How Does Government Information Access Interplay with Resources of Emerging Market Small and Medium-Sized Enterprises for Innovation? Evidence from Vietnam

Abstract

This study integrates the (extended) resource-based view (ERBV/RBV) and non-market strategy (i.e., corporate political activity) to investigate the role of internal and external resources as drivers of innovation in small and medium-sized enterprises in emerging markets (ESMEs). Using primary data from 192 Vietnam SMEs collected between 2014 and 2016, we adopted a partial least square estimation to examine our hypotheses, supplemented by ordinary least square and unobserved heterogeneity tests for robustness. Our PLS-SEM results reveal that firms with intangible resources, from human capital investment to political connections, are more likely to innovate. While the positive effect of human capital investment on innovation is not moderated by government information access, the impact of political connections as an extended resource is significantly enhanced by access to government information. On the other hand, international export experience is not related to innovation by itself, but it is positively moderated by government information access, suggesting that engaging in international markets alone is insufficient. This study contributes to the emerging market literature by examining the effects of intangible resources and political connections on ESMEs’ innovation and highlighting the role of government information as a non-market strategy access in enhancing these relationships.

1. Introduction

Vietnam’s economy is growing yearly, and the economic growth rate reached 6.76% in the first half of 2019, with limitless growth potential. Still, new investment, increased investment, and official development assistance (ODA) are pouring into Vietnam from Korea, Japan, Singapore, Taiwan, Hong Kong, and China, leading to accelerated growth in the innovation sector. Since becoming a developing country in 2010, Vietnam has maintained an average annual GDP growth rate of 6% until 2018, with per capita GDP reaching $2600. The World Innovation Index ranked it 42nd out of 129 countries and third among ASEAN countries. The export strategy also contributed to Vietnam’s outgoing industrial innovation structure, achieving remarkable economic growth. However, since Vietnam is an economy integrated with a plethora of small and medium-sized enterprises (SMEs) that emphasize innovation, firms are inherently limited in size, thus striving to obtain the utmost performance. Naturally, Vietnam is so engrossed in innovation because there is no better instrument than innovation for SMEs to leapfrog large firms and enter the competition. Thus, innovation is the most rational and decisive outcome for SMEs in emerging markets such as Vietnam to technologically grow and socially change.

Since innovation is a resource-intensive activity, it is highly costly for many SMEs in emerging markets [1]. Compared with large multinational enterprises (MNEs) or SMEs in developed economies, emerging market SMEs (ESMEs) are subject to more severe resource constraints, thus facing more barriers to innovation [2]. Yet, innovation is essential for the survival of firms, especially in the midst of fierce global competition [3]. Thus, fostering innovation, especially for small firms in developing countries, has remained an important public policy issue. Although there are several ways to define innovation performance, we define it as a firm’s innovative changes in process and production [4,5]. Product and process innovation are proven to be important for the sustainable growth of firms because they contribute to higher profits, increased market share, and eventually, a higher chance of survival [6,7,8,9].

Because of their significance, there has been extensive literature on the drivers of product and process innovation. Some look at internally intangible resources, such as human capital [1] and social capital [10], while others consider external resources, such as locations, networks, and related institutions. Among the abovementioned factors critical for innovation, we put forth intangible resources and corporate political activity [11,12]. However, less attention has been paid to studies examining the importance of these intangible resources in EMSEs, particularly regarding the contingencies and the integration of political connections and access to political information. To fill this research gap, we analyze the effect of three internal intangible resources on innovation using the firm-level data across the ten provinces in the Red River Delta region in Vietnam from 2014 to 2016. From a resource-based perspective (RBV), we assumed human capital investment and international experiences as central factors. This study selects political connections as a firm’s internal intangible resource for firm competence as a theoretical lens for extended RBV. In other words, we endeavor to verify the intangible resources’ effect on the innovation capacity of ESME based on the theoretical lens of RBV and ERBV. In addition, we pay attention to the moderating role of government information as a corporate political activity.

To address these calls, we attempt to fill the research gap with existing research by examining the mechanism through which intangible resources operate among the various resources that ESMEs adopt to induce innovation, and why corporate political activities play an important role in emerging markets. In emerging economies with inadequate market information, the usefulness of government information is undeniable. In Vietnam, a transitional, single-party economy, the public sector still controls the allocation of many key resources, including information [10]. We observe how much government information gained through corporate political activity interacts with firms’ intangible resources to innovate. We propose that firms equipped with compatible corporate capabilities can understand the importance of information from the government and internalize it to conduct innovative activities [13,14,15,16,17].

