Relationship between innovation, regional institutions and firm performance: Micro-evidence from Africa

Abstract

The main aim of this paper is to examine empirically the extent to which firm-specific action, innovation, and an external factor, the quality of regional institution, affect firms’ productivity. To this end, a firm-level dataset from 15 African countries was used and the three-step CDM approach was employed to estimate the relationship. The empirical results of the study highlight that firms that invest in R&D are more innovative and innovative firms are more productive, indicating the importance of innovation for firms’ productivity. In addition, regional institutional quality has a significant positive effect on productivity, particularly for firms at low productivity distribution. Therefore, improving the quality of regional institutions, in addition to promoting R&D and innovation, is essential to enhance the productive capacity of firms in Africa. The present study contributes to the existing literature by providing new empirical evidence on the role of regional/local institutions on firm productivity for a region, Africa, which needs productivity enhancement the most. It is also recommended that further studies, similar to this one, should be carried out in order to better understand the topic in the context of African firms.

Keywords:

• innovation
• productivity
• regional institutions
• CDM model
• Africa

JEL Classification:

• L26
• O31
• D24

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Real GDP growth data is calculated based on data retrieved from IMF Data Mapper https://www.imf.org/external/datamapper/datasets/WEO/1.

2 The 2030 Agenda for Sustainable Development was adopted by all United Nations Member States in 2015 as a successor of Millennium Development Goals. Its first goal is zero poverty by 2030. Agenda 2060, on the other hand, is Africa’s long-term vision for the next 50 years adopted by African leaders in 2015. It is a master plan for transforming Africa into the global powerhouse of the future by optimizing the use of Africa’s resources for the benefit of the continent’s people.

3 For instance, the Europe 2020 strategy, which was adopted by the European Council in 2010, sets R&D spending target of three percent of GDP for EU members to achieve its goal of smart, sustainable and inclusive growth (Andersson, Johansson, Karlsson, and Lööf 2012, 2).

4 Crepon et al. (1998) estimated the four equations jointly using asymptotic least square, but later studies used an instrumental variable method by using the predicted values. According to Mohnen and Hall (2013), the estimation results using either of the estimation methods produce similar results provided that endogeneity and selection bias is taken into account.

5 Sample countries are selected based on availability of R&D data. The WBES asks firms whether and how much they spend on R&D. Although data for Yes/No question is available for all countries, data for how much firms spend on R&D is only available for selected countries. It should be noted that some of the countries in our sample are too small to be divided into regions. Therefore, the World Bank didn’t divide these countries into regions during the survey.

6 Since R&D expenditure is reported in local currency in the WBES, this variable is converted to US dollars using annual average exchange rates obtained from the IMF.

7 The generalized Tobit model has been used by a large number of empirical studies for innovation input equation, including Crespi and Zuniga (2012), Crespi, Tacsir, and Vargas (2016), Raffo, Lhuillery, and Miotti (2008).

8 Note that a detailed description of the measurement of each independent and dependent variable is provided in Table 2.

9 Mlozi et al. (2018) showed that impact of innovation on performance varies across the African countries because of the differences among them.

10 Cirera, Lage de Sousa, and Sabetti (2016) maintain that even though some firms did not report spending on R&D, they may carry out innovation output. Mohan, Strobl, and Watson (2016) also argue that innovation is a multidimensional and complex process, and R&D expenditure alone underestimates the true cost of innovation.

11 Since WBES collects monetary data using local currency, the total value of sales is converted to US dollars using annual average exchange rates obtained from the IMF to ensure the comparability of sales data for the sampled countries.

12 We have followed a similar approach as that of Hussen and Çokgezen (2020) to calculate the regional/local institutional quality level. For further detail on the measure, see Hussen and Çokgezen (2020).

13 The reported coefficients in the table are marginal coefficients.

14 See UNESCO Institute for Statistics (http://uis.unesco.org/) for more detail.

15 See for instance Guellec and van Pottelsberghe de la Potterie (2000), Nicolaides (2013), and Veugelers (2016).

16 https://au.int/en/agenda2063/overview.