How does political instability affect firm innovation?

(1)

How does political instability affect firm innovation?

Evidence from Sub-Saharan Africa

By Koen Oudehand

S1959530

k.j.oudehand@student.rug.nl

Date: 04-01-2016

University of Groningen Faculty of Economics and Business Master: International Economics and Business

(2)

Abstract

The goal of this study is to examine the effect of political instability on firm innovation. Next to this, the study examines whether certain firm characteristics alter the relationship between political instability and firm innovation. The data from enterprise surveys in 15 Sub-Saharan African countries with a total of 3007 different firms, in 2006 and 2007. The results show a significantly negative effect of political instability on the propensity to innovate. Next to this the study shows that exporting firms are less sensitive to political instability compared to non-exporting firms. Foreign-owned firms are more sensitive to political instability when measured as political violence. Internal training programs for employees can help to reduce the negative effects of political instability on firm innovation.

(3)

1. Introduction

Innovation is often mentioned as one of the key drivers of economic development. Rosenberg, 2004, even stated innovation to be the single, most important component of long-term economic growth. Different studies tried to explain how institutions and policies stimulate or hamper innovation (Edquist and Johnsson, 1997, Johansson and Lööf, 2009, Krammer, 2009). While most scholars focused primarily on the effect of different policies on innovation, a new stream in literature is shifting focus to the stability of political regimes as a driver of innovation (Waguespack et al., 2005, Allard et al., 2012). These studies, however, focus on national systems of innovation and lack firm-level evidence. The main goal of this study is to investigate the effects of political instability on firm innovation.

There are different arguments predicting a negative effect of political instability on firm innovation. Firstly, political instability can lead to uncertainty of the future demand for innovative products. This will make firms more hesitant to innovate. Uncertainty about future demand can also increase the cost of external capital because of increased risk of default (Sandstrom, 2014). The second argument relates to the quality of current and future policies. On the one hand, political instability can decrease the quality of current institutions (Varsakelis, 2006). On the other hand, political instability increases uncertainty about the nature of future policies (Rodrik, 1991). This uncertainty will again decrease the propensity of firm innovation. The last argument states that political instability can decrease the availability of the high-skilled human capital crucial for innovation. (Aguero et al., 2014, Aisen and Veiga, 2011).

(5)

5 2003). Secondly, foreign-owned firms are less affected by the negative effects of political instability on high-skilled human capital. This is because they are more attractive to the remaining high-skilled labour pool (Görg and Greenaway, 2004), and because they have the possibility to shift knowledge and human capital in a parent-subsidiary relationship (Song, 2014). There are however also several arguments predicting foreign-owned firms to be more sensitive to political instability in their innovative decision. Firstly, political instability is often spurred by anti-foreigner sentiments. Because of this foreign firms can be harmed disproportionally (Klapper et al., 2015). Secondly, political instability often increases the risk of expropriation of foreign firms (Kobrin, 1980). Thirdly, foreign-owned firms, in contrast to domestic firms, have the possibility to withdraw innovative activities from politically unstable countries. The third moderator is the presence of internal training programs. For two reasons internal training programs can reduce the negative effects of political instability on firm innovation. Firstly, when high-skilled human capital becomes scarce internal training programs can be a substitute for external human capital (Collier and Duponchel, 2010). Secondly, internal training programs can reduce the costs of knowledge required for innovation and therefore reduce the innovation threshold (Bidwell, 2012).

One of the world’s developing regions with the most problematic innovation numbers is the Sub-Saharan region of Africa. Innovation levels in Sub-Saharan Africa are lagging behind the rest of the developing regions of the world. Graph 1 compares innovation figures for the developing areas Sub-Saharan Africa, Europe and Central Asia, Latin-America & the Carribean, Middle-East and North-Africa, South-Asia and East-Asia & Pacific. All developed countries are bundled in the high-income group. The graphs compare innovative inputs and outputs. Sub-Saharan Africa is performing the worst in terms of R&D expenditures as a percentage of GDP. In terms of R&D personnel per 1000 inhabitants only the South-Asian region is underperforming Sub-Saharan Africa. Also in terms of innovative output the region is performs poorly. With 0.14 US and 0.04 EU patents per million inhabitants the region is performing significantly worse than for example the Latin-American or the Europe and Central Asian region. Because of this the focus of this study will be on the Sub-Saharan region of Africa.

---- INSERT GRAPH 1 ---

(6)

6 instability on the propensity for firms to innovate? The three sub-questions are: Is the effect of political instability on the propensity to innovate different for exporting versus non-exporting firms? Is the effect of political instability on the propensity to innovate different for foreign-owned versus domestic firms? Can internal training programs help to reduce the effect of political instability on firm innovation? This study contributes to the literature in several ways. Firstly, the study provides new evidence on the divers of innovation in developing countries. Secondly, the study will shed new light on literature on the effects of political instability on firm level outcomes. Thirdly, the study will delve into the differences between firms in how they are affected by political instability in their innovative decisions.

The remainder of this paper is organized as following: section 2 will firstly look at the different theories of the effect of political instability on firm and country level economic outcomes. After this the different drivers of firm innovation are elaborated. Based on this the different hypotheses are constructed. Section 3 will elaborate on the data, different variables and the methods used. The results are given in section 4, and finally, section 5 will present the conclusions, limitations and ideas for future research.

2. Literature review and hypotheses

This section will firstly summarize the literature on the effects of political instability. The theoretical effects of political instability on both country and firm level outcomes will be discussed. Secondly, the different firm, market and national drivers of innovation are elaborated. Based on this I will conceptualize the mechanisms through which political instability can affect firm innovation and how firm characteristics can alter the effect of political instability on firm innovation.

2.1 The effect of political instability on country and firm level outcomes.

(7)

7 Because of this the risk of investment increases. This results in a so-called capital flight, where investors invest in other, more politically stable, countries. The second way in which political instability affects economic growth is through the over accumulation of public debt. The theory behind this is that current governments do not internalize the costs of borrowing because they know that there is a possibility that the debt will be serviced by the future governments. This over borrowing will again have negative consequences for economic growth since the loans eventually have to be repaid.

A different mechanism through which political instability affects economic growth is explained by Darby et al., 2004. The baseline of their model is that political instability reduces the possibility of re-election. Because of this governments will be less interested in implementing long-term policies and will choose short-term spending over long-term investment. Less government investment will, in the long run, result in less economic growth. Svensson, 1998, for example found that political instability can reduce the incentives of governments to invest in legal systems needed to execute property right protection laws. Also Devereux and Wen, 1998, argue that political instability will increase government spending over GDP, which will increase public debt and will encourage future governments to increase capital taxation. This increasing taxation will in the long-run decrease private investment.