Our results suggest that SMEs which invest in human capital and SMEs which have political connections are likely to engage in product and process innovation, whereas international experience is not a significant prerequisite for innovation. If those SMEs with internal resources receive valuable government information through external knowledge sources, the likelihood of innovation increases for firms with international experience and political connections. In other words, if a firm’s owner or family has both political experience and access to information from the government, the positive impact of political experience is amplified.

The contributions of this study are threefold. First, we contribute to the RBV and the ERBV literature by explaining the moderating effect of corporate political activity on various types of intangible resources. Although there are many previous findings on the role of internal resources and external networks, there is a research gap in considering the combined effect of the two sources. We assume that, because of different levels of corporate capabilities based on intangible resources, access to government information will have heterogeneous effects across firms.

Second, we portray the driving factors of innovation from the perspective of early-stage ESMEs. Although there has been much empirical research among large firms or firms in developed countries such as France [18] and Canada [19], not much empirical evidence is available for family-oriented, underdeveloped, traditional ESMEs. The potential marginal benefit that ESMEs can yield from innovation is supposedly high; thus, this research can provide managerial and policy implications for those firms struggling to innovate.

Third, we uncover the importance of government information. The literature that analyzes the relationship between social capital from government/political ties and innovation presents several significant gaps, since it mostly focuses on institutional support, such as financial benefits (tax, subsidy, and loans) and legal protection. Thanks to the primary survey, which collected direct data on whether each firm could access valuable and important information from the government, this paper accurately measures the effect of government information as a corporate political activity.

The following sections comprise the rest of the paper. Section 2 presents the theoretical framework and hypotheses development based on the literature review. Section 3 describes the data and the methodology. The main results are presented in Section 4, and Section 5 concludes the paper with managerial and policy implications.

2. Theoretical Background and Hypotheses Development

To investigate the factors influencing innovation in ESMEs, this study employs the extended resource-based view (ERBV) and non-market strategy to examine how firm-level resources, rather than individual attributes, influence innovation. Innovation performance refers to new or significantly improved products or processes [20]. While the extant literature has extensively explored individual characteristics of top managers, such as age, education, and experience, via the upper echelons theory [21], this research deliberately concentrates on firm-level data. By focusing on intangible resources such as human capital investment, international experience, and political connections, this study endeavors to provide a sophisticated and comprehensive analysis of how these firm-specific resources significantly impact innovation performance in emerging markets.

RBV theorists assume that the tangible or intangible resources accumulated by a firm determine its strategic competitiveness [22,23,24]. Unlike tangible resources, it is difficult to imitate or transfer intangible resources. Being rare and valuable, intangible resources become a firm-specific resource for innovation and financial performance [25,26,27]. In emerging markets where weak institutional regulations are prevalent, private firms rely more on close relationships with politicians and governments as intangible resources than legal and formal systems [28,29]. Despite the vital role of intangible resources, especially in the ESME context, there are few studies on the relationship between intangible resources and innovation in terms of ERBV [30]. ERBV emphasizes intangible social capital that can enhance a corporate entity’s capabilities to outperform its competitors in terms of value enhancement, along with close relationships with key stakeholders such as customers, suppliers, and communities [31]. Thus, small businesses need to maintain friendly and cooperative relationships to remain competitive; their interdependent relationships with local communities can be seen as important resources directly linked to their survival [32].

Hence, our focus is to examine which intangible resources of SMEs are critical for innovative activities. We choose human capital investment, international experience, and political connections as intangible, but valuable and rare resources for ESMEs. The reasons for our variables of interest are as follows. A firm’s human capital is a valuable resource for innovation because it cannot be imitated or replaced [33]. Additionally, firm-specific human capital with expertise and technical capabilities is difficult to transfer to other companies [34]. Similarly, international experience equips firms with the ability to learn and identify possibilities and develop strategies accordingly in a foreign market with higher risks [35]. Political connections, which are also a form or rare social capital, influence a firm’s innovation by means of technological subsidies, tax incentives, and deregulation provisions [36].

Additionally, we investigate how corporate political activities improve a firm’s ability to use their internal resources more effectively for innovative activities [13,14,15,16,17]. It is well established that expanding the firm’s engagement with key external information sources will enhance its innovation potential [37,38,39]. Based on the literature, we further explain why corporate political activities can interact with each of the three intangible resources and affect the likelihood of innovation.

