy in terms of commercial agreements and enterprise growth. According to the Secretary of the Economy, these enterprises generally only maintain their success in the national market for two or three years. The reasons for this include a lack of planning, data, operational problems, production, finances, vision, intern management, export support, and corruption (^{Dussel, 2004}; ^{Secretaría de Economía, 2013}).

Accordingly, the generation and participation of micro, small and medium businesses in the Mexican territory is extremely relevant. In this study, we analyze Mexico’s micro-business. Micro-businesses include a team of up to 10 people in extractive, constructive, service, transport and commercial industries and up to 16 people in the manufacturing industry. These teams include the owner (^{ENAMIN, 2012}).

INEGI conducted a national poll of micro-businesses (^{ENAMIN, 2012}) to collect information about the principal characteristics of micro-business’ productive resources, organization, and expense and income amounts and distributions. We thoroughly reviewed ^{ENAMIN (2012)} to determine the reasons people start micro-businesses and their expectations for continuing with it. The present study aims to determine the factors that influence micro-business owners to start a micro-business and to keep operating their micro-businesses in the future. Qualitative and quantitative data were reviewed. A binary discrete choice model¹ is used to obtain the estimations of the probability that Mexican micro-business owners will continue operating their micro-business, taking into account business performance and expectations.

Literature review

The creation, development and support of micro, medium and small businesses is a common area of investigation, especially in developing economies. These businesses promote employment, investment, innovations and competition. They significantly contribute to economic growth that, in turn, can increase social benefits.

^{Audretsch and Fritsch (2002)}, ^{Foelster (2000)}, ^{Kleppleer (1996)}, ^{Audretsch (1995)}, ^{Hopenhayn (1992)} and ^{Lambson (1991)} maintain that entrepreneurial activity plays an important role in the economic growth of a country. If the government supports and promotes the creation of new companies, their level of economic growth will be greater. In this sense, ^{Reynolds et al. (2002)} points out that the central government of every country will need to formulate more effective public policies for the creation of new enterprises. These policies should focus specifically on the entrepreneurial activities of each society. Entrepreneurial activity is a topic that has had a variety of theorical-methodogical discussions. Hence, there are many ideologies that use different methods to identify and quantify social entrepreneurship. A great diversity of studies that try to explain these issues exist. In this research, the focus is on the business sector, particularly on the entrepreneurial activity of micro-businesses.

^{Audretsch and Fristsch (1994)} and ^{Keeble and Walker (1994)} state that measuring the term “entrepreneurship” in any commercial activity is not an easy task, because of its intangible nature. Hence, the definition and measurement of this term can include structural, social and/ or personal factors. Structural factors refer to the factors in the environment that can directly or indirectly influence the entrepreneurial activity (e.g., company culture, economic activity, economic conjecture, financial environment, social capital). The personal factors refer to dissatisfaction with the wage earned job, opportunities, financing capability, the need for achievements, familiar environment, age, academic formation and personal security.

Some studies about entrepreneurial activity, influence, motivation and business opportunities have focused on identifying the aspects or reasons business owners consider relevant for carrying on with the creation of a company. ^{Alonso and Galve (2008)}, ^{Álvarez and Valencia (2008)}, ^{García et al. (2010)} and ^{Kantis (2003)}, point out that every initiative or entrepreneurial activity related to the creation of a new enterprise depends on the motivations and observations the person has in relation to changes in their environment. Entrepreneurs are constantly looking for opportunities. They have a knack for finding social relations that depend on their business. They strive to develop loyal clients and suppliers (^{Gomez, 2007}).

Entrepreneurs typically have seven characteristics that influence the creation or successful process of the new enterprise: 1) enthusiasm, 2) risk takers, 3) spirit of sacrifice, 4) need, 5) ambition, 6) independence, and 7) knowledge (^{Garcia et al., 2007}). ^{Entrialgo et al. (1999)}, ^{Fernández and Junquera (2001)}, ^{Quintana (2001)} and ^{Mintzberg et al. (1999)} concur and argue that the personality characteristics of the most successful business owners include a need for success, tolerance to ambiguity, internal control, and a willingness to take risks.

^{Elston and Audretsch (2011)} and ^{Rodríguez (2006)} found that in the United States and Venezuela, economic factors or sources of financing are essential for carrying out an entrepreneurial project. The three sources of initial financing of an entrepreneur are their own resources, their family’s resources, and the resources of their partners and friends. ^{Gelderen et al. (2001)}, ^{Quintana (2001)}, and ^{Pinillos (2001)} maintain that the demographic characteristics of the entrepreneur, from a physiological perspective, are also relevant. These characteristics include age, marital status, level of education, experience in the creation of companies, work experience and a family background in creating companies.