A different stream of literature examines the negative effects of political instability on economic growth through the impact of instability on the highly educated workforce (Fosu, 1992, Adebayo, 1985). According to Adebayo, 1985, political instability can result in so-called ‘brain drain’, a situation where especially highly educated workers emigrate to countries with better economic opportunities. Recent studies by Aisen and Veiga, 2011, and Aguero et al., 2014, provide empirical evidence for the claim that political instability is harmful for the highly-educated human capital stock. Section 2.3 will elaborate further on the effect of political instability on human capital.

(8)

8 ‘capital flight’, where investors tend to shift their capital to more stable countries. Next to this the destructive effect of conflict will also increase the depreciation rate of capital.

The previous models focus on the effects of political instability on economic growth. There are however also several authors trying to explain the relationship between political instability and firm-level outcomes (Petracco and Schweiger 2008, Klapper et al., 2015, Cerra and Saxena 2008, Collier and Duponchel 2010, Ksoll et al., 2009). These studies work from the same theoretical background in which there are two ways in which political instability can affect firm performance, via demand changes and via higher unit costs of output. The first way in which firms are affected is through demand changes. Political instability and especially war often decline the disposable income of households. This because households often shift their capital to other countries in war, or even use their money for basic subsistence expenditures. This reduces the demand for firms and thus reduces the firm income and performance. Another way in which firms are directly affected by political instability is because of higher unit costs of output. Instability can result in destruction of physical capital: production plants, storage centers and spoliation of transport. Next to this instability can also cause human capital problems (Petracco and Sweiger 2008). In the short run employees might be killed or displaced to another area. In the long-run highly educated human capital might be scarce because of decreasing education levels. The destruction of physical and human capital will result in higher costs of output.

2.2 Drivers of innovation

The main interest of this study is to see how political instability affects innovative activities of firms. For this it is first important to see what the theoretical drivers of innovation are and especially what drives some firms to innovate more than other firms. First it is important to define innovation. A clear definition is given by the OECD. Innovation refers to: all scientific, technological, organisational, financial and commercial activities which lead to, or are intended to lead to, the implementation of technologically new or improved products or services (OECD/Eurostat, 1997). There are several drivers of firm innovation. The drivers can be divided into firm-level, market-level and country-level drivers. Table 1 summarizes all the drivers of innovation.

(9)

9 important driver of innovation. There are several studies providing evidence that young firms are more innovative then older firms (Goedhuys, 2012, Ayyagari et al., 2007). The reason for this is that older firms often stick to their old routine and are hesitant to change. Ownership structures can also affect innovation. According to the transition report written by the European Bank for Reconstruction and Development (EBRD, 2014) state-owned firms are less innovative than privately owned firms. Also, foreign-owned firms tend to be more innovative compared to domestic firms (Johansson and Lööf, 2009, Falk, 2008, OECD, 2008, EBRD, 2014). The mechinism behind this is that foreign owned firms are often part of larger multinationals, and because of this have better access to global knowledge and technologies. Export orientation is another driver of innovation. Exporting firms, like foreign-owned firms, have better access to global knowledge and technologies (Hanley and Monreal-Perez, 2011, Seker, 2011). The presence of high-skilled human capital can is also an important driver of innovation. Several scholars have found evidence that better educated employees provide more innovative output than less well-educated employees (Van Uden et al., 2014, Schneider et al., 2010, Mahemba and Bruijn 2003, EBRD, 2014). Another important firm-level driver of firm innovation is access to finance. Innovation is often associated with investment. For these investments firms often rely on external capital. Evidence showed that innovative firms consider access to finance the most important obstacles to innovation (OECD, 2011). Especially small firms tend to have less access to external financing sources (Beck and Demirguc-Kunt, 2006). The last firm-level driver of innovation is managerial experience. Managers with experience in the sector can affect innovation positively by reducing the uncertainty about the future returns of the innovation (Balsmeier and Czarnitzki, 2014).

(10)

10 value of the investment. This net present value depends largely on the future cash flows. Because of this the decision to innovate depends largely on the expectation of future demand for the innovative product.

Next to firm and market level drivers there are also several country-level characteristics that can stimulate innovation. Obviously, one of the most important institutional drivers of firm innovation is protection of intellectual property rights. Property right laws are introduced to reduce the uncertainty of imitation and expropriation of property in the future. Different studies found a positive relationship between the quality of property rights protection and the degree of innovation (Lai, 1998, Woo et al., 2015, Naghavi and Tsai, 2015, Chen and Puttitanun, 2005). Another country-level driver of firm innovation is the quality of the educational system. Different scholars have found a positive relationship between the quality of the educational system and the level of innovation (Krammer, 2009, Varsakelis, 2006). Especially universities play a crucial role in enhancing and facilitating innovation (Krammer, 2009). The development level of a country is also an important driver for innovation. More developed countries tend to be better able to support innovative activities (Ayyagari et al., 2007). The last country-level driver of innovation is foreign direct investment. Especially in developing countries FDI is often associated with positive spill over effects for local firms (Wooster and Diebel, 2010, Krammer, 2009).

---- INSERT TABLE 1 ---- 2.3 Hypothesis building

Based on the previous sections there are three different ways in which political instability can affect firm innovation. Firstly, political instability can create uncertainty about future demand for innovative products. This can also result in decreasing access to finance. Secondly, political instability can decrease the quality of current policies and institutions and increase the uncertainty about future policies and institutions. Finally, political instability can create scarcity of highly educated employees.

(11)

11 uncertainty in household income increases the uncertainty of demand for innovative products and will thus increase the risk of innovation. The uncertainty of future demand will not only make firms more hesitant to innovate, it will also the increase the costs of outside capital to finance innovation. The uncertainty about future demand will increase the risk of default (Sandstrom, 2014). Because of this increased default risk banks will increase the risk premium on loans and will thus increase the cost of external capital. As a result, the threshold to innovate is even higher.

(12)

12 All three arguments predict a negative effect of political instability on firm innovation. Therefore the first hypothesis states:

Hypothesis 1: Political instability decreases firms’ innovation propensity.

There are several firm characteristics possibly moderating the negative effect of political instability on firm innovation. The first moderator is export orientation. Exporting firms are affected less by political instability in their propensity to innovate because of three mechanisms: they are better able to spread sales risk across countries, have better access to finance and they are less dependent on domestic institutions.

