Does Innovation Spur Export Performance across Countries? An Investigation from Meta-regression Analysis

Abstract

Innovation plays a major role in determining exports of a country by strengthening domestic industries. The narratives of innovation and exports are important for national competitiveness. Therefore, this study, utilizing a meta-regression analysis, examines the impact of innovation on export performances. We conduct a meta-analysis from 27 empirical studies that contain 249 estimates undertaken during 1996–2019 with an aim to test whether the results of empirical studies are sensitive to various measures utilized and recognize possible policy implications. This study finds that innovation affects export performance across countries. We find strong evidence that developed countries’ domestic innovation enhance their exports; however, for developing countries, innovation does not contribute to their exports. It indicates that within developing countries, the level of innovation efforts varies and concomitantly their inability to translate such efforts into exports.

Keywords

Innovation exports meta-analysis cross-country

I. Introduction

Endogenous growth theory postulates that technology adaptation, innovation and imitation are the key factors for the technological progress of a country (Romer, 1990). Such progress contributes to the economic growth of an economy by determining its international competitiveness. Vernon (1966) stresses the importance of technology factor in international competition based on country-specific advantage, and such competitive advantages result in innovating new products and processes. Studies consider innovation as an exogenous variable and envisage that innovation affects export performance of a country (Krugman, 1979; Posner, 1961; Vernon, 1966).

Schumpeter (1912) defines innovation, as an introduction of new product, better method of production, newer market, novel source of raw materials and better organization technique. Furthermore, innovation is categorized into technological innovation (such as product and process innovation) and non-technological innovation (such as market and organization change). Innovation of a country depends not only on input innovation (R&D) but also on output innovation (patent) that influence creation, adoption, adaptation, assimilation, diversification of technology and technological learning of the country that further stimulates knowledge intensive product. Considering innovation (namely, R&D, patent, product innovation and process innovations) and its connection to export decision, numerous studies underscore that innovation and new technology adoption enable firms to enter foreign market and enhance their export performances in developed countries (Basile, 2001; Dhanaraj & Beamish, 2003; Ganotakis & Love, 2011; Rodríguez & Rodríguez, 2005; Roper & Love, 2002; Yang et al., 2004). For example, few advanced economies, namely, the USA, Japan and Germany are identified as technology leaders and such economies are moving towards the world technology frontier, by original innovation and invention (Hofmann, 2013; Shin et al., 2016).

The existing evidence on the role of innovation in developing countries exports is mixed. For instance, in case of Brazil, Willmore (1992) finds that research and development (R&D) expenditure has no significant effect on its exports, whereas Kumar and Siddharthan (1994) suggest that the technology plays an important role in explaining the export performance of Indian enterprises. Chadha (2009) finds that foreign patent rights (PRs) (technology proxy) have a positive impact on Indian generic pharmaceuticals exports, by considering the later stage of product cycle development. She also suggests that developing countries have a potential to establish in the international market through innovation skills (by using patents). Developing countries’ innovation activity is a process of learning to use imported technologies efficiently rather than to innovating on the technological frontier (Lall, 2003) and its exporters face many problems in order to enter into the global market and access information, due to higher production cost. Hence, they are not directly involved in innovating and pushing the frontiers of knowledge. Instead, such economies acquire, adapt and improve the existing technologies from the international technology market. Moreover, Shin et al. (2016) argue that within the developing countries, the technology levels vary leading to a complex picture. For example, developing countries’ innovation is based on adaptive R&D for high-technology products. Evidently, countries export such products if these are not protected in the international markets.¹ Moreover, another developing countries innovation have evolved up the technology ladder in a few sectors to actually produce and export patented product that may be in the first stage of product cycle development.² Based on these above discussions, we conclude that the impact of innovation on export performance is not clear across developing countries, and empirical studies in this area are yet to reach a consensus about the relationship between innovation and exports. Therefore, this study, utilizing a meta-regression analysis (MRA), raises the following question: Does innovation spur export performance across countries?

This study aims to investigate such relationship by carrying out a review of previous empirical literatures using a MRA and tests whether the findings are sensitive to various measures employed, and hence identifying possible policy implications across countries. We went through 27 empirical studies containing 249 estimates during 1996–2019. We find that many indicators quantify innovation, namely, patent, R&D, product innovation, process innovation and other variables. Our analysis suggests that certain aspects of measuring innovation are crucial in explaining the significance of these findings. Our results indicate that innovation determines the export performance across countries. We also find strong evidence that developed countries’ domestic innovation enhance their exports; however, for developing countries, innovation does not contribute to their exports. It indicates that within developing countries, the level of innovation efforts varies that influence their ability to translate such efforts into exports.

This article is organized as follows. Following this introduction, Section II introduces the existing literatures, Section III presents the MRA database and econometric method and Section IV summarizes the regression results. Finally, Section V concludes and outlines the policy implication of the work.

