Determinants of Innovation and Productivity in the Service Sector in Mexico

Abstract

Drawing on survey data about firms in Mexico, we investigate the determinants of innovation and the linkages between innovation and productivity in the services sector. We apply a three-stage Crépon-Duguet-Mairesse (CDM) econometric model; the use of manufacturing firms as a benchmark helps to better appreciate our findings. We find that a series of structural, performance, and behavioral factors increase a firm’s propensity to invest in innovation, but some differences arise between services and manufacturing firms. Intensive investment in innovation leads to superior innovation performance, while innovation output has a positive effect on labor productivity.

KEY WORDS:

• innovation
• labor productivity
• services
• manufacturing
• Mexico

Acknowledgments

The authors are indebted to CONACYT for facilitating access to the data used in this study. The authors have benefited enormously from comments by four anonymous reviewers to this journal.

Notes

1. Traditional services include trade, transportation, posting, and storage; mass media information; financial services and insurance; real estate services and rental of property and intangible goods; recreational, cultural, and other recreational services; temporary accommodation and food and beverage preparation; and financial intermediation services. KIBS include scientific and technical services; management of corporate enterprises and support services for businesses; waste management; and remedial services. Other Services include educational services; health care and social assistance; and government activities.

2. Diverse services include real estate services and rental and intangible goods; recreational, cultural, and other recreational services; and other services except government activities. Social services include educational services and health care and social assistance.

3. Barras understood services innovation as a complementary process with other sectors of the economy, notably manufacturing. The model contends that the life cycle of services runs opposite to the cycle of industrial products; the development and subsequent adoption of the latter by services firms contribute to innovation in services.

4. INEGI applies expansion factors that produce a distribution of firms that better reflects the larger share of services firms in the total population of firms. See: INEGI: Conversión SCIAN 2002_2007, Mexico (http://internet.contenidos.inegi.org.mx/contenidos/productos//prod_serv/contenidos/espanol/bvinegi/productos/derivada/cuentas/bienes%20y%20servicios/2011/CByS2007-2011.pdf); also UNSTATS United Nations Statistics Division, Correspondencia entre SCIAN 2002 (México) y ISIC (http://internet.contenidos.inegi.org.mx/contenidos/productos//prod_serv/contenidos/espanol/bvinegi/productos/derivada/cuentas/bienes%20y%20servicios/2011/CByS2007-2011.pdf).

5. ESIDET 2010 follows the North American Industry Classification System (NAICS) 2007, not fully compatible with standard industry classifications based on a firm’s technology intensity. INEGI provides the equivalence between ISIC Rev. 3.1 and NAICS 2007. The NAICS codes for the KIBS industries included in this study were identified according to Institut national de la recherche scientifique (INRS)—Urbanisation Culture Société (2012).

6. We also perform a robustness analysis by including in the equation the predictors from Equation (2):

Additional information

Funding

The authors acknowledge funding and support from the Competitiveness and Innovation Division of the Inter-American Development Bank.