One way in which political instability affects the propensity to innovate is through demand shocks. As mentioned before, political instability decreases household disposable income and thus decreases the current demand for innovative products and makes the future demand more uncertain. Exporting firms are however less dependent on domestic sales since part of their sales is from demand in other countries. Since exporting firms are less dependent on domestic sales these firms are also less sensitive to the negative effects of political instability on household disposable income. Atkin et al., 2014, support this claim in their study on the differences between exporting and non-exporting firms in Egypt. They found that exporting can work as a mechanism to decrease the volatility in demand caused by political instability. Secondly, exporting firms, compared to non-exporting firms, are less capital constraint in politically unstable times. As seen in the previous section, banks are more hesitant to grant loans when political instability increases the risk of default (Sandstrom, 2014). This risk of default is the consequence of the demand uncertainty for innovative products. However, since exporting firms are better able to spread demand uncertainty across different countries they also have lower default risk. This results in banks being more willing to grant loans to exporting firms compared to non-exporting firms. Different studies have proven the claim that exporting firms have more access to finance compared do non-exporting firms (Castagnino et al., 2013, Beck et al., 2005).

(13)

13 decreases the quality of property right protection in country X, an exporting firm can still innovate in country X but register and sell the innovative product in country Y. In this way the firm can keep innovating without experiencing the negative policy changes associated with political instability. Paunov, 2015, shows that exporting firms are indeed able to circumvent the negative effects of institutional and political quality on firm innovation.

Concluding, I posit that exporting firms are less sensitive to political instability in their innovative decision because they are better able to circumvent the negative effects of instability on the home market.

Hypothesis 2: Exporters, compared to non-exporters, are less sensitive to political instability in their decision to innovate.

The second possible moderator is the degree of foreign ownership. There are three arguments why foreign-owned firms are less sensitive political instability in their innovation decision. Foreign-owned firms are less sensitive to political instability because they are less capital constraint, are more attractive to highly educated human capital and have the possibility of internal knowledge mobility.

One of the arguments in the previous section for political instability having a negative effect on firm innovation is that costs of external capital increase because of higher risk of default (Sandstrom., 2014). Due to this the investment costs of innovation increase. However, foreign-owned firms are often subsidiaries of larger multinationals. Because of this, they have more internal financing possibilities and thus depend less on external loans. Due to this, foreign-owned firms have the possibility to keep innovating even when political instability increases the cost of external capital. Several authors (Harrison and McMillan 2003, Mickiewicz et al., 2004, Arbeláez and Echavarria, 2002) found evidence for the fact that foreign owned firms are indeed less capital constraint compared to domestic firms.

(14)

14 2010, Aitken and Lipsey, 1996). The latter argue that foreign-firms can pay higher wages because they often have higher productivity and are more capital-intensive. Hijzen et al., 2010, show that the wage differential between foreign and domestic firms is especially large for highly educated workers. This enforces the claim that foreign-owned firms are more attractive for high-skilled employees.

The third mechanism for foreign firms to be less sensitive to political instability relates to the parent-subsidiary relationship. As mentioned before, foreign-owned firms are often subsidiaries of larger multinationals. This relationship enables the possibility of internal transfer of resources from the parent to the foreign subsidiary. When political instability causes scarcity of domestic highly educated employees, foreign-owned firms have the possibility to attract employees from the parent company. In this way foreign-owned firms can still acquire the necessary human capital needed for their innovative activities. There are several studies showing that parent-subsidiary transfers of employees can indeed fill the gap when the stock of qualified employees on the home market is scarce (Edström and Galbraith, 1977, Harzing, 2001).

Concluding, there are several arguments why foreign-owned firms are less sensitive to political instability in their innovative decision. Because of this the third hypothesis states:

Hypothesis 3a: Foreign-owned firms are, compared to domestic firms, less sensitive to political instability in their decision to innovate

There are however also several arguments for why foreign-owned firms are more sensitive to political instability in their innovation propensity. Foreign-owned firms are often harmed disproportionally by political instability, have higher expropriation risk and have the possibility to withdraw innovative activities from unstable economies.

(15)

15 firms and therefore these firms are, compared to domestic firms, more hesitant to invest in innovative activities.

Another reason for foreign-owned firms to be especially reluctant to innovate in politically unstable times is increased risk of expropriation (Kobrin, 1980, Azzimonti and Sarte, 2007). The OECD, 2004, defined expropriation as “an act where there is a compulsory transfer of property rights by the host state”. This implies that foreign firms involuntarily lose all rights of property. Azzimonti and Sarte, 2007, explain that countries with a high level of political instability often experience high rates of expropriation of foreign firms. Because of this foreign-owned companies are more hesitant to invest in innovations since there is an increased possibility of compulsory transfer of the property rights. Kobrin, 1984, shows that the problem of expropriation has been especially severe in the Sub-Saharan region of Africa. In the period between 1960 and 1979, 43.4% of all nationalizations of foreign companies occurred in this region.

The third argument for foreign-owned firms to be more sensitive to political instability again relates to the parent-subsidiary relationship. As mentioned before, foreign-owned firms are often part of larger multinationals. Fisch, 2011, found that multinationals are hesitant to invest in subsidiaries that are located in politically unstable countries. Especially irreversible investments, like R&D, seem to decrease when political situation in the subsidiary home country worsens. Multinationals often have the possibility to reallocate innovative activities between subsidiaries in different countries and therefore prefer to invest in politically stable economies. Due to this, foreign-owned firms withdraw from innovative activities quicker compared to domestic firms.

Concluding, there are also several arguments why foreign-owned firms might be more sensitive to the effects of political instability on innovation. Because of this I introduce the competing hypothesis:

Hypothesis 3b: Foreign-owned firms are, compared to domestic firms, more sensitive to political instability in their decision to innovate.

(16)

16 programs can substitute external human capital and secondly, internal training programs can decrease the investment costs of innovation.

As mentioned before, one of the most important drivers of firm innovation is the availability of high-skilled human capital. In the previous section the argument was made that political instability decreases the availability of adequately educated workers. There are two mechanisms through which political instability can decrease the availability of adequately educated employees. Firstly, political instability can lead to violence and can result in death or migration of people (Collier and Duponchel, 2010) and secondly, political instability can lead to decreasing education quality and thus to a smaller highly educated human capital pool (Aisen and Veiga, 2011). Because of this it becomes harder for firms to acquire the skilled human capital needed for innovation. A substitute for external human capital acquisition could be internal mobility. Less educated employees can be transformed to highly educated employees through training and education. Internal training programs can in this way work as a substitute to external accumulation of high-skilled employees. Collier and Duponchel, 2010, confirm this claim by noticing that the instability in Sierra Leone between 1991 and 2002 indeed increased the willingness of firms to invest in training programs for employees.