II. Literature Review

The conventional trade theory (neo-classical international trade model) postulates that the difference in factor endowment measures country’s export, and establishes comparative advantage of countries based on factor endowments. It assumes that technology has no role to play, as it is freely available to all. However, such models have failed to provide due attention to the effect of technology on international trade. Later developments in the trade theory recognizes technology as an important determinant of international competitiveness and exports (Krugman, 1979; Posner, 1961; Vernon, 1966). According to the international trade models, developed by Vernon (1966), Krugman (1979), among others, suggest that innovation activities play a significant role for success in international markets, and these models argue that there is a positive linkage between innovation and exports. Trade and growth models envisage the relationship between innovation and exports. Usually, it is argued that countries’ export promotion strategy plays a significant role in economic growth prospects. Innovation plays a major role in determining exports and hence economic growth of a country by strengthening domestic industries.

There has been assumed that firms’ R&D investments benefit their competitiveness and long-run economic growth. However, effective innovation could only be occurred when firms’ innovations inputs translate into innovations outputs. A country involved in adaptive R&D will have high expenditure on research with fewer patents whereas novel products would require high spending on research leading to patenting. For example, this case represents countries in the second or third stage of the product cycle hypothesis. In the knowledge production function, the investment into R&D strengthens the stock of knowledge in a country that leads to innovation, and further raises their output (Griliches, 1979). The motivation is to patent, the principal innovation output of the R&D investment—in such cases, higher investment in innovation input would encourage inventions, innovations and patents (Griliches, 1990). A nation’s innovation capabilities, technology growth and knowledge capital would improve through effective R&D activities. Moreover, as countries with patent protection develop greater technology. Hence, such protection further stimulates domestic innovation (Ginarte & Park, 1997; Park, 2008). Therefore, this study, based on the primary survey literature, find many indicators that measure innovation, namely, patent, R&D, product innovation, process innovation and other variables.

According to the product cycle concept, ‘the product cycle hypothesis begins with the assumption that the stimulus to innovation is typically provided by some threat or promise in the market. But according to the hypothesis, firms are acutely myopic; their managers tend to be stimulated by the needs and opportunities of the market closest at hand, the home market’ (Vernon, 1979, p. 256). Krugman (1979) suggests that technological diffusion is an important element of international trade, primarily, North (developed countries) innovates that gets diffused to the South (developing countries), and it is shaping the trade. In the South, the diffusion of technological innovation occurs through a certain level of imitative, adaptive and absorptive capability. A study by Lall (2000), through an analysis of the relationship between technological structure and manufactured exports performance, finds that developing countries are exporters of high-tech products. This study concludes that there is a significant performance of high-tech exports, which may be ‘something statistical illusion’ following from the specialization in the labor-intensive processes within high-tech-intensive industries. Interestingly, the growth of high-tech exports is because of the technology spurts or international production sharing (Mani, 2000; Srholec, 2007). Mani (2000) finds that majority of developing countries’ high-tech exports are due to multinational enterprises with very little local R&D. Srholec (2007) suggests that international fragmentation of production plays an important role for the significant performances of developing countries’ in high-tech exports. Singh and Chawla (2018) found a rising trend of high-tech and medium-high-tech industrial R&D in total industrial R&D. They suggested that structural transformation in industrial R&D is positively correlated with structural change in industrial output in India.

The narratives of innovation and exports, in empirical studies, are important for national competitiveness at the country (macro) and firm (micro) levels. At country levels, innovation stimulates industrial productivity and exports growth, and innovation measures firms’ competitiveness at the micro level. From the systematic review, we find that empirical studies used survey data and firms level data. Zhao and Li (1997) study the relationship between R&D and export propensity of China’s manufacturing firms. The study finds that there is a positive influence of R&D on export propensity and export growth. Beise-Zee and Rammer (2006) investigate the impact of local adaptation of innovation on exports and test whether certain local market’s characteristics influence exportability of innovation by using survey data from the German innovation survey of 4,786 firms in the manufacturing and service industries. They find that domestic demand structure and export orientation encourage exports success. By examining the impact of innovation on exports of Vietnam small- and medium-sized enterprises (SMEs), Nguyen et al. (2008) suggest that innovation determines Vietnamese SMEs exports. AñónHigón and Driffield (2011) examine the relationship between innovation activities (distinguishing product from process innovation) and export performance of the UK SMEs. Their study indicates that innovation activities stimulate exports, conditional upon the independent of product and process innovation. Nevertheless, they do not find a positive link between process innovation and exports beyond the product innovation by considering interdependence between both the innovation activities. In case of China and Spain, Aw et al. (2007) do not find a significant relationship between firm-level R&D and export performances.

