An integrated analysis of the impact of gender diversity on innovation and productivity in manufacturing firms: Prepared for the institutions for development sector

Gallego, Juan Miguel; Gutiérrez Urdaneta, Luis H. Working Paper An integrated analysis of the impact of gender diversity on innovation and productivity in manufacturing firms: Prepared for the institutions for development sector IDB Working Paper Series, No. IDB-WP-865 Provided in Cooperation with: Inter-American Development Bank (IDB), Washington, DC Suggested Citation: Gallego, Juan Miguel; Gutiérrez Urdaneta, Luis H. (2018) : An integrated analysis of the impact of gender diversity on innovation and productivity in manufacturing firms: Prepared for the institutions for development sector, IDB Working Paper Series, No. IDB-WP-865, Inter- American Development Bank (IDB), Washington, DC, https://doi.org/10.18235/0000987 This Version is available at: https://hdl.handle.net/10419/208101 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/3.0/igo/legalcode IDB WORKING PAPER SERIES Nº IDB-WP-865 An Integrated Analysis of the Impact of Gender Diversity on Innovation and Productivity in Manufacturing Firms Prepared for the Institutions for Development Sector by: Juan Miguel Gallego Luis H. Gutierrez Inter-American Development Bank Institutions for Development Sector January 2018 January 2018 An Integrated Analysis of the Impact of Gender Diversity on Innovation and Productivity in Manufacturing Firms Prepared for the Institutions for Development Sector by: Juan Miguel Gallego Luis H. Gutierrez Inter-American Development Bank. This work is licensed under a Creative Commons IGO 3.0 Attribution- NonCommercial-NoDerivatives (CC-IGO BY-NC-ND 3.0 IGO) license (http://creativecommons.org/licenses/by-nc-nd/3.0/igo/ legalcode) and may be reproduced with attribution to the IDB and for any non-commercial purpose, as provided below. No derivative work is allowed. Any dispute related to the use of the works of the IDB that cannot be settled amicably shall be submitted to arbitration pursuant to the UNCITRAL rules. The use of the IDB's name for any purpose other than for attribution, and the use of IDB's logo shall be subject to a separate written license agreement between the IDB and the user and is not authorized as part of this CC-IGO license. Following a peer review process, and with previous written consent by the Inter-American Development Bank (IDB), a revised version of this work may also be reproduced in any academic journal, including those indexed by the American Economic Association's EconLit, provided that the IDB is credited and that the author(s) receive no income from the publication. Therefore, the restriction to receive income from such publication shall only extend to the publication's author(s). With regard to such restriction, in case of any inconsistency between the Creative Commons IGO 3.0 Attribution-NonCommercial-NoDerivatives license and these statements, the latter shall prevail. Note that link provided above includes additional terms and conditions of the license. The opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the Inter-American Development Bank, its Board of Directors, or the countries they represent. Cataloging-in-Publication data provided by the Inter-American Development Bank Felipe Herrera Library Gallego, Juan Miguel. An integrated analysis of the impact of gender diversity on innovation and productivity in manufacturing firms / Juan Miguel Gallego and Luis H. Gutierrez p. cm. — (IDB Working Paper Series ; 865) Includes bibliographic references. 1. Women industrialists-Colombia. 2. Diversity in the workplace-Colombia. 3. Manufacturing industries-Technological innovations-Colombia. 4. Industrial productivity-Colombia. I. Gutierrez, Luis H. II. Inter-American Development Bank. Competitiveness, Technology and Innovation Division. III. Title. IV. Series. IDB-WP-865 http://www.iadb.org Copyright © 2018 Abstract* This paper presents evidence of the effects of gender diversity on firm innovation outcomes and their productivity in Colombian manufacturing firms, by extending a CDM model to include women’s participation in science, technology and innovation (STI) activities and production processes. The paper makes a methodological contribution by taking into account potential endogeneity issues of women’s participation in STI and innovation behavior. An upward bias of the impact of gender diversity on innovation might arise because more dynamic and innovative managers, knowing the value of gender diversity, may also hire more women in STI. The paper addresses the endogeneity concerns with a Tobit specification of the firm’s decision to employ women in STI by instrumenting it with the share of total women in the workforce in the industry or the region where the firm is located. The main results indicate that firms with a larger share of women in the knowledgecreation and innovation process might increase their innovative behavior. It also presents evidence of a differentiated effect of gender diversity by type of innovation. Women’s participation has a larger effect on technological innovation than on organizational innovation. Finally, gender diversity drives firm productivity, even after controlling for the effect of innovation on the production process. These results are important for developing countries where women’s participation in STI activities is extremely low. In Colombia, for example, only 6 percent of STI employees are women, a very low figure compared with the 21 percent participation of women in industry and even lower compared with the 52 percent of women in Colombia’s total work force. These figures and this paper’s results should open up