2.1. Human Capital Investment

There is a broad consensus that human capital investment is related to productivity and innovation [40] and ultimately to actual innovation performance [41,42]. Especially, the augmentation of human capital knowledge is a resource factor that affects innovation [43]. Given training, employees better adapt to new technological changes and use knowledge and experience to solve problems in adopting new technologies [44,45]. Knowledge sharing can be spotted as a culture of social interaction in which employees’ knowledge, experiences, and skills are exchanged within organizations, leading to solutions and efficiencies that organizations can maximize [46]. It increases the know-how embedded in each employee as the experience-based knowledge of each worker is organized, reused, and transferred to co-workers [47,48].

From the perspective of ESMEs, the benefits of human capital investment can be even greater because of the lack of standardized know-how and expertise among employees [49,50]. Human capital investment creates opportunities to discuss ideas among employees via informal assemblies in developing countries [51,52]. In addition, it increases the chance of finding creative solutions, increases organizational stickiness, and improves the problem-solving capacity of employees. In sum, we hypothesize the following:

Hypothesis 1 (H1).

Human capital investment as a human resource has a positive effect on innovation performance by ESMEs.

2.2. International Experience

Much international business literature that underscores organizational learning as a prerequisite for innovation shows that firms learn and expand their knowledge base as they expand globally [53]. Such international experiences result in more skills, expertise, and innovation [54]. Exporting firms are more likely to gain technical expertise, such as designing and packaging new products, compared to non-exporters [55]. With multidimensional approaches to innovation by using local information [56,57,58], exporting firms are highly motivated to acquire new technologies to improve productivity because only productive companies can survive in the fiercely competitive overseas market [59,60].

For ESMEs, because it is relatively difficult to accumulate knowledge and skills in the home market, expanding their business arena overseas can be a channel to obtain new knowledge. The case of Chinese SMEs confirms that firms that had an opportunity to interact with international buyers and suppliers through export gained high-tech knowledge. As a result, international transactions increased the likelihood of innovation [61,62]. Thus, we hypothesize the following:

Hypothesis 2 (H2).

International experience as organizational learning has a positive effect on innovation performance by ESMEs.

2.3. Political Connections

Political connections are often perceived as ‘helping hands’ for companies. By adopting Faccio [63]’s definition of political connection and applying it to ESMEs, we assume that a political connection exists when at least one of a firm’s owners or shareholders has served in the government. Because having political experience among influential personnel is rare, valuable, and firm-specific, we assume that political connections are firms’ intangible resources embedded among decision-makers. For example, firms with political connections may have advantages such as tax cuts [63,64], bank loans [65,66,67], and party nominations [68]. From the perspective of the ERBV, it is emphasized that these potential effects are realized by the exclusive connection with external organizations, communities, and societies.

In emerging markets, the rule of law and formal institutions are weak. In addition, most of the market infrastructure, such as capital, labor, and product markets, remains underdeveloped [69]. In such a setting, because governments play a key role in allocating resources and economic opportunities [70], a strong tie to the government is even more important for any business activity. To illustrate, the legal system in emerging economies is so vulnerable that firms in developing countries face challenges in securing property rights and enforcing contracts fairly [71,72,73]. On the other hand, if firms have political connections, a favorable business environment can help them avoid or recover from potential losses caused by innovation failures [74]. Hence, we propose the following third hypothesis:

Hypothesis 3 (H3).

Political connection as social capital has a positive effect on innovation performance by ESMEs.

2.4. Government Information Access

Unlike developed countries, where firms can easily obtain necessary information from the public and private domain, a lack of information acts as a bottleneck for firm growth in developing countries [75]. Because of the highly centralized control of information, only firms that have informal ties as a corporate political activity with government officers can gain valuable information that other competitors cannot obtain by means of the official lines [76]. Firms with access to key government information can anticipate and proactively respond to prospective policies [77] and reduce uncertainties [78] by leveraging the wealth of their families or the economic and social background of their relatives [79]. Due to incomplete institutionalization, low marketization, and implicit cultures or traditions, ESMEs are more eager to form political connections and seek to receive more information via personal channels [80,81,82].

Firms with government information can receive favorable regulatory, operational, and financial support from the government, contributing to their stability against uncertain innovation pursuits [83,84]. Bai, et al. [85] verified that corporate access to government information effectively promotes corporate innovation and protects property rights.