Empirical studies focus on the role of innovation in advanced countries’ trade. Soete (1987) studies organization for economic co-operation and development (OECD) countries’ 40 industries and suggests that patents play a significant role in a country’s export performance. Van Hulst et al. (1991) find that there is a positive association between the pattern of export specialization and the technology specialization in the case of Germany, the Netherlands and Sweden. Caldera (2010), investigates the relationship between innovation and export behavior of Spanish firms, finds a positive effect of firm innovation on export performances. Ganotakis and Love (2011) study the relationship between R&D, product innovation and exporting for a sample of new technology-based UK firms. By using a recursive system of the R&D-innovation-exporting relationship, this study finds innovator are certainly expected to export; however, there is no evidence about the positive impact of innovation on successive export intensity.

In the context of developing countries, studies find mixed evidence on the relationship between innovation and export performances. Dasgupta and Siddharthan (1985) suggest that largely goods of Indian exports consist of low technology. In case of Brazil, Willmore (1992) finds that R&D expenditure has no significant effect on its exports. Guan and Ma (2003) find that innovation capability dimensions are important in determining Chinese firms’ export performances. Moreover, Bhat and Narayan (2009) argue that an achievement of technological capabilities (in-house R&D) is a significant in determining export performances of Indian chemical industry. Furthermore, Molina-domene and Pietrobelli (2012) find that technological capabilities positively influence export performance in Latin American countries. Moreover, Montobbio and Rampa (2005) find the relationship between technological activities and export performances are different in low-tech, medium-tech and high-tech exports, as these are dissimilar in terms of learning potential, growth opportunities, scope of upgrading and spillover to the rest of the economy.

Studies show that the influence of patents on innovation rather varies considerably across sectors (Allred & Park, 2007; Sharma et al., 2018). Such innovation capabilities are expected to be translated into country’s export competitiveness (Panda & Sharma, 2020). Moreover, PRs protection is an institutional factor that supports the innovation of a country. As suggested in the promotional channel of gains from PRs, strong protection is expected to stimulate domestic innovation, whereby a firm may invest more in R&D in the expectation that it will profit from the newly developed product or process. The empirical evidence underscores that there is no straightforward answer as the net impact of the patent protection on innovation is conditioned by different factors. Developing countries are also not directly involved in innovating and pushing the frontiers of knowledge. Instead, such economies acquire, adapt and improve the existing technologies from the international technology market. Therefore, it is interesting to examine the impact of innovation on export performances across countries by using an MRA.

III. Data and Methodology

This study analyzes the impact of innovation on exports by utilizing a MRA. MRA harmonizes empirical survey results, combining the findings of various studies that use different data and methodologies and present a clear and consensual descriptive result to provide clear visions by challenging estimations with actual research. According to Glass,

meta-analysis refers to the statistical analysis of a large collection of results from individual studies for the purpose of integrating the findings. It connotes a rigorous alternative to the causal, narrative discussions of research studies which typify our attempts to make sense of the rapidly expanding research literature. (Glass, 1976, p. 3)

MRA is a specific method that explains the heterogeneity in the effect sizes reported by primary studies. Following Stanley and Jarrell (2005), MRA comprises the estimation of a standard regression model as follows:

b_{j} = β + \sum_{k}^{k} α_{k} Z_{k j} + e_{j} (j = 1, 2, \dots L),

(1)

where, b_j is the reported estimates of β of the jth study in the literature included of L studies; Z_kj corresponds to the K meta-independent variables, which measures relevant characteristics of the empirical studies and it shows the systematic difference from other studies’ results. a_k is the MRA coefficient that represents the bias effect of specific study characteristics, and e_j is the meta-regression disturbance term. MRA produces the empirical studies by identifying pertinent characteristics of the empirical studies and showing those changes in Z_kj. For the empirical economic investigations, reference of data determines the final specification of the model. Based on the available literature, we capture these differences by including dummy variables of primary studies, namely, journal; this variable is a binary that takes the value 1 if study is published in journal article, 0 if otherwise; Country level, takes value 1 if country level data is used; 0 if otherwise. And GDP, size, distance and profitability are also other dummy variables of primary studies (Table 1).

Table 1.

Moderating Variables

Variables	Type	Measures
Journal	Dummy	1 if study is published in journal article, 0 if otherwise
Country level	Dummy	1 if country level data is used, 0 if otherwise
GDP	Dummy	1 if study includes GDP as a control variable, 0 if otherwise
Distance	Dummy	1 if study includes distance as a control variable, 0 if otherwise
Profitability	Dummy	1 if study includes profitability as a control variable, 0 if otherwise
Size	Dummy	1 if study includes size as a control variable, 0 if otherwise

Source: Authors’ calculation.