discussion on policies to promote women’s participation in the labor force in manufacturing firms, specifically in the knowledge creation processes. JEL codes: J16, J24, J82, O30, O31 Keywords: CDM, diversity, gender, innovation, women * Authors’ affiliation: Department of Economics, Universidad del Rosario. Calle 12c #4-59, Bogota, Colombia. This paper is part of the research project, Science, Technology and Innovation Gender Gaps and their Economic Costs in Latin America and the Caribbean, led by the Competitiveness, Technology, and Innovation Division of the Inter-American Development Bank (IDB) and financed by the IDB Gender and Diversity Fund. The information and opinions presented here are entirely those of the authors, and do not imply endorsement of the IDB, its Board of Executive Directors, or the countries they represent. We are indebted to Diego Aboal, Roberto Álvarez, Matteo Grazzi, Jacques Mairesse, Philip E. Keefer, Jocelyn Olivari, and Janet Stotsky for their comments on earlier versions of this paper. The information was provided by the National Bureau of Statistics (Departamento Administrativo Nacional de Estadísticas-DANE) and the information was processed in the workplace provided by this institution. ! 2 1. Introduction Can the presence of women be a factor that contributes to a firm’s productivity and propensity to innovate? Research on the impact of women’s participation on firm performance and innovation has produced mixed results. Some studies that use employer-employee or firm-level datasets have found that women’s participation in research and development (R&D) teams increases the likelihood that a firm innovates (Díaz, González, and Sáez, 2013; Fernández, 2015; García, Zouaghi and García, 2017; Østergaard, Timmermans, and Kristinsson, 2011; Teruel, Parra and Segarra, 2015). Other studies find that the empirical evidence of the impact of gender diversity on firm productivity is mostly negative (e.g., Parrotta, Pozzoli, and Pytlikova, 2014), although some (e.g., Garnero, Kampelmann, and Rycx, 2014) find a positive relationship. The involvement of women in knowledge creation and production processes within firms has interested both academics and experts. For example, renowned consulting companies such as McKinsey Global Institute (MGI) and specialized business magazines have recently called attention to the importance of women in the firms’ labor force. MGI argues that advancing women’s equality in the workplace could add as much as $12 trillion of additional global annual GDP worldwide by 2025, and a recent insight by Forbes (2015) concludes “... a diverse and inclusive workforce is necessary to drive innovation and foster creativity, and guide business strategy.” We present evidence about the effects of gender diversity on innovation outcomes and their productivity in Colombian manufacturing firms. Despite recent results on the virtuous relationship between innovation investments, innovative behavior, and productivity in developing economies, no research has delved into the extent to which having women involved in innovation activities might change the pace of both how firms innovate and their expected gains in productivity (Crespi and Zuñiga, 2012; Crespi, Taczir and Vargas, 2016; Gallego, Gutiérrez and Taborda, 2015). Specifically, we examine the impact of the presence of women involved directly in STI activity teams on the probability that a firm innovates. In addition, we link the predicted innovation behavior, controlling for women in STI activities and R&D expenditures, with gender diversity and labor productivity to emphasize the importance of women on productivity within firms. We argue that an integrated view is well-suited to capture the contribution of gender diversity both to the innovation and production processes. The main results indicate that firms with a larger share of women in the knowledge creation and innovation process might increase their innovative behavior and, consequently, may exhibit greater labor productivity. An increase of six percentage points in the share of women in STI would increase the innovative behavior of firms by two percentage points. In addition, we find evidence of a differentiated effect of gender diversity by type of innovation: women’s participation affects ! 3 three times more technological innovation outcomes than organizational innovation. In particular, an increase of six percentage points would increase technological innovation by three percentage points and the organizational innovations of firms by less than one percentage point. In addition, we ran regressions where the mode of innovation is whether the firm achieved incremental or radical innovations (that is, innovations new to the market); interestingly, women’s inclusion in STI activities has a similar effect on either type of innovations. As Page (2017) argues, the diversity premium is stronger in complex, non-routine, and cognitive tasks precisely the group problemsolving contexts. Technological innovation, specifically radical innovations, are by nature complex and based upon knowledge. The more cognitively diverse the teams, in our case in terms of gender, the stronger the effects. STI teams within firms obtain that premium if they are more diverse. Hence, the effect of diversity is stronger on technological than on non-technological innovation. These results are new evidence of the positive impact of having a more genderdiversified workforce on knowledge production processes in developing countries, whose economies exhibit low ratios of women in STI. In Colombia, for example, women’s participation in manufacturing is only around 6 percent of the total number of employees in STI activities, a very low