How does critical government information affect SMEs that invest in human capital? Firms training their employees are more likely to have a capable human resource pool. Eventually, a firm with trained workers can easily seize and use knowledge from outside, so, when valuable information from the government is given, such firms would be able to take full advantage when conducting innovative activities. Moreover, while ESMEs are susceptible to the sudden leakage of innovative employees [86], government information can help SMEs proactively lock in some safeguards on the resources and capabilities required for innovative performance [87,88]. In sum, our hypothesis is the following:

Hypothesis 4a (H4a).

Government information access positively moderates the positive relationship between ESME’s human capital investment and innovation performance.

Although the relationship between international experience and innovation by emerging market firms has been studied [89], few studies are interested in reinforcing such association with access to government information. Among these few, Piperopoulos, et al. [90] found that emerging market firms that have been dependent on government information in the domestic market can leapfrog through internationalization because fresh products, technologies, and demand in overseas markets trigger innovative stimulants. Conversely, internationalized firms that have absorptive capacity because of their exposure to the competitive global market would be likely to pursue innovation with government information for two reasons: first, because internationalized firms better understand the importance of innovation for productivity, survival, and profit; and second, because they are better able to recognize and capitalize on external knowledge [91,92]. As a result, we establish the following hypothesis:

Hypothesis 4b (H4b).

Government information access positively moderates the positive relationship between ESMEs’ international experience and innovation performance.

Finally, we make a distinction between political connection and government information and examine how these two seemingly similar, but different political resources affect innovation together. As mentioned above, we assume that firms have political connections if their board members or CEO/owner has worked for the government. Naturally, firms with politically experienced leadership are more likely to access government information. However, it is possible that even firms with politically experienced personnel no longer have ties with the current government, so they cannot access government information. Similarly, firms without any political background can newly establish a social tie with government officials and can acquire valuable information. Therefore, there is a need to distinguish political connections as internal resources and government information as corporate political activities seeking external knowledge sources [93].

So far, few innovation studies have explored the role of government information as a corporate political activity. Instead, we can indirectly infer that firms with access to valuable government information would have tax reliefs, preferential treatment for government contracts, reduced regulatory oversight over competitors, or relaxed regulatory oversight on bank loans [94,95,96]. The combined effect of political connections and access to government information means more access to resources and opportunities [97] because those who have worked in the government can detect the value and relevance of government information and mobilize resources efficiently. To summarize, political connections further improve firms’ ability to innovate when additional external information is given because political connections can develop the ability to process and transmit government information. Thus, the following hypothesis is proposed:

Hypothesis 4c (H4c).

Government information access amplifies the positive relationship between ESME’s political connections and innovation performance.

Figure 1 comprehensively summarizes the logical backdrop discussed above.

3. Methodology

3.1. Research Context and Data Collection

This study is based on firm-level data collected from small and medium-sized enterprises in the Red River Delta. One unique feature of these firms is that those within the adjacent village produce similar products, forming an industry cluster. Unlike industrialized zones initiated by the government, these village clusters do not receive any given benefits from the government because of their locations. Some of these village clusters are historical, dating back a few hundred years, and are based on traditional manufacturing, whereas some are relatively new and developed. The firm size, level of technology, productivity, and quality of products also differ across firms within a village.

To identify these village clusters, we used the 2010 Vietnam Enterprise Survey (VES) data. The VES is conducted annually by the General Statistical Office (GSO) in Vietnam, covering all the formally registered firms under the Law on Enterprises, except for micro-firms. Using VES data, we selected 16 villages or communes, the smallest administrative units, with more than five clothing and textile industry enterprises (industrial codes 13 and 14 in Vietnam’s Industrial Classification System) in six provinces/cities (Hanoi, Hung Yen, Ha Nam, Thai Binh, Hai Duong, and Bac Ninh) in the Red River Delta region. Since there is a time lag between the VES data of 2010 and the actual survey, which was conducted from December 2014 to January 2015 for the first round and from July to August for the second round, we acquired an updated list of all enterprises registered in each of the 16 villages from the commune’s administration. The number of companies registered in each village ranged from 1 to 74, with the total number of companies standing at 354. We additionally gathered our third-round data in 2017 and ended up with 192 firms. Most firms in our sample are privately owned limited liability companies or household establishments, with capital sourced primarily from personal savings or family investments.