Data

This study aims to examine the relationship between exports and innovation by utilizing MRA. We follow the four strategies of preferred reporting items for systematic reviews and meta-analysis (PRISMA) method to select articles, that is, identification, screening, eligibility and the inclusion. With regard to the identification or literature search, several strings/keywords reference to ‘innovation’ and ‘exports’; ‘innovation and exports’ since 1996 were used to search articles in the Google Scholar, JSTOR and other sources, and builds the sample from the English language sample. In doing so, the records identified through systematic searching a total of 325. Records after duplication removed, we have identified a total of 193. We have screened 193 paper and we find full-text articles assessed for eligibility is 55 after excluding criteria based on two reasons, namely, (a) other than empirical paper and (b) published before 1995 (TRIPs). After this exclusion and inclusion criteria, this study includes 27 studies for the MRA.

Keeping in mind the objectives of our study, we include 27 empirical studies after screening and eligibility strategies that contain 249 estimates fulfill the criteria for our MRA. Out of the 27 papers, two are conference papers. Interestingly, maximum studies employ panel data analysis. Most papers are country-level studies and firm-specific studies, and they use survey data as well. Fifteen studies are on the developed world, eight studies are on developing countries and only four studies are on mixed (including both developing and developed countries). Table 2 presents the list of papers included in the meta-regression database, number of estimates and average effect size of the studies. The estimates of innovation on exports vary between −11.012 and 36.22, with 231 positive value and only 18 negative values.

Nature of the Studies Selected

The shortlisted papers are published in a wide range of academic journals (Table 3). Half of the studies are found in economics and development journals, with Review of World Economics (3) followed by World Development (2), Oxford Development Studies (2) and others. Innovation journals publish almost a fifth of the studies, with Economics of Innovation and New Technology (2) followed by Research Policy (1) and Technological Forecasting & Social Change (1). Moreover, management studies cover 26% of the studies, with Small Business Economics (2) and International Business Review (2). Nonetheless, working paper and conference proceeding publish less than 10% of sample total publications.

Table 2.

List of Papers Included in the Meta-analysis Database: Innovation and Exports

Studies	No. of Country	Time Period	Measurement of Innovation	No. of Estimates	Ave Effect Size
AñónHigón and Driffield (2011)	1	2004	Mixed	8	0.207
Andersson and Ejermo (2008)	1	2003	Patent	4	0.165
Bhaduri and Ray (2004)	1	1994–1995	R&D	8	2.275
Beise-Zee and Rammer (2006)	1	1999	R&D	6	2.263
Bhat and Narayan (2009)	1	2001–2007	R&D	5	2.804
Bravo-Ortega et al. (2014)	1	1997–2004	R&D	4	7.06875
Caldera (2010)	1	1991–2002	R&D	23	0.124
Chadha (2009)	1	1989–2004	Patent	2	0.067
Filipescu (2013)	1	1994–2005	Mixed	6	0.209
Franco and Sasidharan (2010)	1	1994–2006	R&D	12	0.8925
Ganotakis and Love (2011)	1	2004	Mixed	8	2.658
Girma et al. (2008)	2	1996–2003	R&D	12	0.202
Goldar (2013)	1	1999–2010	R&D	3	−0.115
Lachenmaier and Woßmann (2006)	1	2002	Mixed	17	5.544
Montobbio and Rampa (2005)	9	1985–1998	Patent	4	1.84
Nguyen et al. (2008)	1	2005	Mixed	12	0.643
Panda et al. (2020)	67	1996–2014	Technological effort	6	18.807
Pla-Barber and Alegre (2007)	1	2002	Innovation outcome	2	0.5905
Rasiah et al. (2016)	1	2013	Innovation capabilities	1	0.628
Rodríguez and Rodríguez (2005)	1	1998–1999	Mixed	16	2.101
Roper and Love (2002)	2	1991–1994	Mixed	12	3.193
Sandu and Ciocanel (2014)	26	2006–2010	R&D	4	8.87
Shin et al. (2016)	70	2000–2007	Patent	42	1.411
Srholec (2007)	111	2003	Technological capabilities	3	0.277
Wang and Guan (2009)	17	1991–2005	Patent	4	1.103
Yang et al. (2004)	1	1996	Mixed	23	0.092
Zhao and Li (1997)	1	1991	R&D	2	0.335
Total				249

Source: Authors’ calculation.

Most studies use patent and R&D data as the measure of innovation, some studies use mixed measures (including R&D and product innovation, R&D and process innovation and other variables), studies also use different indicators to measures innovation, that is, technological efforts, innovation outcome, innovation capabilities and technological capabilities.

Table 3.