figure compared with 21 percent of the total workforce in industry and even lower compared with the 52 percent of the women in Colombia’s total workforce. Furthermore, we make a methodological contribution by including an integrated analysis of the effect of gender diversity on the relationship among innovation investments, innovation outcomes, and productivity in a CDM model (Crepón, Duguet, and Mairesse, 1998), taking into account potential endogeneity issues of women’s participation in STI and innovation behavior. An upward bias of the impact of gender diversity on innovation might arise because more dynamic and innovative managers, knowing the value of gender diversity, will also hire more women for STI. We address the endogeneity concerns with a Tobit specification of the firm’s decision to employ women in STI, by instrumenting it with the share of total women in the workforce in the industry and in the region where the firm is located. We then include the predicted women’s participation in STI activities at the CDM model’s second stage. By addressing the endogeneity of a firm’s decision to hire (more) women in STI teams, we identify an unbiased effect of women on the knowledge equation of the CDM. Finally, in the third stage, we estimate a Cobb-Douglass production function augmented with gender diversity measures such as the share of women in the firm’s total employees and two indicators (i.e., coefficient of variation and standard deviation) of women’s participation in the different departments of the firm. We also contribute to the current discussion about the function of team collaboration in knowledge creation. Innovation activities are complex processes that need, as the main input, a ! 4 dedicated team of employees to perform very specialized tasks. A more diverse labor force may benefit firm performance if it encourages complementarities, but these beneficial effects can be offset by the potential communication costs that might arise (Lee and Farh, 2004; Prat, 2002). Our results show that the inclusion of gender diversity could produce value gains for manufacturing firms in Colombia, an industry with low female participation, and potential gains in terms of new products and higher productivity at the firm level. The remainder of the article is organized as follows. Section 2 presents the analytical framework that guides this research as well as the working hypothesis. Section 3 provides details on the data, the sources of information and empirical methodology we employ to test our hypotheses. Section 4 presents the results, and Section 5 concludes. 2. Theoretical Framework and Hypotheses We analyze an integrated view of the relationship between women’s participation in the workplace and innovation and firm performance. This paper is related to two branches of literature: on innovation and productivity, and diversity. With respect to the former, researchers from different fields largely acknowledge that innovation is a key driver of firm performance and economic development (Fagerberg, Srholec, and Verspagen, 2010; Nelson, 2005). At the firm level, studies using the CDM structural framework have confirmed that enterprises that are more intensive in innovation investment have a higher propensity to be innovative and that firms that innovate more are also more productive (Crespi and Zuniga, 2012; Gallego, Gutiérrez and Taborda, 2015; Griffith et al., 2006). Most of this literature has identified key determinants of innovation efforts and innovation, but the inclusion of workforce diversity, specifically gender diversity, as a critical aspect in innovative behavior has been absent from an integrated analysis using the CDM model. However, studying the role of women in the workplace and whether their presence can be a factor that contributes to a firm’s propensity to innovate and its productivity are relevant questions that we assess within the framework of CDM models. To the best of our knowledge, no research has followed this approach. With regard to the role of diversity in the workplace, there is an extensive literature on the positive or negative consequences of having a more diversified labor force. The general idea is that firms seek to bring products and services to the firm or to markets and to implement production processes. These tasks are usually carried out by teams of workers or by specialized people instructed to perform them. Teams may be diverse in terms of age, gender, race, education, or nationality. These characteristics set in motion different economic and social forces that may affect the functioning of teams. Some scholars have argued that diversity in the labor force, and perhaps ! 5 more so in specific teams, can be beneficial for firm performance if certain conditions are present and if there are complementary skills (Alesina and La Ferrara, 2005; Grund and Westergaard- Nielsen, 2008; Lazear 1999; Osborne, 2000; Parrotta, Pozzoli, and Pytlikova, 2014; Skirbekk, 2003). However, diversity may also have some disadvantages, since people may face greater communication and cooperation problems. In some cases, the costs of diversity outweigh the benefits ( Lee and Farh, 2004; Prat, 2002). One important dimension of workforce diversity is more equal participation of women in the workplace. In this respect, gender diversity has been considered a key factor not only in the production process but also in the knowledge creation process, with strong evidence in the case of developed economies. However, it is surprising that scant research has been done that provides evidence of whether women’s involvement in firm innovation endeavors has positive or negative effects for emerging economies. There is even less evidence, if