In the case of Mexico, ^{Peña et al. (2012)} remarks that not all micro-businesses are similar. This is because the reasons for starting them are different and depends, in many cases, on the opportunities that arise. Some goals include being self-employed and independent. As well, growth perspectives are limited because there is no inversion in capital assets (e.g., infrastructure, equipment, tools). Most micro-business owners are not interested in obtaining the economic support of a financial entity. In this sense, ^{Boenfil (2003)} establishes that the micro-business owner goes out to the entrepreneurial world in the quest of the best opportunities and to have more income. Hence, the first financial source they appeal to is their family and friends. ^{Lecuona (2009)} mentions that the suppliers are agents of big importance in the success of the micro-business. Through them, they obtain their company’s raw materials. They also occupy the principal position as the source of the micro-business funding.

^{Casas and Ibarra (2013)}, ^{Nava (2013)}, ^{Gómez et al. (2009)} and ^{Dussel (2004)} analyzed the qualitative factors that influence the creation of companies. They show that business entrepreneurs must have three different perspectives: motivational, economics, and financial factors. These authors conclude that the creation and success of an enterprise depends on several socio-economic factors, including having a well-functioning capital structure. They also need support and financing for the development of innovative companies and services. Accordingly, qualitative and quantitative factors influence the operation or creation of a micro-businesses.

Data

To determine the factors that influence the reasons to operate a micro-business, we use the ENAMIN of 2012.² This survey is published by INEGI. It contains information about the primary characteristics of a Mexican micro-business’ productive resources, organizations, expense amounts, the distribution of expenses and income. ENAMIN collected specific information about the reasons for starting the surveyed businesses. Approximately 27,000 micro-business owners around the country provided the documentation for this database.

For our investigation, 24,975 observations were useable for the analysis. The collected data highlights the main reasons for how the micro-business began, its financing limitations, whether it is facing problems, salary, personnel employed and the plans of the owner to continue with the micro-business.³

Empirical analysis

This section presents the empirical model. The model’s purpose is to quantify the effects of the explanatory variables on the probability that a Mexican micro-entrepreneur will continue to operate their micro-business. To achieve this, a binary discrete choice model (logit) is applied. This model allows us to obtain estimates of the probability of an event, identify the risk factors that determine these probabilities, and determine the influence or relative weight that these have on themselves. This procedure is used when the numbers of alternatives are two, qualitative and mutually exclusive.

The logit model allows for the calculation for each sector of the total business (enterprises sample) and the probability of belonging to one or another category established for the dependent variable. For example, if the probability is high, the event has a high chance of occurring, otherwise it has reduced probability values (i.e., close to zero) or a zero chance of occurrence, since the cut-off value is 0.50. ^{Pérez (2005)} points out that for a vector X = (X ₁ , X ₂ , ... ,X _n ) of independent variables, the probability of event occurrence is calculated by a conditional probability, using the following expression:

(1)

Where: ₀ is the independent term, ₁ X _{1 2} X ₂ … _k X _k are the linear combinations of the independent variables.

Equation (1) represents the probability of the occurrence of the event, given the characteristics determined by the independent variables (X ₁ , X ₂ , ... ,X _n ) for each micro-entrepreneur in the business (enterprises sample). This is obtained by an expression involving the exponential functions of base “e” (2.7182...). It can be observed that the denominator of the equation is always greater than the numerator. Hence, in all cases, it is always positive and less than unity.

Rearranging Equation (1), we get:

Using Equation (2), the results can be narrowed between 0 and 1, since:

Therefore, the logit regression model will be as follows:

where: Z _i = β ₁ + β ₂ X _2i + …+ β _k X _ki Given the values of the independent variables X ₂ , … ,X _k the probability that the dependent variable takes on the values of 1 and 0 are:

Finally, the marginal effects of the model are calculated from Equation (3), obtaining the (partial derivative) of the variables, resulting in the following form:

Results

Table 6 presents the results of the logit model in relation to continuing to operate a microbusiness in Mexico. The alternatives chosen by the micro-entrepreneur according to the ENAMIN are:

Y _i = 1, if the micro-entrepreneur plans to continue with the micro-business in the year following the period of study;

Y _j = 0, if the micro-entrepreneur does not plan to continue with the micro-enterprise in the year following the study period.

Each of the explanatory attributes of the probabilities of choice have been incorporated in terms of a set of dummy variables that take on the value of one, if the attribute in question takes a determined modality, and zero otherwise (Table 4).