Articles by Journal Categories

Economics and Development Studies	Innovation Journals	Management Studies	Working Paper/Conference Proceeding
Oxford Development Studies (2)	Economics of Innovation and New Technology (2)	International Business Review (2)	Procedia Economics and Finance (1)
Emerging Markets Finance and Trade (1)	Innovation and Development (1)	International Small Business Journal (1)	IEEE International Conference on Intelligent Computing and Intelligent Systems (1)
Review of World Economics (3)	Technological Forecasting & Social Change (1)	Journal of International Marketing (1)
World Development (2)	Research Policy (1)	Small Business Economics (2)
Economic Systems (1)		Journal of High Technology Management Research (1)
Oxford Economic Papers (2)
The European Journal of Development Research (1)
Journal of Developing Areas (1)
The World Economy (1)
Total: 13 (48%)	Total: 5 (18)	Total: 7 (26%)	Total: 2 (7%)

Source: Authors’ calculation.

Note: Number of studies in parenthesis.

Publication Bias and Estimation of Effect Size

There is a need to test the publication trends adopted by authors or journals that relate to the direction of results or significance of coefficients (Neves et al., 2016). Hence, a funnel plot is used to graphically examine this issue. Egger et al. (1997) explain this by using a funnel plot, cited by Neves et al. (2016), a scatter plot of effects sizes (in the horizontal axis) against their precision, 1/SE (in the vertical axis). If the graph appears in the shape of an inverted funnel with no asymmetries, then there may be no publication bias. Figure 1 is the funnel plot of the estimated effect sizes against their precision, 1/SE and it presents the impact of innovation on exports meta-sample. This graph seems to be asymmetric about the ‘true effect’ sizes, and it reveals the existence of a publication bias (Figure 1).

Figure 1.

Source: Authors’ calculation.

The graphic examinations may sometimes be misleading and may not be accurate. Therefore, a funnel asymmetric test (FAT) is conducted to explore this systematically, and it is like the funnel plot, however, with more robustness (Stanley, 2005). Consequently, this study presents the following equation by using a simple regression of the effect sizes with respect to the respective SE,

E f f e c t S i z e_{i j} = α_{1} + α_{2} S E_{i j} + µ_{i t}

(2)

where i = 1,…, 249 the individual regression estimates reported, j = 1,…, 27 the studies in the meta-database and μ_it is the error term.If the effect size will be correlated with standard errors, then there is a publication bias, which leads to higher standard errors and higher value of the effect sizes as well. In order to solve this problem of heteroscedasticity, usually, Equation (2) is weighted by the SEs associated with each observation (Stanley, 2005). So, following Equation (3) is the weight least square (WLS) of Equation (2) by dividing both sides by SE_ij that yields more efficient estimates:

t_{i j} = β_{1} + β_{2} {\frac{1}{S E}}_{i j} + e_{i t}

(3)

where t_ij is the t-value of the estimated coefficient from estimate i of study j. The intercept, β₁ (β₁ ≠ 0), and slope, β₂, coefficients are to be tested if these are statistically different from zero. Nevertheless, there is no issue of heteroscedasticity in Equation (3), and we estimate the equation by an ordinary least squares (OLS) regression of the t-statistics with respect to their precision $({\frac{1}{S E}}_{i j})$ .

Table 4 shows that the coefficient is positive and significant, which shows an existence of publication bias in these empirical studies. The possible reason of the publication bias is journalistic trends for publishing papers considering statistically significant results of innovation on exports, as discussed earlier that the estimates, from the primary studies (27 studies), are having more than 90% of positive values.

IV. Empirical Results

To examine the contributions of innovation to export performances, we utilize a simple OLS regression model for MRA, and we also employ a random effect maximum likelihood (REML) model to examine the robustness of the coefficients. Table 5 presents the results for the MRA.

Table 4.
Estimation of Equation (3)

Variables Coefficients

Precision 0.00006*

(0.00003)

Constant 2.98*

(0.15)

Observation Studies 249
27

Variables	Coefficients
Precision	0.00006*
	(0.00003)
Constant	2.98*
	(0.15)
Observation Studies	249 27

Source: Authors’ calculation.

Note: *Denote statistical significance at the 0.1 level. Standard errors are in parentheses.

We find strong evidences that the existing studies’ precision coefficient is positive and significant which suggest a positive effect of innovation on export success. By the type of countries, we find robust evidence that the sign of developed countries’ coefficient is positive and highly significant in both the models. This reveals that innovation plays an important role in stimulating developed countries exports. However, developing countries’ coefficient is insignificant, highlights that innovation does not contribute to their exports. In other words, their innovation falls short of some threshold level. In fact, some countries may even export such products if they are not protected by patents or are off-patents in the international markets. Our results corroborate earlier results of a study, Panda et al. (2020). A probable reason is that, within developing countries, the innovation varies, and some developing countries may be engaged in adaptive R&D for high-technology products that are in the second or third stage of product cycle development. Park (2008) also suggests that the adoption of stronger patent protection laws and the usage of PRs vary across countries according to their levels of economic development.

Table 5.