any, for any Latin American country. Since our contribution is to explore women’s participation in innovation and production processes in an integrated way, it is important to understand how the literature on gender diversity has assessed the role of women in innovation and production. Østergaard, Timmermans, and Kristinsson (2011) have studied how women’s participation can affect the propensity of firms to innovate. Using an employer-employee dataset, the authors analyzed the effect of gender (and age) diversity and human capital on innovation in a sample of Danish firms using a diversity measure, the Shannon–Weaver entropy index. They found a strong positive and significant relationship between gender diversity and innovation, which indicates that a gender-diverse workforce composition is positively associated with the likelihood of introducing an innovation. Pfeifer and Wagner (2014) also found a positive correlation in their research on German manufacturing firms. They found that women’s participation in a firm’s labor force contributed to higher profits as well as more investment in R&D activities. Using data on Spanish firms, researchers have studied the effect of gender diversity in R&D teams on innovation (Díaz, González, and Sáez, 2013; Fernández, 2015a; García, Zouaghi and García, 2017; Teruel, Parra, and Segarra, 2015). The focus of their research was to study the relationship between gender distribution of R&D groups and the propensity to achieve radical, incremental, or any type of innovation. In general, these studies found that diversity exerted positive effects on the probability of introducing any of those types of innovations. In developing countries, it would be important to analyze not only radical and incremental technological innovation, but also non-technological (i.e., organizational and marketing) innovation, since this type of innovation is particularly important in Latin American firms (Crespi and Zuñiga, 2012). ! 12 only 5 percent approached research centers or used scientific sources to find information for their innovation activities. Table 1 presents the main statistics of women’s participation in STI, the whole workforce, and the two diversity measures. First is direct women’s involvement in STI activities, which was surprisingly low. It only amounted to 6.0 percent of the total employees on STI activities. However, women’s involvement in the entire manufacturing workforce averaged 21 percent. Both indicators, especially the first one, clearly show that there is ample room for firm managers to hire more women. The means of the coefficient of variation and standard deviation coefficient are 0.5 and 0.42, respectively. These two last indicators confirm that the gender distribution in manufacturing firms is still unequal. 4. Results This section provides empirical evidence of the relationship of having a diversified gender workforce in Colombian manufacturing industries on the likelihood that a firm innovates and on the value-added per worker. In this way, we will test the two hypotheses proposed above. 4.1. How Women’s Involvement in STI Activities Contributes to Innovation As was shown in Table 1, the share of women working in STI areas is relatively low (6 percent compared to 21 percent in the overall labor force at the industry, on average) but their contribution to a firm’s propensity to innovate can only be assessed empirically. We will focus our analysis on our main variables of interest and on the predicted R&D. The results of the other coefficients are presented in Table 2. Table 2 presents four panels, each of them presenting the marginal effects of explanatory variables on the propensity of innovate in: (i) any mode, (ii) technological innovations, (iii) non-technological types, (iv) radical innovations, and (v) imitation ones. Furthermore, in each panel, we first present the results of the second stage of the CDM model assuming non-endogeneity of our variable of interest by including directly the share of women working in STI activities, followed by the results when we instrument that variable. ! 13 Table 2. The Knowledge Production Function - Biprobit Regressions - Marginal Effects (Stage 2) Source: Authors’ elaboration. Notes: Bootstrapping errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1. All regressions include dummies controlling for region, year, and ISIC-2 digits. One outstanding result across all variables through what we call naive specification, or the IV regressions that control for potential sources of endogeneity of women’s participation in STI activities, is that regardless of the mode of innovation, firms with greater women’s participation in STI activities (𝑠𝑤_𝑆𝑇𝐼) had a higher propensity to innovate.1 It confirms hypothesis 1, which states that gender diversity increases the likelihood of being innovative. We also provide evidence of a differentiated effect of gender diversity by type of innovation, as was stated in hypothesis 1a. Women’s participation has a positive and statistically significant effect on technological innovation outcomes that doubles the effect found in non-technological ones. In our data, that type of innovation includes both radical and incremental innovations, which were also found to be positive and significantly affected by gender diversity in another empirical context (Díaz, González, and Sáez, 2013; Fernández, 2015a and 2015b; García, Zouaghi and García, 2017; Teruel, Parra and Segarra, 2015).2 Not controlling for potential endogeneity issues of women’s participation in STI activities led to an overestimation of the effect of that participation. The overestimation effects were greater proportionally for non-technological innovation regressions. These results provide 1 The results of the Tobit regression used to control for the potential endogeneity of the gender variables are presented in Table A2 (see Annex). In Table A3 (see Annex), we also present results for regressions checking the validity of our instrument. It runs the effect of our instrument the women share on STI and the industry and region level on the three types of firm innovation. Clearly, after controlling for sector, industry and year effect, women share at industry region level does not affect any mode of firm’s innovation outcomes. 