Introduction

According to ^{Teece, Pisano, and Shuen (1997)}, the distinctive skills and capacities of the organizations are grouped into three categories: particular resources or assets, trajectory of the organization, and organizational and management processes, among which learning stands out. Moreover, in increasingly changing environments, the theory of dynamic capabilities recognizes that companies need to renew their capabilities, and for this, adaptation, absorption, and innovation are essential factors (^{Wang and Ahmed, 2007}). At the same time, being able to learn is key, especially for innovation, which is highly demanding in terms of knowledge generation and application (^{Senge, Roberts, Ross, Smith and Kleiner, 1994}). In this logic, it can be expected that organizations that adequately manage the learning process, including knowledge acquisition, information distribution and interpretation, and organizational memory are able to develop competitive advantages (^{Huber, 1991}) that impact on their performance indicators.

The empirical evidence on the relationship between dynamic capacities and performance depends on the type of observed capacities (generic or specific), on the inclusion in the models of variables coming from the environment, and on the measurement or analysis methods used (^{Pezeshkan, Fainshmidt, Nairm Frazier and Markowski, 2016}). However, positive relationships between learning, outcomes, and organizational survival, as well as with innovation, are often observed in more learning-oriented enterprises (^{Keskin, 2006}; ^{Alegre and Chiva, 2008}, ²⁰¹³; ^{Torres and Jasso, 2009}; ^{Rhee, Park and Lee, 2010}; ^{Wang and Ellinger, 2011}; ^{Sok, O’Cass and Sok, 2013}; ^{Gomes and Wojahn, 2017}; ^{Liu, 2017}; ^{Tajeddini, Altinay and Ratten, 2017}).

In Mexico, as in other emerging economies, the average results of innovation and business performance are not encouraging¹, and this situation is more critical among smaller companies. The causes of this reality are multifaceted, but triggering learning processes in organizations should have positive impacts at the micro and macro levels.

In this sense, research on the consequences of innovation learning in micro, small, and medium-sized Mexican companies is scarce, and some of them focus on economic sectors with very particular characteristics, as we can observe below.

^{Mendoza and Valenzuela (2014)} analyze learning, innovation, and technological management in micro, small, and medium enterprises in the metalworking and information technology industries. The authors conclude that there is an intrinsic relationship between technical knowledge and its management; they affirm that, in a learning framework, this relationship generates technological innovation. ^{Estrada and Dutrénit (2007)} studied knowledge management in micro and small machining companies in two Mexican cities. They found that human capital, expressed through tacit and individual learning variables, has a positive influence on almost all innovation indicators. Therefore, they suggest that there is no process of change or improvement without the intervention of the skill, experience, and knowledge of human resources; in general, their results reaffirm the influence of intellectual capital on performance.

The works of ^{Martínez Serna, Vega Martínez, and Vega Martínez (2016)} and ^{Martínez Serna, Vega Martínez, and Eternod Domenech (2018)} on micro, small, and medium manufacturing, commercial, and services enterprises in the city of Aguascalientes explore the relationships between learning orientation, organizational commitment, innovation, and performance, concluding that learning orientation is a good predictor for improving innovation and business performance. However, they do not find a direct relationship between learning orientation and performance, but rather indirectly through innovation.

San Luis Potosi is one of the Mexican states whose growth in recent years is above the national average and its unemployment rate is lower. However, there are no works that analyze the innovation capacity of the productive plant focusing on learning and orientation to innovation of state enterprises².

Speaking of the performance of micro, small, and medium enterprises, ^{Ndiaye, Razak, Nagayev, and Ng (2018)} point out that, although research on the performance of these enterprises has increased, there is no consensus on the factors that drive them and emphasize that academic literature focuses disproportionately on developed economies.

Derived from the above, within the framework of the theory of dynamic capabilities, the underlying research question in this paper is: What is the role of learning orientation in innovation and the performance of micro, small, and medium-sized enterprises in the city of San Luis Potosí? The motivation of this is to increase the empirical evidence on the subject from the little explored context of an emerging economy like Mexico, and the smaller-sized enterprises of the commerce, industry, and services sectors.