Estimation of Meta-regression Analysis

Variables	All Studies		Developed Countries Studies		Developing Countries Studies		Mixed Studies
Variables	1	2	1	2	1	2	1	2
Precision	0.0001***	0.00008**	0.00007***	0.00007***	−0.02	−0.02	−0.12**	−0.12***
	(0.00001)	(0.00003)	(0.00001)	(0.00003)	(0.02)	(0.02)	(0.03)	(0.03)
Country level	0.33	0.33	0.61	0.61	1.27	1.27
	(0.55)	(0.52)	(0.55)	(0.55)	(0.91)	(0.91)
Journal	0.61	0.61	−0.77	−0.77	0.89	0.89	−0.12	−0.12
	(1.05)	(0.81)	(0 .64)	(1.19)	(1.34)	(1.34)	(1.75)	(1.75)
GDP	1.34	1.34*	0.53	0.53
	(0.89)	(0.73)	(1.47)	(1.17)
Size	0.29	0 .29	−0.42	−0.42			9.03***	9.03***
	(0 .45)	(0.45)	(0.45)	(0.5)			(2.06)	(2.06)
Profit	0.89**	0.89*	1.82**	1.82**	0.39	0.39
	(0.44)	(0.47)	(0.66)	(0.59)	(0.69)	(0.69)
distance	0.21	0.21	2.2	2.2*			2.3**	2.3**
	(0.94)	(0.75)	(1.69)	(1.33)			(1.16)	(1.16)
Constant	1.5	1.5	2.9	2.9**	0.74	0.74	2.38*	2.38*
	(1.03)	(0.75)	(0.49)	(1.1)	(1.59)	(1.59)	(1.39)	(1.43)
Observation	249	249	146	146	48	48	55	55
Studies	27	27	15	15	8	8	4	4

Source: Authors’ calculation.

Note: *, ** and *** Denote statistical significance at the 0.1, 0.05 and 0.01 levels, respectively. Robust standard errors are in parentheses. Model (1) is the OLS regression results; model (2) is the REML regression results, respectively.

The coefficient of journal is insignificant in all the models. Reason for the result could be very less variations since most studies are from journal articles with less than 10% of total publications as working papers and conference proceedings. Our result shows that GDP is positive and significant highlighting that countries’ economic activities play a major role in exports performances. We also find that firms’ profitability also enhances developed countries’ export performances. Interestingly, we also find that countries’ size of the firm, as measured by net fixed assets in most of the primary studies, positively stimulate export performance.

V. Conclusion and Policy Implications

This study examines the impact of innovation on exports performances across countries, using a MRA from 27 empirical studies during 1996–2019. We provide a quantitative synthesis of the effect size estimates reported in empirical survey and find reasons for the inconsistencies in the empirical findings. Furthermore, this study demonstrates the policy implications about different measures of innovation and its impact on exports by the types of countries in analysis. This study finds that empirical results are driven by publication bias, which means that journal editors are attracted to publish studies with significant positive results. Furthermore, adjusting for this bias, we find that countries’ innovation contributes their export performances.

The synthesis of narrative reviews and MRA reveals that innovation determines exports success across countries. By the form of countries, we find strong evidences which indicate that developed countries’ domestic innovation enhances their exports; notwithstanding, for developing countries, innovation does not contribute to their exports. It indicates that within developing countries, the level of innovation efforts varies, and concomitantly these countries are unable to translate such efforts into exports. The results are also not as compelling for developing countries owing to the limited variation in their innovation efforts.

We suggest that there is divergence in innovation efforts among middle income countries and concomitantly their inability to translate these same efforts in exports. Developing economies are not directly involved in innovating and pushing the frontiers of knowledge. Instead such economies acquire, adapt, and improve the existing technologies from the international technology market. Hence, there appears to be some implied minimum economic development that needs to be reached for innovation efforts to be a determining factor. An avenue for future work is to study source country outward foreign direct investment (FDI) activities and export performances by utilizing MRA. Under theories of internalization, it is well known that firms choose among different modes of entry into foreign markets, with exporting and FDI being the key modes.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Notes

References

Allred

B. B.

, & Park

W. G.

(2007). Patent rights and innovative activity: Evidence from national and firm-level data. Journal of International Business Studies, 38(6), 878–900.

Andersson

, & Ejermo

(2008). Technology specialization and the magnitude and quality of exports. Economics of Innovation and New Technology, 17(4), 355–375.

AñónHigón

, & Driffield

(2011). Exporting and innovation performance: Analysis of the annual Small Business Survey in the UK. International Small Business Journal, 29(1), 4–24.

B. Y.

, Roberts

M. J.

, & Winston

(2007). Export market participation, investments in R&D and worker training, and the evolution of firm productivity. World Economy, 30(1), 83–104.

Basile

(2001). Export behaviour of Italian manufacturing firms over the nineties: The role of innovation. Research Policy, 30(8), 1185–1201.

Beise-Zee

, & Rammer

(2006). Local user-producer interaction in innovation and export performance of firms. Small Business Economics, 27(2–3), 207–222.