2 The EDIT survey allows separating the measure of technological innovation between those innovations that were introduced by the firm as new to the firm and new to the market. Table 2. The Knowledge Production Function - Biprobit Regressions - Marginal Effects (Stage 2) Variable Naive IV Naive IV Naive IV Naive IV Naive IV RD_p (Predicted R&D expenditure per employee) 0.079*** 0.102*** 0.14*** 0.175*** 0.08*** 0.097*** 0.06*** 0.07*** 0.071*** 0.09*** (0.044) (0.033) (0.005) (0.0063) (0.003) (0.004) (0.003) (0.004) (0.012) (0.0032) Size 0.02*** 0.003 0.022*** 0.012 0.009*** -0.007 0.012*** -0.011 0.011*** -0.014 (0.0013) (0.009) (0.002) (0.0012) (0.002) (0.007) (0.0013) (0.010) (0.0012) (0.012) Log_export per employee -0.033*** -0.054*** -0.064*** -0.095*** -0.038*** -0.053*** -0.022*** -0.037*** -0.026*** -0.042*** (0.0056) (0.0045) (0.0063) (0.008) (0.0063) (0.005) (0.006) (0.006) (0.0033) (0.005) Human capital 0.0072 0.0018 0.031 0.0025 0.011 0.016 0.019** 0.023*** 0.026 0.002 (0.0085) (0.0013) (0.012) (0.0015) (0.013) (0.017) (0.008) (0.01) (0.018) (0.011) Cooperation -0.0088 -0.0053 -0.021** -0.018 -0.038*** -0.035*** -0.012 -0.01 -0.017** -0.02 (0.0084) (0.006) (0.0094) (0.014) (0.0085) (0.010) (0.009) (0.0097) (0.026) (0.011) Foreign ownership > 10% -0.034*** -0.04*** -0.056*** -0.067*** -0.026*** -0.0302*** -0.029*** -0.033*** -0.03*** -0.035*** (0.0067) (0.0084) (0.0095) (0.01) (0.0063) (0.0057) (0.0065) (0.006) (0.010) (0.011) sw_STI 0,203*** 0,26*** 0,15*** 0,11*** 0,13*** (0.0115) (0.011) (0.007) (0.007) (0.006) sw_STI_hat 0,106*** 0,16*** 0,075*** 0,10** 0,11** (0.038) (0.049 (0.025) (0.047) (0.04) Constant Observations 17055 17055 17051 17051 17050 17050 17020 17020 17055 17055 Imitation INN TI NTI Radical ! 14 new evidence of the positive impact of having a more gender-diverse workforce in knowledge production undertakings in developing countries, whose economies exhibit low ratios of women working in STI activities. To further explore the robustness of these findings, we ran regressions where the mode of innovation is whether the firm achieved incremental or radical innovation. The findings are presented in Table 2. Interestingly, the inclusion of women in STI activities seems to affect similarly either mode of innovation. As Page (2017) argues, the diversity premium is stronger in complex, non-routine, cognitive tasks, which are the group problem-solving contexts. In more cognitively diverse teams, the effect of diversity, in this case gender diversity, is stronger. STI teams within firms gain their bonus if they are more diverse. Hence, the effect is stronger on technological than on non-technological innovation. Finally, Table 2 confirms the importance of the (predicted) intensity of R&D investment. The results show a positive and significant correlation between R&D investment and the probability that a firm innovates, even after controlling for gender diversity. Of the remaining control variables, human capital has a positive sign for any type of innovation, but it is only statistically significant for radical innovations. The presence of foreign ownership and whether a firm cooperated with internal and external partners both have a negative impact on the likelihood that a firm innovates. In the first variable, foreign companies undertake innovation efforts at their headquarters, which may explain their nil impact. With respect to the second variable, more research is warranted. Being an exporting firm decreases the chances of innovating. This result calls for more exploration of the way the indicator is built; for example, the geographic destination or R&D intensity of goods (Crespi and Zuñiga, 2012) should be considered. Last, the effect of size was not statistically significant in any of the IV-controlled regressions. How do the findings of gender diversity compare to those found in the literature? This result is in the same vein with the findings of recent research that have tested whether having gender diversity is beneficial for getting more innovation. We have found a similar positive effect in an integrated analysis, despite using direct women’s share in STI activities instead of a diversity index. For example, Díaz et al. (2013) found that gender diversity in R&D team contributes positively to a firm’s achievement of radical innovations. Fernández (2015a and 2015b) found a positive contribution of gender diversity on four modes of innovations. Finally, Teruel et al. (2015) and García et al. (2017) also found a positive contribution of greater gender diversity on a firm’s propensity to innovate either in product, process, marketing, or organizational innovations, or to ! 15 make incremental or radical innovations.3 Consequently, we have shown that for a developing economy, having (more) women directly employed in STI activities brings ideas, insights, and creativity that enhance a firm’s propensity to be more innovative. 