Below, based on the review of the literature, the hypotheses and the proposed theoretical model are justified. Subsequently, the method used to give way to the results is explained. The last section sets out the conclusions and limitations.

Review of the literature and development of the hypothesis

Learning orientation and business performance

In the literature, the topic of the performance of companies that promote organizational learning stands out. ^{DiBella et al. (1996)} conceptualize learning orientation as the ability or process of an organization to maintain or improve its performance based on experience, involving the acquisition, dissemination, and use of knowledge. Such knowledge could potentially be the most productive resource of an organization and an indispensable factor in achieving sustainable competitive advantage (^{Barney, 1991}; ^{Day, 1994}; ^{Slater and Narver, 1995}; ^{Grant, 1996}; ^{Sinkula et al., 1997}; ^{Hult, Snow and Kandemir, 2003}; ^{Wang and Wei, 2005}). This is due to the fact that a learning-oriented organization adequately processes the information it receives from its clients, competitors, and distribution channels, creating a critical competition to improve performance and, thanks to the information gathered, it is able to foresee changes in the environment and the market, making the necessary adjustments in time (^{Calantone et al., 2002}).

Based on the above, the first hypothesis is presented:

H1: The learning orientation of a company positively influences its performance.

Learning orientation and focus on innovation

Hand in hand with the relationship between learning and the generation of new knowledge, the approach to innovation is conceptualized as the extent to which an organization shows willingness to accept new concepts by changing the schemes it uses (^{Hurley and Hult, 1998}). When called an approach to innovation it also refers to the set of values and beliefs that stimulate creativity (^{Walter, Auer and Ritter, 2006}), invention (^{Hult et al., 2003}), and experimentation, promoting and supporting new ideas and openness towards them (^{Lumpkin and Dess, 2001}; ^{Keskin, 2006}), often assisted by the use of new technological resources or resulting in new products, services, or processes (^{Lumpkin and Dess, 2001}).

Learning orientation represents the desire of the organization for knowledge, and the focus on innovation emphasizes the search for change through new knowledge. Both capabilities are related because an organization committed to learning enhances its innovative potential since it is more likely to be committed to innovation, invest in cutting-edge technology and uses it to create and sell new products. In turn, learning organizations possess knowledge and skills to understand and anticipate consumer needs and also tend to develop greater innovative capacity than their competitors because they monitor and learn from their actions in the marketplace (^{Damanpour, 1991}; ^{Cahill et al., 1996}; ^{Calantone et al., 2002}).

^{Slater and Narver (1995)} suggest that the main link between innovation and learning is due to the fact that the values that promote learning orientation create in the company a tendency to go into a process of seeking knowledge and encourage the concern to break the status quo of the company to achieve better performance, encouraging the exploration of new markets and the development of new products, services, and technologies.

From the above, the following hypothesis is proposed:

H2: The learning orientation of a company positively influences its performance.

Differences associated to size and age of the company

Empirical works show interest in analyzing whether the age and size of the organization influence the relationship between learning and innovation. In most cases, it is observed that there are differences but it is not possible to generalize a conclusion on the matter, because this also depends on other factors such as the order in which companies enter the market, the economic sector, the diversity in the portfolio of research and development projects, the degree of competition in the market, the presence of economies of scale, the life cycle of the market, the degree of technological development, learning rates, or the network of contacts (^{Balasubramanian and Lee, 2008}; ^{Bianchini, Krafft, Quatraro and Ravix, 2015}; ^{Ciriaci, Moncada-Paternó-Castelo and Voigt, 2012}; ^{Coad, Segarra and Teruel, 2016}; ^{Hansen, 1992}; ^{Rogers, 2004}; ^{Hui, Radzi, Jenatabadi, Kasim and Radu, 2013}; ^{Prajogo, McDermott and McDermott, 2013}; ^{Sirén, Hakala, Wincent and Grichnik, 2017}; ^{Swee Lin Tan, Smyrnios and Xiong, 2014}; ^{Torres and Jasso, 2009}).

In this context, this work explores the following hypotheses.