There is another reason why internal training can reduce the negative effect of political instability on firm innovation. One of the arguments for political instability negatively affecting the innovation propensity was that instability increased the uncertainty about future demand for innovative products. This uncertainty can increase the threshold to invest because the costs of investment might be bigger than the future revenues. Internal mobility of low educated to high educated employees is less expensive than external accumulation of skilled labor (Bidwell, 2012). Because of this internal mobility through training programs can reduce the investment costs of innovation and so reduce the threshold to innovate. In this way internal training can help to create the high-skilled labor force required for innovation, without excessive investments.

Summarizing, both arguments predict that internal mobility through training programs can help to reduce the negative effects of political instability on firm innovation. Because of this, the fourth hypothesis predicts:

(17)

17

3. Methodology

3.1 The sample

The data on firm characteristics is from the Enterprise Survey initiative from the World Bank Group. The surveys come from face-to-face interviews with top managers and business owners over 130.000 companies in 135 countries. The surveys cover several business topics like finance, corruption, crime, infrastructure, competition and performance measures. From these surveys I use the data on Sub-Saharan countries from 2006 and some from 2007. This because this is the only year in which questions related to innovation were asked. All the selected data is from the manufacturing sector because this is the only sector with data for innovation available. Although Sub-Saharan Africa counts a total of 50 countries, there is data on firm innovation for 15 Sub-Saharan countries. The sample consist of the following countries: Angola, Botswana, Burundi, Democratic Republic Congo (DRC), Gambia, Guinea, Guinea Bissau, Kenya, Mauritania, Namibia, Nigeria, Rwanda, Swaziland, Tanzania and Uganda. The total sample is 4737 firms large, however because of empty values and unwillingness or inability to answer the number of useful observations is 3007. Table 2 shows a division of the sample per country and table 3 shows a division of the sample per industry.

---- INSERT TABLE 2 --- ---- INSERT TABLE 3 --- 3.2 Dependent variables

The dependent variable in this study is firm innovation. Innovation will be measured as the output of innovation activities. Because of lack of data it is not possible to measure the innovation input. Following Fritsch and Görg, 2013, the first measure of firm innovation will be the introduction of new products or services. The second measure for innovation is the measure of new or improved production processes.

3.3 Independent variables

(18)

18 in a country. Table 4 explains all the factors included in this measure. The values are normally distributed and range from -2.5, being very politically unstable, to 2.5 being very politically stable. Since this study is on instability it would make more sense to multiply all the values by -1, so that 2.5 would mean highly unstable. Because of this it is easier to remove all the negative values. I do this by adding 2.5 to all the values. This means that 0 becomes politically very stable and 5 becomes politically very unstable. To give an example, for the period of 2000-2004, the most politically stable country in my sample is Botswana with a value of 1.57 and the most unstable country is Burundi with a value of 4.84.

---- INSTERT TABLE 4 --- 3.4 Moderating variables

In the model there are three possibly moderating variables, namely export orientation, foreign ownership and internal training programs. The degree of export orientation will be measured as a share of the total sales. For the measure we sum up the percentage of direct exports with the percentage of indirect exports. Export orientation will be transformed into a dummy being 1 if the firm is involved in exports and 0 if not. The degree of foreign ownership is measured as the percentage of the firm that is owned by foreign entities. Foreign ownership will be transformed into a dummy being 1 if the firm is more than 50% foreign owned and 0 if not. The presence of internal training programs is a dummy of 1 if the establishment organized training programs in the last year and 0 if not.

3.5 Control variables

There are several control variables that have to be taken into account. These variables can be divided into firm-specific controls, market specific and country-specific controls.

(19)

19 is the experience of managers. This is measured as the years the manager has experience in the sector. The last firm control variable is the degree of government ownership. There are also two controls for the competitiveness of the market. I control for the amount of competition and the presence of foreign competitors.

Next to firm- and market-level controls there are also a couple of country-specific controls. These are also based on the drivers of innovation in the previous section. First I want to control for the level of innovativeness of the country. I do this by controlling for the number of patent applications of the specific country in the U.S. I take all the patents between 1960 and 2004. The second country control is FDI stock. FDI can be an important driver of firm performance and innovation because of spill-over effects. Finally, I also want to control for the development level of the country. I control for GDP per capita as a measure for development.

All variables are listed in table 5. Here the different measurements and the sources are also explained in further detail.

--- INSTERT TABLE 5 ---- 3.6 Model

This study tests for two different innovation activities, namely product/service and process innovation. Product/service and process innovation are often complementary. The introduction of new products is often followed by the introduction of new processes. Different studies found evidence for this interdependency (Miravete and Pernias, 2006, Rouvinen, 2002, Leo, 1996). To show this interdependency in my sample, column 8 of table 2 shows the percentage of product innovating firms who also innovated their processes. The totals show that 87% of all the product innovators also innovated processes and that 77% of all the process innovators also innovated products. Because of this complementarity I introduce a bivariate probit model. This kind of probit model takes into account that product/service and process innovation are often jointly determined (Andersson et al., 2012). The significant rhos in the model justify the use of a bivariate probit instead of two separate probit models.

𝑏𝑖𝑝𝑟𝑜𝑏(𝐼𝑛𝑛𝑜𝑣) = 𝛽0 + 𝛽1𝑃𝐼 + 𝛽2𝐹𝑖𝑟𝑚𝑆𝑖𝑧𝑒 + 𝛽3𝐹𝑖𝑟𝑚𝑃𝑒𝑟 + 𝛽4 𝐹𝑖𝑟𝑚𝐴𝑔𝑒 +

(20)

20 In this model Innov will measure the propensity to innovate in either products and services or production processes. The value is a binary variable being 1 if the firm innovated in the last three years and 0 if not. PI is the measure for political instability. FirmSize measures the total number of employees in the firm. Firmper measures firm profit. FirmAge is equal to the years the firm exists in the country, Rellskill measures the firm reliance on high-skilled labor. ForCom measures to what extent the firm experiences pressure from foreign competitors, Pat measures the innovativeness of the country, NumComp meaures the competitiveness in the market. Manexp measures the managerial experience in the firm. FDI measures the country’s FDI stock, GovOwn measures the degree of government ownerhip. GDP measures the country’s GDP per capita.

To test for the moderating effect of export orientation, foreign ownership and internal training programs interactions term will be included. ExpDummy is equal to 1 if the firm exports and 0 if not. PI*ExpDummy is the interaction term between the political instability measures and the export dummy. ForDummy is equal to 1 if the firm is foreign owned and 0 if not. PI*ForDummy is the interaction term between the political instability measures and the foreign ownership dummy. Trainpro is equal to 1 if the firm organized training programs for employees, and 0 if not. PI*Trainpro is the interaction term between the political instability measures and the presence of training program dummy.