4.2. Women’s Contribution to Productivity We have found that women’s engagement in STI activities within the Colombian manufacturing firms contributed importantly to the propensity of firms to innovate. But can women’s participation in the whole firm, i.e., within any functional domain, also be a contributing factor to enhancing firm productivity (our second hypothesis)? This question is answered in Table 3, which shows the result of the econometric specification of equation (2). We recall that equation (2) is an augmented Cobb- Douglas production function that includes common inputs, capital, labor, human capital, plus a measure of gender diversity in the overall firm labor force and the predicted innovation results obtained in stage 2. Before presenting and analyzing the main findings, some explanations about Table 3 are in order. First, the estimation of the production function is the third stage of a structural model in which we first estimate the intensity of R&D investment (reported in Table A1 in the Annex), followed by introduction of the predicted value into the knowledge production function, which in turn is calculated for three measures of innovations, that is, the innovative firm (INN), the realization of technological innovations (TI), and the achievement of non-technological innovations (NTI) (Table 2). The estimation of these three innovation measures was made using the gender variable of interest at stage 2, that is, women’s participation in STI activities. These predicted values, plus the standard inputs of a production function, led to the results in Table 3. To illustrate, columns 1 to 4 present the results of the standard CDM augmented by the predicted innovation variable –𝐼𝑁𝑁, plus the direct contribution of three measures of women’s participation in firm productivity (Columns 2 to 4). Columns 5 to 7, and columns 8 to 10, replace the 𝐼𝑁𝑁 variable with 𝑇𝐼 and then with 𝑁𝑇𝐼 plus the contribution of the main variables of interest (women’s participation in the overall workforce of the firm, the coefficient of variation, and the standard deviation). 3 Most papers use the Blau Index as a measure of gender diversity because it accounts for the distribution of women on different teams involved on STI activities (Harrison and Klein, 2007). The information about the distribution of women on different teams on innovation activities is not available on the EDIT survey in Colombia. We measure the role of women on innovation outcomes by the ratio of women participating in STI. ! 16 Table 3. Impact of Innovation and Gender on Labor Productivity Source: Authors’ elaboration. Notes: Robust standard errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1. All regressions include dummies controlling for region, year, and ISIC-2 digits. We begin by focusing on our variables of interest, that is, the share of women in the overall labor force of a firm, and the two diversity measures. First, regardless of the mode of (predicted) innovation obtained by firms, the higher the share of women in the overall labor force of a firm, the greater its labor productivity. Second, regarding the two gender diversity measures, only the standard deviation seems to contribute to increased labor productivity, and the results do not depend on the mode of innovation achieved by the firm. When we utilize the coefficient of variation instead of the standard deviation, we find that this diversity measure impacts productivity when the innovation outcome is defined broadly, that is, either type of innovation. No (significant) effect of this indicator on firm’s labor productivity was found when we only used technological or nontechnological innovation. Our second hypothesis says that a more gender diverse workforce, or a labor force in which more women participate, can offer a broader pool of skills and capabilities which can manifest itself in greater productivity. Thus, the main variable of interest in Table 3 is women’s overall participation within the firm (Columns 2, 5, and 8 in Table 3). Consequently, we are assuming that regardless of the functional area where women work, their impact as a whole should be positive. Indeed, this is what we found in those three columns. This result is contrary to some results found in the labor research of the impact of gender on productivity (and wages). For Table 3. Impact of Innovation and Gender on Labor Productivity (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) Capital stock per employee 0,185*** 0,186*** 0,185*** 0,185*** 0,185*** 0,185*** 0,185*** 0,185*** 0,185*** 0,185*** (0.008) (0.010) (0.010) (0.009) (0.010) (0.009) (0.009) (0.008) (0.010) (0.010) Size 0,122*** 0,120*** 0,121*** 0,119*** 0,111*** 0,113*** 0,111*** 0,114*** 0,115*** 0,113*** (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.007) (0.006) Age -0,0459*** -0,0453*** -0,0444*** -0,0458*** -0,0499*** -0,0495*** -0,0500*** -0,0509*** -0,0506*** -0,0510*** (0.012) (0.011) (0.012) (0.010) (0.011) (0.014) (0.010) (0.014) (0.011) (0.011) Human capital 0,476*** 0,480*** 0,475*** 0,477*** 0,480*** 0,476*** 0,477*** 0,444*** 0,442*** 0,444*** (0.039) (0.043) (0.050) (0.036) (0.050) (0.042) (0.041) (0.044) (0.050) (0.037) Share_women in firm 0,048*** 0,041** 0,041** (0.017) (0.018) (0.019) Coefficient of variation 0,022** 0,079*** 0,0070 (0.006) (0.008) (0.009) Standard deviation 0,0094** 0,0086** 0,0085*** (0.004) (0.004) (0.004) INN (Predicted)- sw_STI 0.437*** 0.437*** 0.383*** 0.405*** (0.042) (0.037) (0.049) (0.045) TI (Predicted) - sw_STI 0.102*** 0.096*** 0.095*** (0.009) (0.015) (0.011) NTI (Predicted) - sw_STI 0.147*** 0.14*** 0.137*** (0.01) (0.02) (0.02) Constant 3.28*** 3,27*** 3,28*** 3,29*** 3.45*** 3.44*** 3.45*** 3.54*** 3.54*** 3.54*** (0.09) (0.09) (0.10) (0.09) (0.10) (0.10) (0.10) (0.09) (0.12) (0.11) Observations 17055 17055 17055 17055 17067 17067 17067 17067 17067 17067 R-squared 0.223 0.223 0.224 0.224 0.224 0.224 0.225 0.224 0.224 0.225 ! 