H3: There are significant differences in the approach to innovation derived from the size of the company.

H4: There are significant differences in the approach to innovation derived from the age of the company.

Figure 1 shows the model that represents the relations presented in the hypotheses, as well as the four dimensions that comprise learning orientation.

Source: Own elaboration

Figure 1 Hypothetical model. 

Method

Results analysis

Measurement model and descriptive analysis of scales

The results of the measurement model, which used the maximum likelihood method, indicate that the model fitted well when performing the chi-squared test, X²(N= 253, df=84) =132.724, p<.001.

For the learning orientation, innovation focus, and performance constructs, the standardized items and loads are specified in Table 4. It can be observed that all of them are significant at the p<.001 level and are above 0.5, which indicates that latent variables are significantly represented by their respective observed variables. From the learning orientation construct, items OA1, OA5, OA6, OA7, OA8, OA9, OA10, and OA11 were eliminated because they had standardized loads lower than 0.5 and in some cases because they were redundant; thus, only the dimensions of commitment to learning and shared intra-organizational knowledge were represented. For the same reasons, in the case of performance, items D1, D5, D7, and D9 were also eliminated.

In the OA scale, it is observed that the highest means greater than 6 with a “strongly agreed” interpretation are for items OA4 and OA3, which refer to “Learning in your organization is seen as a necessary element to ensure the survival of the organization” and “The feeling in the company is that employee learning is an investment not an expense”. As well as OA15 and OA2 that mention that “The basic values of the organization include learning as a key to improvement” and “Top management emphasizes the importance of sharing knowledge in our company”. The rest of the OA items are placed in a mean that is interpreted as “agree”. This reflects that companies claim to be committed to learning and have it clear in the discourse, but the consensus diminishes when they value specific practices of developing collective knowledge capabilities that are essential for organizational survival.

In the innovation variable, the mean decreases, with the highest being of 5,091- interpreted as “in agreement”-for the item “In general, the management of my company favors a strong emphasis on research and development, technological leadership, and innovations”. On the other hand, the item “The company has launched to the market many new lines of products or services in the last five years” reaches an average interpreted as “neither agree nor disagree”. Once again, it is observed that, faced with discourse, consensus decreases when it comes to implementation. Moreover, innovation reaches lower averages than learning; this can be explained by the greater complexity involved in innovating, which requires and produces learning, but the inverse relationship is not always true.

In terms of performance, it is shown that, with the exception of one item (D4), all have an average greater than 5 interpreted as “satisfied” with the aforementioned criterion. The indicator with the lowest average is the one referring to the return on capital of the partners. The reason could be that the key informants were owners or partners, so this question has more subjective connotations for them.

Structural model

The results of the structural equation model, in which the maximum likelihood method was used, indicate that the model fitted well when performing the chi-squared test, X²(N= 253, df=112) =159.149, p<.002.

Additionally, in the case of the incremental goodness of fit indices (see Table 6) IFI, TLI, and CFI are all above 0.9 in a range of 0.976 to 0.980, which is considered acceptable. The same is true of the RMSEA, which has a value of 0.041 and is at a level below the recommended maximum of 0.08 (^{Hair et al., 1999}; ^{Schumacker and Lomax, 2004}; ^{Pesämaa, Shoham, Wincent, and Ruvio, 2013}). Likewise, the results indicate that the weights of the standardized regressions are significant for the test of the t value (>=1.96, p<=.05).

Since it is suggested in the literature that the size and age of the company could have an impact on innovation, these two variables were tested as control variables. No significant effect was found for the age of the company. However, there was a significant positive effect (p<0.05) for the size of the company on the approach to innovation and, considering the standardized regression coefficient (0.182), it can be said that the effect is weak positive (see Table 5).

Beyond the fact that the sample has a great dispersion in the age of the companies measured in years, the reason for the foregoing has to do with the fact that the use of external factors and the development of organizational capacities, which facilitate innovation, are more feasible in larger companies and are less explained by the permanence in the market-time-(^{Torres and Jasso, 2009}). That the relationship is weak could be due to the fact that the variable does not maintain the dimensions related to shared vision and open mindedness. In addition, a lack of systematization of learning practices is suspected, which undermines innovative results.