Bhaduri

, & Ray

A. S.

(2004). Exporting through technological capability: Econometric evidence from India’s pharmaceutical and electrical/electronics firms. Oxford Development Studies, 32(1), 87–100.

Bhat

, & Narayanan

(2009). Technological efforts, firm size and exports in the basic chemical industry in India. Oxford Development Studies, 37(2), 145–169.

Bravo-Ortega

, Benavente

J. M.

, & González

Á.

(2014). Innovation, exports, and productivity: Learning and self-selection in Chile. Emerging Markets Finance and Trade, 50(Suppl. 1), 68–95.

10.

Caldera

(2010). Innovation and exporting: Evidence from Spanish manufacturing firms. Review of world Economics, 146(4), 657–689.

11.

Chadha

(2009). Product cycles, innovation, and exports: A study of Indian pharmaceuticals. World Development, 37(9), 1478–1483.

12.

Dasgupta

, & Siddharthan

(1985). Industrial distribution of Indian exports and joint ventures abroad. Development and Change, 16(1), 159–174.

13.

Dhanaraj

, & Beamish

P. W.

(2003). A resource-based approach to the study of export performance. Journal of Small Business Management, 41(3), 242–261.

14.

Egger

, Smith

G. D.

, Schneider

, & Minder

(1997). Bias in meta-analysis detected by a simple, graphical test. BMJ, 315(7109), 629–634.

15.

Filipescu

D. A.

, Prashantham

, Rialp

, & Rialp

(2013). Technological innovation and exports: Unpacking their reciprocal causality. Journal of International Marketing, 21(1), 23–38.

16.

Franco

, & Sasidharan

(2010). MNEs, technological efforts and channels of export spillover: An analysis of Indian manufacturing industries. Economic Systems, 34(3), 270–288.

17.

Ganotakis

, & Love

J. H.

(2011). R&D, product innovation, and exporting: Evidence from UK new technology-based firms. Oxford Economic Papers, 63(2), 279–306.

18.

Ginarte

J. C.

, & Park

W. G.

(1997). Determinants of patent rights: A cross-national study. Research Policy, 26(3), 283–301.

19.

Girma

, Görg

, & Hanley

(2008). R&D and exporting: A comparison of British and Irish firms. Review of World Economics, 144(4), 750–773.

20.

Glass

G. V.

(1976). Primary, secondary, and meta-analysis of research. Educational Researcher, 5(10), 3–8.

21.

Goldar

(2013). R&D intensity and exports: a study of Indian pharmaceutical firms. Innovation and Development, 3(2), 151–167.

22.

Griliches

(1979). Issues in assessing the contribution of research and development to productivity growth. The Bell Journal of Economics, 10(1), 92–116.

23.

Griliches

(1990). Patent statistics as economic indicators: A survey part 2. National Bureau of Economic Research.

24.

Guan

, & Ma

(2003). Innovative capability and export performance of Chinese firms. Technovation, 23(9), 737–747.

25.

Hofmann

(2013). The impact of international trade and FDI on economic growth and technological change. Springer Science & Business Media.

26.

Krugman

(1979). A model of innovation, technology transfer, and the world distribution of income. Journal of Political Economy, 87(2), 253–266.

27.

Kumar

, & Siddharthan

N. S.

(1994). Technology, firm size and export behaviour in developing countries: The case of Indian enterprises. The Journal of Development Studies, 31(2), 289–309.

28.

Lachenmaier

, & Wößmann

(2006). Does innovation cause exports? Evidence from exogenous innovation impulses and obstacles using German micro data. Oxford Economic Papers, 58(2), 317–350.

29.

Lall

(2000). The technological structure and performance of developing country manufactured exports, 1985–98. Oxford Development Studies, 28(3), 337–369.

30.

Lall

(2003). Indicators of the relative importance of PRs in developing countries. Research Policy, 32(9), 1657–1680.

31.

Mani

(2000). Exports of high technology products from developing countries: Is it real or a statistical artifact? (Discussion Paper No. 2000-1). United Nations University, Institute for New Technologies.

32.

Molina-Domene

M. A.

, & Pietrobelli

(2012). Drivers of technological capabilities in developing countries: An econometric analysis of Argentina, Brazil and Chile. Structural Change and Economic Dynamics, 23(4), 504–515.

33.

Montobbio

, & Rampa

(2005). The impact of technology and structural change on export performance in nine developing countries. World Development, 33(4), 527–547.

34.

Neves

P. C.

, Afonso

Ó.

, & Silva

S. T.

(2016). A meta-analytic reassessment of the effects of inequality on growth. World Development, 78, 386–400.

35.

Nguyen

A. N.

, Pham

N. Q.

, Nguyen

C. D.

, & Nguyen

N. D.

(2008). Innovation and exports in Vietnam’s SME sector. The European Journal of Development Research, 20(2), 262–280.