Some variables are transformed to logarithms because of skewed distribution. Table 5 indicates for which variables this is applicable.

4. Empirical results

4.1 Descriptive statistics

Table 6 gives the descriptive statistics for all of the variables in the model. On average, 50% of all the firms introduced new products or services in the previous three years. For production process innovation the percentage is 58%. 15% of the firms where involved in exports. 11% of all firms where owned by foreign entities for more than 50%. Next to this we see that 30% of the firms offered internal training programs to employees.

(21)

21 Table 7 reports the correlation matrix. As expected there is a high correlation between the different types of innovation. Next to this the different measures of political instability also have a high degree of correlation. The rest of the values are all within acceptable ranges.

---- INSERT TABLE 7 ---- 4.2 Results

(22)

22 coefficient is negative and significant at a 1% level. This supports the first hypothesis, political instability negatively affects the firm propensity to innovate.

--- INSERT TABLE 8 ---

The magnitude of the political instability coefficients in models 3 and 4 are hard to interpret. Because of this table 9 reports the marginal effects. In models 15 and 16 we see that the coefficient of political instability for product/service innovation is equal to -0.066. This implies that an increase in the political instability measure of 1 decreases the propensity to innovate with 6.6% point on average. The effect of political instability on process innovation is fairly similar. An increase of 1 in the political instability measure will decrease the innovation propensity with 6.3% point. To illustrate these marginal effects I take as an example Angola and Botswana. Angola is one of the most politically unstable countries in Africa (WGIn 4.2) and Botswana is one of the most politically stable countries (WGI 1.6). The difference between the political instability measures of these two countries is 2.6. This would indicate that Botswanan firm has a propensity to innovate that is 21% higher than an Angolan firm, keeping all other variables equal. Of course this is far from realistic because the countries differ at some more factors, but it gives an illustration of the magnitude of the effect of political instability on firm innovation.

--- INSERT TABLE 9 ---

To test hypothesis 2 columns 5 and 6 in table 8 introduce an interaction term between the political instability measure and the export dummy. The positive and significant coefficient in column 5 implies that there is a significant different between exporting and non-exporting firms in the effect of political instability on the propensity of product/service innovation. Graph 2 is a graphical representation of this interaction effect. The graph clearly shows that exporting firms are affected less by political instability in their innovation probability. This supports the second hypothesis. The interaction is however only significant for product and service innovation.

(23)

23 effect of training programs, hypothesis 4. The interaction term between political instability and the training dummy is significant for both product/service and process innovation. This supports the fourth hypothesis, external training programs help to reduce the negative effect of political instability on firm performance.

4.3 Robustness checks

For robustness I introduce two different measures of political instability. The WGI political instability measure is a comprehensive, multidimensional measure of instability. In this section I will introduce two one-dimensional measures. One measure mild form of political instability, namely government change (Haber et al., 2004, Gurgul and Lach 2013, Aisen and Verga 2005), and one more severe measure of political violence (Imai and Weinstein 2000, Serneels and Verpoorten., 2013), namely the number of battle-related deaths. The degree of government change is a measure of the number of government changes multiplied by the magnitude of policy change associated with the change of government. The results can be found in table 10. Government change has a negative effect on innovation for both product/service and process innovation. This enforces the evidence for hypothesis 1. The fourth column shows that the interaction term of export orientation and political instability is again significant for product/service innovation. This supports hypothesis 1. The marginal effects of government change are presented in models 17 and 18 in table 9. The marginal effect of government change on product/service innovation for example is equal to -0.002. This indicates that for example the military coup in Uganda in 1985, with a policy change magnitude of 7, reduced the average innovation propensity of Ugandan firms with 1.4% point. The introduction of government change does not provide evidence for hypothesis 2, the interaction terms models 27 and 28 remain insignificant. Model 29 and 30 do enforce the evidence for hypothesis 4. Training programs do still work as a mechanism to reduce the negative effects of political instability on firm innovation.

(24)

24 casualties because of violence will decrease the probability of a firm to innovate in new products, services or processes. This again enforces the evidence for hypothesis 1. The marginal effect of the BRD measure is equal to -0.078 for product/service innovation, implying that 1000 extra casualties decrease the firm innovation propensity with 7.8% point on average. For process innovation the marginal effect is stronger. An increase of 1000 extra casualties will decrease the average propensity to innovate processes with 10.2% point. The introduction of the new political instability measures did also change the interaction terms. The interaction between the BRD measure and the export dummy is not significant, implying that there is no significant difference between exporting and non-exporting firms in the way that increasing casualties affect their innovation probability. However, the interaction term between BRD and the foreign ownership term did become significant for product/service innovation at a 10% level. The negative coefficient of this interaction term implies that foreign owned firms are affected stronger by the BRD measure in their product/service innovation propensity. This suggests that there is some evidence for hypothesis 3b, foreign owned firms are more sensitive to political instability in their innovation probability.

--- INSERT TABLE 11 ---

Both export orientation and foreign ownership are measured as dummies. For robustness I also include continuous measures of the share of exports in the total sales and the share of foreign ownership. The results are not shown in a table but the interaction term between the WGI measure and the share of exports is still significantly positive for product and service innovation. This enforces the evidence for hypothesis 2. For hypothesis 3 include a continuous percentage of foreign ownership. The interaction term between the WGI measure and the foreign ownership share remain insignificant. The interaction term between the BRD measure and the continuous measure of foreign ownership is again significantly negative at a 10% level. This enforces the earlier found evidence that the degree of foreign ownership weakly moderates the relationship between political violence and product/service innovation.

Since Nigeria is 31% of the total sample, the results could be biased towards Nigeria. Because of this table 12 reports the same results as table 8, only now leaving out Nigerian firms. When comparing table 8 with table 12 we can see that the results hold. There are no major differences in the coefficients or the significance.

(25)

25 for innovation are all measurements for the three years before 2005 or 2006. The firm characteristics are however measures of only one year, the year 2005 or 2006. Because of this inconsitancy I transform all the firm controls to categorical variables with three categories. In this way I can get around the endogeneity problem, since the probability that a firm shifts between categories is very small. Table 13 reports the results. Again we see that the results of table 8 hold.

5. Conclusion and implications

Innovation is considered to be one of the most important drivers of economic development. Several studies investigated the relationship between institutions and policies to explain differences between country and firm innovativeness (Edquist and Johnsson, 1997, Johansson and Lööf, 2009, Krammer, 2009). Another stream in literature shifted focus to the stability in the political environment as driver of innovation (Waguespack et al., 2005, Allard et al., 2012). These studies are limited to the macro-economic consequences of political instability. There is, however, no evidence on the effect of political instability on firm-level innovativeness.