It is important to mention that when analyzing the quality of the model's adjustment, the indicator of the correctly adjusted values indicates that the model predicts 94.2% of the observations correctly. Similarly, the Wald chi2 function (341.0) shows that the value of the coefficients are jointly significant to explain the probability that micro-entrepreneurs plan to continue with the micro-business. The value of Prob>Chi2 (0.0000) indicates that the hypothesis of all the coefficients that are equal to zero can be rejected at the 1 percent level. Finally, the interpretation of Pseudo R2 (0.0309) establishes that 3.09 percent of the variation of the dependent variable can be explained by the variation in the explanatory variables.

In relation to the individual characteristics used as a control, we highlight that age is a factor that explains the probability that a micro-entrepreneur wants to continue with the micro-enterprise. Since it is a statistically significant variable, it can be inferred that a microentrepreneur that is one year older increases their opportunity to continue with the microbusiness by 1.05 times more than that of the younger micro-entrepreneur, keeping all other factors constant.

As for the schooling level, the signs found in the variables (Sch1 and Sch5) are negative, which means that there is a negative relationship between the level of schooling and the reason for continuing with the micro-business. In other words, if a micro-entrepreneur has professional studies (e.g., bachelor’s degree of engineering, Sch5) their opportunity to continue with the micro-business are, on average, 2.65 lower than if they did not have that degree. In general, it is appreciated that as the micro-entrepreneurs have one more year of schooling, the opportunities to continue with the micro-enterprise in Mexico are lower than if they did not. As such, the opportunities practically double if the other constant factors are maintained (Sch1=1.88, Sch2=1.98, Sch3=2.08 and Sch4=2.20).

The results also show that single micro-entrepreneurs (Status5) are less likely to continue with the micro-business. Hence, the estimate of -0.2152 indicates that single micro-entrepreneurs decrease by an average of 0.2 units, suggesting a negative relationship between marital status and continuing to maintain a micro-business. This implies that if a micro-entrepreneur is single, their opportunity to continue or create a micro-business are, on average, 1.24 lower than that of an engaged person, a person in a relationship or a married person.

In our analysis of economic sectors, it was observed that the micro-entrepreneurs most interested in continuing with a micro-enterprise are in the sectors of manufacturing and construction (Sec1 and Sec3). Hence, the opportunity to continue with the business is, on average, 1.01 higher in manufacturing and construction than if they were in another sector.

Other interesting results that should be mentioned include the main reasons microentrepreneurs start a micro-business. Micro-entrepreneurs who inherit the micro-business are more able to continue with it, because it is a family tradition. This means that if a microentrepreneur falls into this category, the opportunity to grow the micro-enterprise is, on average, 1.89 higher. In addition, micro-entrepreneurs who undertake a micro-business to exercise their trade or career (TEC) are motivated and inspired to continue with their micro-business by up to 1.29 times more than those who do not do this.

On the other hand, the LOE and NJO variables decrease by an average of 0.58 and 0.25 units, suggesting a negative relationship between the micro-entrepreneurs who did not have the schooling required for a job and did not find one with the reason to continue attending the micro-business. Micro-entrepreneurs who are within these categories are 1.79 and 1.29, respectively, less likely to continue with the business.

Regarding Finance and Workers, if micro-entrepreneurs plan to continue with the microbusiness, these variables will have a positive and significant impact. The micro-entrepreneurs who undertake a micro-business with the support of a credit or subsidy are motivated to continue their micro-business by up to 1.18 times more than those who do not apply for financing. In addition, micro-entrepreneurs who are in charge of support staff (workers) will continue to operate their micro-enterprise to be more profitable up to 1.39 times more than those who work alone.

By other way, the interpretation of the marginal effects is different regarding the estimates of the logit model. In addition, the calculations for the different types of micro-businesses in the economic sectors applied in this study are different [P(Y =1); P(Y =2); P(Y =3); P(Y =4)]⁶.

Therefore, for reasons of space, we explain the variables that were significant for those microentrepreneurs who decided to continue with their micro-business in Mexico [P(Y =Global)]⁷.

However, the reader may make his or her own inferences after reading the following interpretations for the missing variables or economic sectors.

In relation to the variable Age, it is important to mention that an increasing age of the micro-entrepreneur increases the probability of continuing with the micro-business (0.24%), keeping the rest of the variables constant. The age squared coefficient (Age2) indicates that after a certain age, the intention to continue with the micro-business began to decrease by 0.00253%.

The negative sign on the schooling variables (Sch) show that as one year of schooling increases, the impact of continuing and growing a micro-enterprise is lower (Sch1=3.44%, Sch2=3.64%, Sch3=3.90%, Sch4=4.19% and Sch5=5.18%). These results are coherent from the perspective that the more educated individuals are looking for a well-paid salaried job. For this reason, they will begin to neglect their micro-businesses to dedicate themselves to other activities that are more profitable.