17 example, Ilmakunnas and Ilmakunnas (2011) found that the influence of women’s participation in total factor productivity had a negative sign but had become insignificant in fixed effects estimation. However, in their GMM specification, the effect is definitely negative. On the other hand, Garnero et al. (2014), in their study of Belgian companies, found that the diversity index of gender was negatively related to both average labor productivity and average wages when taking the whole industry into account, while gender diversity increased productivity when looking only at high-tech industries. In our case, we only tested the effect of women’s participation in all manufacturing sectors. Contrary to Ilmakunnas and Ilmakunnas (2011) and Garnero, Kampelmann, and Rycx (2014), the result was that women’s participation supports productivity. Hypothesis two is therefore accepted. Furthermore, the positive contribution of either mode of innovation achieved by firms to firm labor productivity is consistent with findings in related research for both industrialized and emerging economies. The effects are large and significant. Finally, two results that are worth highlighting refer to the coefficients of human capital and age. The first one is positive, meaning that the higher the skills of the (overall) labor force of a firm, the higher its productivity. The second one is negative, meaning that younger firms, when controlling for innovation and women’s participation, are more productive than mature firms. 5. Concluding Remarks This study has tested two hypotheses regarding the contribution of women’s presence in a firm’s labor force and has found a positive relationship between measures of gender diversity and innovation and productivity. Using data from two innovation surveys and performance surveys for Colombian manufacturing firms for the period 2011–2014 and studying the role of women using an integrated view of a CDM framework as the econometric technique, we found some robust evidence of a positive contribution of women’s participation in STI activities to firm innovation outcomes. Hypothesis 1 is accepted, since it established a positive association between women’s engagement in STI activities and firms’ achievement of innovations. The results in Table 2 also confirm hypothesis 1a, that is, that women’s participation in STI activities impacts differentially the mode in which firms innovate. It has a positive and significant contribution to the likelihood that firms introduce technological innovations but seems not to affect firms’ propensity to achieve nontechnological innovation. Furthermore, using the same data and model, we found a positive contribution of women’s share in overall personnel and of gender diversity measures to labor productivity. In this case, hypothesis 2 is also accepted. ! 18 The results of this research clearly show that women’s participation in the manufacturing labor force, and in scientific, technological, and innovation activities is of foremost importance. The results also bolster the case made by McKinsey about the value women’s participation can add to annual GDP growth. This value could be greater for developing countries that face the challenge of providing better living conditions and more sustainable growth and development and where women’s involvement in STI activities is low. In terms of the cost to firms that do not have (more) women in the workplace, our research can indirectly provide some insights. It is well known that innovation is the key driver of firm productivity, and that for firms to keep or increase their market share, they will begin, or continue, to be more innovative. Since technological innovations (product and process ones) move a firm’s marginal cost function to the right, that is, a reduction of cost for each unit of good produced, innovation can result in larger gains. Finally, we found evidence that a more gender-diverse labor force enhances firm productivity. Thus, firms have lower labor productivity when their labor force tends to be less diverse. Future research can provide us with more accurate results and with them to see the costs that a firm incurs from not having a more balanced labor force. These calculations are left for future research. ! 19 References Adams, R. B., and D. Ferreira. 2009. “Women in the Boardroom and their Impact on Governance and Performance.” Journal of Financial Economics 94(2): 291–309. Alesina, A., and E. La Ferrara. 2005. “Ethnic Diversity and Economic Performance.” Journal of Economic Literature 43(3): 762–800. Becker, W. and J. Dietz. 2004. “R&D Cooperation and Innovation Activities of Firms--Evidence for the German Manufacturing Industry.” Research Policy 33: 209–23. Campbell, K. and A. Mínguez. 2008. “Gender Diversity in the Boardroom and Firm Financial Performance.” Journal of Business Ethics 83(4): 435–51. Crepón, B., E. Duguet, and J. Mairesse. 1998. “Research, Innovation and Productivity: An Econometric Analysis at the Firm Level.” Economics of Innovation and New Technology 7(2): 115–58. Crespi, G. and P. Zuniga. 2012. “Innovation and Productivity: Evidence from Six Latin American Countries.” World Development 40(2): 273–90. Crespi, G., E. Tacsir, and F. Vargas. 2016. “Innovation Dynamics and Productivity: Evidence for Latin America.” in Grazzi M. and C. Pietrobelli (eds.) Firm Innovation and Productivity in Latin American and the Caribbean. Washington, D.C.: Inter-American Development Bank. Czarnitzki, D. and C. Lopes. 2013. “Value for Money? New Microeconometric Evidence on Public R&D Grants in Flanders.” Research Policy 42(1): 76–89. Díaz, C., A. González, and F. J. Sáez. 2013. Gender Diversity within R&D Teams: Its Impact on Radicalness of Innovation. Innovation: Management, Policy and Practice 15: 149–60. Fagerberg, J., M. Srholec, and M. Verspagen. 