(26)

26 The findings of this study suggest different implications on national and firm level. At the country level this study presents two implications. Firstly, in line with Allard et al., 2012, and Waguespack et al., 2005, the findings suggest that stability of the political environment is crucial for firm innovation. Firms tend to be more hesitant to invest in innovative activities when there is high uncertainty about the nature of future governments. Therefore, governments that want to increase firm innovativeness should pay special attention to developing a stable and predictable political environment. Perhaps stability and predictability are even more important than the quality of certain policies and institutions. Secondly, the study shows that foreign-owned firms are affected disproportionally in their innovative decision. This should be a serious cause for concern since innovation in foreign-owned firms is often accompanied by positive technological spill overs for domestic firms (Liu, 2008, Gorodnichenko et al., 2014).

Next to country level implications this study also provides two managerial implications. Firstly, the results show that spreading demand uncertainties across different countries can be a way to limit the impact of political instability. Firms that export tend to be less sensitive to changes in the political environment. Especially in countries that are known to have high levels of political uncertainty firms should operate on different markets to reduce the negative effects of future political risk. Secondly, the results suggest that training of employees can function as a mechanism to circumvent the negative effects of political instability on the availability of highly educated human capital.

(27)

27 declined. Future research must show weather the findings of this study are robust in a panel analyses.

6. References

Adebayo, A. 1985. Brain Drain within the Ecowas Region. Issue: A Journal of Opinion, 14, pp.37-61.

Aghion, P., Bloom, N., Blundell, R., Griffith, R. and Howitt, P. 2004. Competition and Innovation: An Inverted U Relationship. SSRN Electronic Journal.

Aisen, A. and Veiga, F. 2010. How Does Political Instability Affect Economic Growth?. SSRN Electronic Journal.

Aitken, B., Harrison, A. and Lipsey, R. 1996. Wages and foreign ownership. A comparative study of Mexico, Venezuela, and the United States. Journal of International Economics, 40(3-4), pp.345-371.

Alesina, A. and Tabellini, G. 1989. External debt, capital flight and political risk. Journal of International Economics, 27(4), pp.199-220.

Allard, G., Martinez, C. and Williams, C. 2012. Political instability, pro-business market reforms and their impacts on national systems of innovation. Research Policy, 41(3), pp.638-651.

Andersson, M., Johansson, B., Karlsson, C., Lööf, H., (2012). The Importance of Process and Product Innovation for Productivity in French Manufacturing and Service Industries. In Innovation and Growth: From R&D Strategies of Innovating Firms to Economy-wide Technological Change. pp.128-160, Oxford: Oxford University Press.

Arbeláez, M. and Echavarria, J. 2002. Credit, Financial Liberalization and Manufacturing Investment in Colombia. SSRN Electronic Journal.

Atkin, D., Khandelwal, A. and Osman, A. 2014. Exporting and Firm Performance: Evidence from a Randomized Trial. NBER Working Paper, 20690.

(28)

28 Azzimonti, M., Sarte, P., 2007. Barriers to Foreign Direct Investment Under Political Instability. Economic Quarterly, 93(3), pp.287-315.

Balsmeier, B. and Czarnitzki, D. 2014. How Important is Industry-Specific Managerial Experience for Innovative Firm Performance? SSRN Electronic Journal.

Beck, T., Demirguc-Kunt, A. and Maksimovic, V. 2005. Financial and Legal Constraints to Growth: Does Firm Size Matter?. The Journal of Finance, 60(1), pp.137-177.

Becker, G. (1967). Human Capital. A Theoretical and Empirical Analysis with Special Reference to Education. Population (French Edition), 22(2), pp.330.

Bhattacharya, U., Hsu, P., Tian, X. and Xu, Y. 2015. What Affects Innovation More: Policy or Policy Uncertainty? SSRN Electronic Journal.

Castagnino, T., D'Amato, L. and Sangiacomo, M. 2013. How do firms in Argentina get financing to export? ECB Working Paper, 1601.

Cerra, V. and Saxena, S. 2008. Growth Dynamics: The Myth of Economic Recovery. American Economic Review, 98(1), pp.439-457.

Chen, Y. and Puttitanun, T. (2005). Intellectual property rights and innovation in developing countries. Journal of Development Economics, 78(2), pp.474-493.

Collier, P. 1999. On the economic consequences of civil war. Oxford Economic Papers, 51(1), pp.168-183.

Collier, P. and Duponchel, M. 2013. The Economic Legacy of Civil War: Firm-level Evidence from Sierra Leone. Journal of Conflict Resolution, 57(1), pp.65-88.

Cumming, D., Rui, O. and Wu, Y. 2015 Political Instability, Access to Private Debt, and Innovation Investment in China. SSRN Electronic Journal

Darby, J., Li, C. and Muscatelli, V. 2004. Political uncertainty, public expenditure and growth. European Journal of Political Economy, 20(1), pp.153-179.

Devereux, M. and Wen, J. 1998. Political instability, capital taxation, and growth. European Economic Review, 42(9), pp.1635-1651.

(29)

29 Edstrom, A. and Galbraith, J. 1977. Transfer of Managers as a Coordination and Control Strategy in Multinational Organizations. Administrative Science Quarterly, 22(2), pp.248-266. Falk, M. 2008. Effects of Foreign Ownership on Innovation Activities: Empirical Evidence for Twelve European Countries. National Institute Economic Review, 204(1), pp.85-97.

Financing Business R&D and Innovation. 2012. OECD STI Policy Profiles.

Fisch, J. 2011. Real call options to enlarge foreign subsidiaries - The moderating effect of irreversibility on the influence of economic volatility and political instability on subsequent FDI. Journal of World Business, 46(4), pp.517-526.

Fosu, A. 1992. Political Instability and Economic Growth: Evidence from Sub-Saharan Africa. Economic Development and Cultural Change, 40(4), p.829.

Fritsch, U. and Görg, H. 2015. Outsourcing, Importing and Innovation: Evidence from Firm-level Data for Emerging Economies. Review of International Economics, 23(4), pp.687-714. Globerman, S. and Shapiro, D. 2003. Governance infrastructure and US foreign direct investment. Journal of International Business Studies, 34(1), pp.19-39.

Goedhuys, M. and Veugelers, R. 2012. Innovation strategies, process and product innovations and growth: Firm-level evidence from Brazil. Structural Change and Economic Dynamics, 23(4), pp.516-529.

Görg, H., Greenaway, D. 2004. Much Ado about Nothing? Do Domestic Firms Really Benefit from Foreign Direct Investment? The World Bank Research Observer, 19(2), pp.171-197. Gorodnichenko, Y., Svejnar, J. and Terrell, K. (2014). When Does FDI Have Positive Spillovers? Evidence from 17 Transition Market Economies. SSRN Electronic Journal.