On the other hand, the positive signs on the BFT, FGBO and TEC variables show that if the micro-entrepreneur inherits the micro-business, the probability of remaining in this form of association increases by 3.38%. Also, if the micro-entrepreneur has enough capital and finds a good opportunity to start a micro-business, the probability of continuing its operation in the coming years increases by 1.72%. Regarding micro-entrepreneurs undertaking a microbusiness to practice their profession, it is observed that the probability of success or to continue growing is 1.38%.

For the Finance variable, which shows a clear relationship for those micro-entrepreneurs who undertake a micro-enterprise with the support of a bank credit or subsidy, they have more of a probability to grow and continue in the coming years by 0.92%. Similarly, microentrepreneurs in charge of salaried or trustworthy workers are more likely to continue operating their micro-business by up to 1.76% more than those who work alone.

Discussion of the findings

The primary purpose of this paper was to explore the factors that influence the decision to continue operating a micro-business in Mexico once is undertaken. For this purpose, data from the 2012 National Survey of Micro-business, provided by INEGI, was analyzed. The results indicate that most micro-entrepreneurs (92%), despite having limitations, problems and low profits in the operation of their micro-businesses, plan to continue working in the coming years. This discovery is relevant, given the importance of micro-enterprises in the country's economy. Remember that, in Mexico, according to the latest economic censuses of the INEGI in the period of study, economic units total about 5,144,056 units, of which 99.8% employ less than 250 workers. This means that there are about 5,041,175 micro, small and medium enterprises. Of these, 4,886,853 are micro-businesses with less than 10 workers. Hence, of the total number of companies in the country, 95% are micro-enterprises. For that reason, state or federal programs that support the micro-business of the country should be improved or created, since they are strong supporters of the Mexican economy.

Other interesting findings in this paper include the identification of sociodemographic factors and the reasons for undertaking a micro-business and watching how they impact the decision to continue operating in the coming years. For example, the BFT, FGBO, TEC, JFPP, Finance and Workers variables show that there is a statistically significant impact on the probability that micro-entrepreneurs in the study sample will continue to operate their microbusinesses in Mexico. This is interesting if it is discussed from a neoclassical perspective, since economic studies about the theories of profit maximization, utility, and incentives, argue that individuals start a business to generate wealth (^{Hurst & Pugsley, 2011}; ²⁰¹⁵). However, the present study finds that the reasons and performance to start a micro-enterprise in Mexico are different in each micro-business and are not necessarily economic.

The results of this research are consistent with that of ^{Cuevas et al. (2005)}, who indicated that the behavior of micro-enterprises is highly correlated with the sociodemographic characteristics of the pattern. However, economic and market variables are most influential in determining profitability and productivity. ^{Peña et al. (2012)} used a linear regression model of probability and found that micro-businesses were not equal, since the reasons for starting them varied according to the expectations of the owners. The authors also point out that differences in these ratios or expectations imply that the goal that all micro-enterprises receive funding may not be desirable from an efficiency standpoint and depends on the economic sector in which the micro-business is located.

Conclusions

This article conducted a descriptive and inferential analysis of the 2012 National Survey of Micro-businesses in Mexico. The results reveal that the main reasons micro-entrepreneurs start or undertake a micro-business are to complement and improve their family income. 13.7% of micro-entrepreneurs in the sample argue that they did not need any type of financing, since their personal savings allowed them to start their own micro-business (48.2%). Most of the micro-entrepreneurs surveyed seemed to have a positive perception about continuing the micro-business, since they maintained that the Mexican government was implementing policies, programs and support funds to strengthen them. We also point out that the results derived from the inferential analysis provide relevant information about the determinant factors to continue operating a micro-business.

The application of a logit model allowed us to determine the effects of the variables that influence the probability of a micro-entrepreneur deciding to continue with his or her microbusiness. Among the main findings of this analysis, we can remark that:

Although this technique may lack predictive power, we understand that it is a valid procedure to select, from the perspective of the probability of micro-entrepreneurs, as certain significant relationships with the activity of continuing to operate a micro-business were identified. Additionally, due to the rigorous work in the application of a binary discrete choice model, in this study, we are aware of the limitations and opportunities for improvement. For example, at the conceptual level, the results have raised some interesting questions about the main determinant factors that currently influence the entrepreneur. A comparison of the same variables in the different states or regions of the Mexican Republic could improve the knowledge, perception and disjunctives in the business activity of the nation. We also recommend that future studies compare more years of data by gender.