36.

Panda

, & Sharma

(2020). Does technological specialization spur high-technology exports? Evidence from panel quantile regressions. Global Economy Journal, 20(02), 2050013.

37.

Panda

, Sharma

, & Park

W. G.

(2020). Patent protection, technological efforts, and exports: An empirical investigation. The Journal of Developing Areas, 54(2), 145–162.

38.

Park

W. G.

(2008). International patent protection: 1960–2005. Research Policy, 37(4), 761–766.

39.

Pla-Barber

, & Alegre

(2007). Analysing the link between export intensity, innovation and firm size in a science-based industry. International Business Review, 16(3), 275–293.

40.

Posner

M. V.

(1961). International trade and technical change. Oxford Economic Papers, 13(3), 323–341.

41.

Rasiah

, Shahrivar

R. B.

, & Yap

X. S.

(2016). Institutional support, innovation capabilities and exports: Evidence from the semiconductor industry in Taiwan. Technological Forecasting and Social Change, 109, 69–75.

42.

Rodríguez

J. L.

, & Rodríguez

R. M. G.

(2005). Technology and export behaviour: A resource-based view approach. International Business Review, 14(5), 539–557.

43.

Romer

P. M.

(1990). Endogenous technological change. Journal of Political Economy, 98(5, Part 2), S71–S102.

44.

Roper

, & Love

J. H.

(2002). Innovation and export performance: evidence from the UK and German manufacturing plants. Research Policy, 31(7), 1087–1102.

45.

Sandu

, & Ciocanel

(2014). Impact of R&D and innovation on high-tech export. Procedia Economics and Finance, 15, 80–90.

46.

Schumpeter

J. A.

(1912). Theory of economic development. Oxford University Press.

47.

Sharma

, Paswan

A. K.

, Ambrammal

S. K.

, & Dhanora

(2018). Impact of patent policy changes on R&D expenditure by industries in India. The Journal of World Intellectual Property, 21(1–2), 52–69.

48.

Shin

, Lee

, & Park

W. G.

(2016). When an importer’s protection of IPR interacts with an exporter’s level of technology: Comparing the impacts on the exports of the North and South. The World Economy, 39(6), 772–802.

49.

Singh

, & Chawla

(2018). Interconnection between R&D and industrial production structure: The experience of India. Millennial Asia, 9(3), 278–294.

50.

Soete

(1987). The impact of technological innovation on international trade patterns: The evidence reconsidered. Research Policy, 16(2–4), 101–130.

51.

Srholec

(2007). High-tech exports from developing countries: A symptom of technology spurts or statistical illusion? Review of World Economics, 143(2), 227–255.

52.

Stanley

T. D.

(2005). Beyond publication bias. Journal of Economic Surveys, 19(3), 309–345.

53.

Stanley

T. D.

, & Jarrell

S. B.

(2005). Meta-regression analysis: A quantitative method of literature surveys. Journal of Economic Surveys, 19(3), 299–308.

54.

Van Hulst

, Mulder

, & Soete

L. L.

(1991). Exports and technology in manufacturing industry. Weltwirtschaftliches Archiv, 127(2), 246–264.

55.

Vernon

(1966). International trade and international investment in the product cycle. Quarterly Journal of Economics, 80(2), 190–207.

56.

Vernon

(1979). The product cycle hypothesis in a new international environment. Oxford Bulletin of Economics and Statistics, 41(4), 255–267.

57.

Wang

G. B.

, & Guan

J. C.

(2009, November). The impact of technological innovation on export performance: Evidence from a cross-country analysis. In 2009 IEEE International Conference on Intelligent Computing and Intelligent Systems (Vol. 3, pp. 393)–397). IEEE.

58.

Willmore

(1992). Tansnationals and foreign trade: Evidence from Brazil. The Journal of Development Studies, 28(2), 314–335.

59.

Yang

C. H.

, Chen

J. R.

, & Chuang

W. B.

(2004). Technology and export decision. Small Business Economics, 22(5), 349–364.

60.

Zhao

, & Li

(1997). R&D and export: An empirical analysis of Chinese manufacturing firms. The Journal of High Technology Management Research, 8(1), 89–105.

Does Innovation Spur Export Performance across Countries? An Investigation from Meta-regression Analysis

Abstract

Keywords

I. Introduction

II. Literature Review

III. Data and Methodology

Data

Nature of the Studies Selected

Publication Bias and Estimation of Effect Size

Table 4. Estimation of Equation (3) Variables Coefficients Precision 0.00006* (0.00003) Constant 2.98* (0.15) Observation Studies 24927

Footnotes

Declaration of Conflicting Interests

Funding

Notes

References

Table 4.
Estimation of Equation (3)

Variables Coefficients

Precision 0.00006*

(0.00003)

Constant 2.98*

(0.15)

Observation Studies 249
27