2010. “Innovation and Economic Development. In Hall B. H. and N. Rosenberg (eds). Handbook of the Economics of Innovation. Volume 2. Amsterdam: Elsevier. Fernández, J. 2015a. “The Impact of R&D Teams' Gender Diversity on Innovation Outputs.” International Journal of Entrepreneurship and Small Business 24(1): 142–62. Fernández, J. 2015b. “The Impact of Gender Diversity in Foreign Subsidiaries’ Innovation Outputs.” International Journal of Gender and Entrepreneurship 7(2): 148–67. Forbes. 2015. “Global Diversity and Inclusion: Fostering Innovation through a Diverse Workforce.” Forbes Insights. New York, NY: Forbes. Garcia, M., F. Zouaghi, and T. Garcia. 2017. “Diversity is Strategy: The Effect of R&D Team Diversity on Innovative Performance. R&D Management 47: 311–29. Gallego J. M., L. H. Gutiérrez, and R. Taborda. 2015. “Innovation and Productivity in the Colombian Service and Manufacturing Industries.” Emerging Markets Finance and Trade 51(3): 612–34. Garnero, A., S. Kampelmann, and F. Rycx. 2014. “The Heterogeneous Effects of Workforce Diversity on Productivity, Wages and Profits.” Industrial Relations 53(3): 430–77. ! 20 _____. 2016. “Is Workforce Diversity Always Performance-enhancing? A Literature Review.” Reflets et Perspectives 55(4): 81–91. Gómez, J, I. Salazar, and P. Vargas. 2016. “Sources of Information as Determinants of Product and Process Innovation.” PLoS ONE 11(4): e0152743. doi:10.1371/journal.pone.0152743. Griffith, R., E. Huergo, J. Mairesse, and B. Peters. 2006. “Innovation and Productivity across Four European Countries.” Oxford Review of Economic Policy 22(4): 483–98. Grund, C. and N. Westergaard-Nielsen. 2008. “Age Structure of the Workforce and Firm Performance.” International Journal of Manpower 29(5): 410–22. Hall, B. 2010. “The Financing of Innovative Firms.” Review of Economics and Institutions 1(1 Spring): 1–30. Hall, B. and J. Lerner. 2010. “The Financing of R&D and Innovation.” In: Hall, B. and N. Rosenberg (eds.). Handbook of the Economics of Innovation. North Holland: Elsevier. Harrison, D. A., and K. Klein. 2007. “What’s the Difference? Diversity Constructs as Separation, Variety, or Disparity in Organizations.” Academy of Management Review 32(4): 1199–228. Hoogendoorn, S., H. Oosterbeek, and M. van Praag. 2013. “The Impact of Gender Diversity on the Performance of Business Teams: Evidence from a Field Experiment.” Management Science 59(7): 1514–28. Ilmakunnas, P. and S. Ilmakunnas. 2011. “Diversity in the Workplace: Whom Does it Benefit?” The Economist 159(2): 223–55. Kugler, M. and E. Verhoogen. 2009. “Plants and Imported Inputs: New Facts and an Interpretation.” The American Economic Review 99(2): 501–07. _____. 2012. “Prices, Plant Size, and Product Quality.” The Review of Economic Studies 79(1): 307– 39. Lazear E. 1999. “Culture and Language.” Journal of Political Economy Supplement: 95–125. Lee, C. L., and J. L. Farh. 2004. “Joint Effects of Group Efficacy and Gender Diversity on Group Cohesion and Performance.” Applied Psychology 53(1): 136–54. Mckinsey & Company. 2015. “The Power of Parity: How Advancing Women’s Equality Can Add $12 Trillion to Global Growth.” September. Moore, W., A. F. Presbitero, and R. Rabelotti. 2017. “The Gender Gap in the Caribbean: The Performance of Women-led Firms, in Exploring Firm-Level Innovation and Productivity in Developing Countries.” In: Dohnert, S., G. Crespi, and A. Maffioli (eds). The Perspective of Caribbean Small States. Washington, D.C.: Inter-American Development Bank. Nelson R. 2005. Technology, Institutions, and Economic Growth. Cambridge, MA: Harvard University Press. ! 21 Osborne, E. 2000. “The Deceptively Simple Economics of Workplace Diversity.” Journal of Labor Research 21: 463–75. Østergaard, C. R., B. Timmermans, and K. Kristinsson. 2011. “Does a Different View Create Something New? The Effect of Employee Diversity on Innovation.” Research Policy 40(3): 500–09. Parrotta, P., D. Pozzoli, and M. Pytlikova. 2014. “Labor Diversity and Firm Productivity?” European Economic Review 66: 144–79. Pfeifer, C. and J. Wagner. 2014. “Is Innovative Firm Behavior Correlated with Age and Gender Composition of the Workforce? Evidence from a New Type of Data for German Enterprises.” Journal for Labour Market Research 47(3): 223–31. Prat, A. 2002. “Should a Team Be Homogeneous?” European Economic Review 46(7): 1187–207. Reguera, N., P. de Fuentes, and J. Laffarga. 2017. “Does Board Gender Diversity Influence Financial Performance? Evidence from Spain.” Journal of Business Ethics 141(2): 337–50. Robinson, G. and K. Dechant. 1997. “Building a Business Case for Diversity.” The Academy of Management Executive 11(3): 21–31. Srinidhi, B., F. A. Gul, and J. Tsui. 2011. “Female Directors and Earnings Quality.” Contemporary Accounting Research 28: 1610–44. Skirbekk, V. 2003. “Age and Individual Productivity: A Literature Survey, Technical Report 2003-028.” Rostock: Max Planck Institute for Demographic Research. Stahl, G., M. Maznevski, A. Voigt, and K. Jonsen. 2010. “Unraveling the Effects of Cultural Diversity in Teams: A Meta-analysis of Research on Multicultural Work Groups.” Journal of International Business Studies 41: 690–709. Temel, S., A. L. Mention, and M. Torkkeli. 2013. “The Impact of Cooperation on Firms’ Innovation Propensity in Emerging Economies.” Journal of Technology Management and Innovation 8(1): 54–64. Teruel, M., M. D. Parra, and A. Segarra. 2015. “Gender Diversity and Innovation in Manufacturing, Document de treball n.15. Departamento de Economia, Centro de Investigacion en Economía Industrial y Economía Pública (CREIP). Universitat Rovira I Virgili, Tarragona.