Based on the previous analysis, the learning, innovation, and performance orientation model was obtained (see Figure 2). Learning orientation explains 8% of the variance of the dependent variable performance and 31% of the approach to innovation. This is due to the direct relationship between learning, knowledge, and innovation, while performance is influenced by many other factors both internal and external to the organization.

Source: own elaboration.

Note: *** corresponds to p = <0.001; ** corresponds to p = <0.05

Figure 2 Results of the structural learning orientation, innovation and performance model. 

Finally, in sum, the findings allows to affirm that empirical evidence was found in favor of the positive relationship between learning orientation and performance and also with innovation; in addition, the latter is influenced by size, but not by the age of the company (see Table 6).

Discussion and conclusions

This research was conducted in the context of an emerging economy, and with a sample of companies including a larger proportion of the service sector (61.7%), a high presence of companies with no more than 50 workers (77%), and an average age of 22.3 years, indicating that they are not young companies, according to the criteria used in the literature (Lin Tan et al., 2014; ^{Coad et al., 2016}; ^{Sirén et al., 2017}).

The main result is that learning orientation explains 8% of the variance of performance and 31% of the approach to innovation, which is in line with empirical literature that reports positive effects of learning on innovation and organizational outcomes (^{Calantone et al., 2002}; ^{Jiménez-Jiménez and Sanz-Valle, 2011}; ^{Alegre and Chiva 2013}; ^{Martínez Serna et al., 2016} and ²⁰¹⁸). Considering the findings, it is also concluded that learning has a greater impact on the focus on innovation than on performance; in this sense, ^{Gomes and Wojahn (2017)} find that organizational learning capacity influences innovation but not performance.

In this model it should not be ignored that the size of the company, measured in number of employees, is significant at a level p<.050, which suggests that the greater the number of employees the greater the focus on innovation (^{Hui et al., 2013}). For its part, the work of ^{Jiménez-Jiménez and Sanz-Valle (2011)} reports that the positive effect of learning orientation on innovation and performance is greater in smaller companies and argues that larger companies have more resources to invest in innovation and this makes them less dependent on the learning approach when innovating. However, we believe that precisely the fact that they have more means facilitates their approach to innovation, which puts microenterprises at a disadvantage. In this sense, ^{Torres and Jasso (2009)} conclude that shortcomings in the development of internal factors related to innovation-such as training, research, and development-limit the survival and growth of SMEs.

According to ^{Jasso (2004)}, the relationship between the technological trajectory of companies and their life cycle goes beyond technology and proposes to consider the market and production, analyzing management and organizational learning capabilities. The literature states that the life cycle does affect entrepreneurial skills (^{Bianchi et al., 2015}; ^{Coad et al., 2016}; ^{Sirén et al., 2017}), but in this work no significant differences have been found in the approach to innovation derived from the age of the company. This fact deserves further analysis in future research, which may include other variables of the business environment, in order to improve the understanding of the phenomenon of the approach to innovation.

Looking closer at the learning orientation, it can be observed that the commitment to learning, together with the tendency to share knowledge within the organization, encourages research and development actions, as well as the market launch of new products or services (standardized load of learning orientation towards the innovation approach 0.527). However, the dimensions of shared vision and open-mindedness are left out. This reflects the weaknesses in the strategic direction of these companies, which do not share expectations and resist change in a little innovative ecosystem that also does not encourage the need for change.

This study contributes to the literature by providing evidence of an emerging economy and of small businesses that have consolidated in their respective markets. In addition, the results have implications at the management level in the sense that these companies must commit to learning if they want to build better capabilities (^{Torres, 2006}) and be more competitive, but they have to review the way in which they currently learn. In this sense, it is necessary that they promote a more innovative vision and implement internal capacity development actions related to the generation and application of knowledge and organizational learning.

With regard to the limitations of the study, its transactional design should be noted, thus care must be taken when interpreting causality in the relationships posed. Furthermore, in practice, the effect of learning on innovation and performance is usually observed with a delay, which is why longitudinal studies are suggested.