Gorodnichenko, Y., Svejnar, J. and Terrell, K. 2010. Globalization and Innovation in Emerging Markets. American Economic Journal: Macroeconomics, 2(2), pp. 194-226

Gurgul, H. and Lach, A. 2013. Political instability and economic growth: Evidence from two decades of transition in CEE. Communist and Post-Communist Studies, 46(2), pp.189-202. Haber, S., Maurer, N. and Razo, A. 2003. The politics of property rights. Cambridge: Cambridge University Press.

(30)

30 from matched Spanish innovators. Economics Letters, 116(2), pp.217-220

Harrison, A. and McMillan, M. 2003. Does direct foreign investment affect domestic credit constraints? Journal of International Economics, 61(1), pp.73-100.

Harzing, A. 2001. An analysis of the functions of international transfers of managers in MNCS. Employee Relations, 20(6), pp.581-598.

Hijzen, A., Martins, P., Upward, R. and Schank, T. 2010. Do Foreign-Owned Firms Provide Better Working Conditions Than Their Domestic Counterparts? A Comparative Analysis. SSRN Electronic Journal.

Huergo, E. and Jaumandreu, J. 2004. How Does Probability of Innovation Change with Firm Age? Small Business Economics, 22(3/4), pp.193-207.

Indirect Expropriation and the Right to Regulate in International Investment Law. 2004. OECD Working Papers on International Investment.

Innovating to Learn, Learning to Innovate. 2008. OECD, pp.161-172.

Klapper, L., Richmond, C., Tran, T. 2015. Civil Conflict and Firm Performance: Evidence from Côte d'Ivoire. World Bank's Knowledge for Change Program (KCP).

Kobrin, S. 1980. Foreign enterprise and forced divestment in LDCs. International Organization, 34(1), pp.65-88.

Kobrin, S. 1984. Expropriation as an Attempt to Control Foreign Firms in LDCs: Trends From 1960 to 1979. International Studies Quarterly. 28(3), pp.329-348.

Krammer, S. 2009. Drivers of national innovation in transition: Evidence from a panel of Eastern European countries. Research Policy, 38(5), pp.845-860.

Ksoll, C., Macchiavello, R. And Morjaria, A. 2009. Guns and Roses: The Impact of the Kenyan Post-Election Violence on Flower Exporting Firms. CSAE Working Paper Series, 6.

Lai, E. 1998. International intellectual property rights protection and the rate of product innovation. Journal of Development Economics, 55(1), pp.133-153.

(31)

31 Lipsey, R. and Sjöholm, F. 2004. Foreign direct investment, education and wages in Indonesian manufacturing. Journal of Development Economics, 73(1), pp.415-422.

Liu, Z. 2008. Foreign direct investment and technology spillovers: Theory and evidence. Journal of Development Economics, 85(1-2), pp.176-193.

Mahemba, C. and Bruijn, E. 2003. Innovation Activities by Small and Medium-sized Manufacturing Enterprises in Tanzania. Creativity and Innovation Management, 12(3), pp.162-173.

Mickiewicz, T., Bishop, K., Varblane, U., 2004. Financial constraints in investment – foreign versus domestic firms. Panel data results from Estonia, 1955-1999. William Davidson Institute Working Paper, 51.

Miravete, E. and Pernias, J. (2006). Innovation Complementarity and Scale of Production. Industrial Economics, 54(1), pp.1-29.

Naghavi, A. and Tsai, Y. 2015. Cross-Border Intellectual Property Rights: Contract Enforcement and Absorptive Capacity. Scott J Polit Econ, 62(2), pp.211-226.

Paunov, C. 2015. Corruption's asymmetric impacts on firm innovation. Journal of Development Economics, 118(1), pp.216-231.

Petracco, C. and Schweiger, H. 2008. The Impact of Armed Conflict on Firms' Performance and Perceptions. SSRN Electronic Journal.

Rodrik, D. 1991. Policy uncertainty and private investment in developing countries. Journal of Development Economics, 36(2), pp.229-242.

Rosenberg, S. 2004. Innovation and Economic Growth. OECD Working Paper.

Rouvinen, P. (2002). Characteristics of product and process innovators: some evidence from the Finnish innovation survey. Applied Economics Letters, 9(9), pp.575-580.

Schmookler, J. (962. Economic Sources of Inventive Activity. Journal of Economic History, 22(1), pp.1-20.

(32)

32 Seker, M. 2011. Importing, Exporting, and Innovation in Developing countries. World Bank Working Paper.

Serneels, P. and Verpoorten, M. 2013. The Impact of Armed Conflict on Economic Performance: Evidence from Rwanda. Journal of Conflict Resolution, 59(4), pp.555-592. Shefer, D. and Frenkel, A. 2005. R&D, firm size and innovation: an empirical analysis. Technovation, 25(1), pp.25-32.

Song, J. 2014. Subsidiary absorptive capacity and knowledge transfer within multinational corporations. Journal of International Business Studies, 45(1), pp.73-84.

Svensson, J. 1998. Investment, property rights and political instability: Theory and evidence. European Economic Review, 42(7), pp.1317-1341.

The Measurement of Scientific and Technological Activities. 1997. OECD/Eurostat Oslo Manual.

van Uden, A., Knoben, J. and Vermeulen, P. 2014. Human Capital and Innovation in Developing Countries: a Firm-Level Study. Institute for Management Research Working paper. Varsakelis, N. 2006. Education, political institutions and innovative activity: A cross-country empirical investigation. Research policy, 35(7), pp. 1083-1090.

Waguespack, D., Birnir, J. and Schroeder, J. 2005. Technological development and political stability: Patenting in Latin America and the Caribbean. Research Policy, 34(10), pp.1570-1590.

Weinstein, J. and Imai, K. 2000. Measuring the Economic Impact of Civil War. CID Working Paper, 51.

Woo, S., Jang, P. and Kim, Y. 2015. Effects of intellectual property rights and patented knowledge in innovation and industry value added: A multinational empirical analysis of different industries. Technovation, 43(1), pp.49-63.

(33)

(34)

34

7. Appendices

Table 1 Drivers of innovation

Firm-level Market-level Country-level

Age Degree of competition Protection of intellectual

property

Size Presence of foreign

competition

Quality of the educational system

Private-ownership Demand for innovative products

Development level

Foreign-ownership Foreign direct investment

Export orientation

High-skilled human capital Access to finance

(35)

35 Table 2 Comprehensive division of the sample by country and innovation type

Country Number of firms in survey Product/service innovators

Process innovators Product/service and process innovators % of product/service innovators who also innovated in process % of process innovators who also innovated

products/services Count % of total Count % of total

(36)

36 Table 3 Division of sample by industry

Industry Freq. Percentage

of total Product/service innovators Process innovaters Mfg - Food 850 28.17 457 467

Mfg - Wood, wood products & furniture

529 17.53 238 293

Mfg - Garments 511 16.94 246 314

Mfg - Other Manufacturing 486 16.11 460 281

Mfg - Metal & Metal products 322 10.67 144 174

Mfg - Chemicals 133 4.41 91 96

Mfg - Non-metallic minerals 70 2.32 29 34

Mfg - Textiles 67 2.22 38 48

Mfg - Machinery & Equipment 39 1.29 21 24

(37)

37 Table 4 Components of the WGI political stability and absence of violence/terrorism measure Survey source Data provider Concept measured

Political Risk Index and Operational Risk Index (BRI), Quantitative Risk Measure in Foreign Lending (QLM) Business Environment Risk Intelligence (BERI)

Fractionalization of political spectrum and the power of these factions Fractionalization by language, ethnic and/or religious groups and the power of these factions

Restrictive (coercive) measures required to retain power Organization and strength of forces for a radical government

Societal conflict involving demonstrations, strikes, and street violence Instability as perceived by non-constitutional changes, assassinations, and guerrilla wars

Global Risk Service Global Insight, Boston, Mass.

Military Coup Risk

Major Insurgency/Rebellion Political Terrorism

Political Assassination Civil War

Major Urban Riot Country Risk Service,

Country Forecasts Economist Intelligence Unit Armed conflict Violent demonstrations Social unrest International tensions Gray Area Dynamics Cerberus

Corporate Intelligence

Autonomy and Separatism Civil Unrest

State of Emergency/Martial Law

Active Terrorist Groups in the last two years Global Competitiveness

Survey

World Economic Forum

Country terrorist threat: Does the threat of terrorism in the country impose

significant costs on firms? Cingranelli Richards

Human Rights Database

University of Binghamton

Frequency of political killings Frequency of disappearances Frequency of torture

Country Security Risk Ratings

iJET Security Risk Rating Institutional Profiles

Database

French Ministry of the Economy, Finance and Industry and the

Agence Francais de Developpement

Conflicts of ethnic, religious, regional nature

Violent actions by underground political organizations Violent social conflicts

International Country Risk Guide

Political Risk Services

Internal conflict: Assesses political violence and its influence on governance External conflict: The external conflict measure is an assessment both of the risk

to the incumbent government and to inward investment

Government stability. Measures the government's ability to carry out its declared

programs, and its ability to stay in office

Ethnic tensions: This component measures the degree of tension within a country

attributable to racial, nationality, or language divisions Political Terror Scale University of

North Carolina

Political Terror Scale World Markets Online Global Insight

Business Risk and Conditions Global Insight

Civil unrest: How widespread Civil Un Civil Unrest: How widespread political

unrest is, and how great a threat it poses to investors. Demonstrations in themselves may not cause for concern, but they will cause major disruption if they escalate into severe violence.

Terrorism: Whether the country suffers from a sustained terrorist threat, and from

(38)

38 Table 5 Variable list

Variable name Variable type Measure Source

Product of service innovation

Dependent During the last three years, did your establishment: introduce into the market any new or significantly improved products (goods or services)?(Yes=1, No=0)

World Bank Enterprise surveys 2006/2007 Process innovation Dependent During the last three years, did your establishment:

introduce any new or significantly improved production processes including methods of supplying services and ways of delivering products?( Yes=1, No=0)

World Bank Enterprise surveys 2006/2007

World Governance Indicator (WGI) Political Instability and absence of violence/terrorism

Independent See table 4, Average 1995-2004 World Bank Governance Indicators

Battle-related deaths Independent robustness

Deaths in battle-related conflicts between warring parties in the conflict dyad, sum of total number of deaths between 1995-2004 divided by 1000

World Bank Development Indicator

Government change Independent robustness

Number of adverse regime changes times the magnitude of polity change, sum of 1955-2004

Center for Systemic Peace, Conflicts in Africa, Adverse Regime Changes in Africa Export orientation, direct

and indirect export share

Moderator robustness

In 2006(2007), what percentage of your establishment’s sales were: direct exports? + In 2006(2007), what percentage of your establishment’s sales were: indirect exports?

Export dummy Moderator 1 if export orientation >1%, 0 if not World Bank Enterprise surveys 2006/2007 Foreign ownership Moderator

robustness

What percentage of your firm is owned by: private foreign individuals, companies or organizations?

Foreign ownership dummy

Moderator 1 if foreign ownership >50%, 0 if not. World Bank Enterprise surveys 2006/2007 Formal training programs

dummy

Moderator In 2005, did this establishment run formal training programs for permanent, full-time employees? ( Yes=1, No=0)

World Bank Enterprise surveys 2006/2007 Firm size Firm control What was the total number of employees at the end of

2006(2007)?

Log transformed

World Bank Enterprise surveys 2006/2007 Firm performance Firm control Total sales minus total costs of labor minus total costs of

land equipment and buildings

Log transformed

World Bank Enterprise surveys 2006/2007 Firm age Firm control Year of survey - In what year did this establishment begin

its operations in this country?

Log transformed

Reliance on skilled labor Firm control Do you think that the following presents any obstacle to the current operations of your establishment: inadequately educated workforce? (1=totally disagree,

4=totally agree)

Foreign competitors Firm control How important is the influence of foreign competitors on your production costs? (1=not important, 4=very

important)

World Bank Enterprise surveys 2006/2007 Managerial experience Firm control How many years has the top manager worked: in a

managerial function in this sector?

(39)

39

Number of competitors Firm control How many competitors did you face at the end of 2006(2007)?

World Bank Enterprise surveys 2006/2007 State-owned Firm control What percentage of your firm is owned by the

state/government?

World Bank Enterprise surveys 2006/2007 Patent stock Country control (Number of U.S. patent applications, total between 1946

and 2004) divided by total population *10.000, Log transformed

USPTO statistic

FDI inflows Country control Foreign direct investment, net inflows (Current USD), average of 1995-2004

Log transformed

World Bank statistic

GDP per capita Country control GDP/population, World Bank statistic, average of 1995-2004

Log transformed

(40)

40 Table 6 Descriptive statistics

Variables Mean Std. Dev. Min Max

(41)

41 Table 7 Correlation matrix

Product/ service

Process WGI BRD Gov change

Exports %

ExpD For own%

How does political instability affect firm innovation?