3.2. Analytical Method

Partial least-square structural equation modeling (PLS-SEM) estimation is an outstanding analytical technique for proving the relationships between constructs by forming constructs that target observed variables by means of the multivariate technique [98]. Since PLS-SEM discovers components optimized by factoring and maximizes variance explained according to the degree to which the construct affects the endogenous variable, the normality assumption is not necessarily required for multivariate analysis between latent variables and has a little freedom regarding the sample size. We used the estimation by means of the ‘plesem’ command provided by STATA 16 [99]. We tested fitness using confirmatory factor analysis (CFA) in the measurement model and examined hypotheses between constructs in the structural model. CFA evaluates the model’s reliability, convergent validity, and discriminant validity. Then, we re-verified the structural model using bootstrapping with a re-sample.

3.3. Variables

Our dependent variable is the innovation performance of ESMEs. We adopt the definitions of OECD for product and process innovation. Product innovation refers to “a good or service that is new or significantly improved… in technical specifications, components and materials, software in the product, user friendliness or other functional characteristics”. Process innovation is defined as “a new or significantly improved production or delivery method… in techniques, equipment and/or software” [100]. Variables are measured as dummies following Ballot, et al.’s [101] approach.

The explanatory variables in this study are three intangible resources: human capital investment, international experience, and political connections. First, based on the employees’ human capital investment suggested by Rauch, et al. [102], we inquired whether a company conducted or was planning to conduct training to enhance its employees’ skills. Second, we asked if a company had exported directly in the last two years to measure its international experience, following Oura, et al. [103]. Third, following Zhou [97]’s scale, we collected the number of board members that have had experience working as public officials and information on whether the owner or CEO had experience working for the government. Lastly, the moderating variable of this study is valuable government information as an external knowledge source. This moderating dummy variable is comparable to Krammer and Jimenez’s [93] scale, which measures political connection by the time spent by managers with government officials.

3.4. Measurement Assessment

Table 1. Descriptive statistics and confirmatory factor loadings (N = 192).

Construct	Code	Scale Item	SFL
IP	Ballot, Fakhfakh, Galia and Salter [101]
	IP1	Our company conducted product innovations in 2014.	0.74 *
	IP2	Our company conducted process innovations in 2014.	0.72 *
	IP3	Our company conducted product innovations in 2015.	0.77 *
	IP4	Our company conducted process innovations in 2015.	0.75 *
HCI	Rauch, Frese and Utsch [102]
	HCI1	Our company conducted training to improve employees’ skills.	0.93 *
	HCI2	Our company conducted training to improve workers’ knowledge.	0.93 *
IE	Oura, Zilber and Lopes [103]
	IE1	Our company experienced direct export in 2014.	0.91 *
	IE2	Our company experienced direct export in 2015.	0.97 *
PC	Zhou [97]
	PC1	Board members in our company are either current or former public officials.	0.90 *
	PC2	The top decision-makers in our company served in the government.	0.76 *
GIA	Krammer and Jimenez [93], Zhou [97]
	GIA	Our company obtains valuable and important information from the government.	N/A

Notes: IP = innovation performance, HCI = human capital investment, IE = international experience, PC = political connection, GIA = government information access, SFL = standardized factor loading, * p < 0.05.

shows the results verifying whether the observed variables used for the constructs are suitable as confirmatory factors. By excluding items with standardized factor loading thresholds lower than 0.7 [98], we adopted p < 0.05 as the significance level. In alignment with the findings presented in Table 1,

Table 2. Results of cross-loadings.

Loadings	IP	HCI	IE	PC
IP1	0.7410	0.1906	0.0792	0.0551
IP2	0.7238	0.2055	0.0802	0.1489
IP3	0.7645	0.2244	0.0639	0.0703
IP4	0.7462	0.1439	−0.0118	0.1200
HCI1	0.2509	0.9264	−0.0491	0.0264
HCI2	0.2482	0.9247	0.0030	0.0466
IE1	0.0506	−0.0286	0.9074	−0.0517
IE2	0.0899	−0.0211	0.9716	−0.0822
PC	0.1343	0.0526	−0.0704	0.9020
PC	0.0886	0.0039	−0.0527	0.7560

Note: Bold values are loadings for items, which are above the threshold value of 0.7.

identified the valid cross-loadings.

Table 3. Inter-construct correlations, convergent, and discriminant validity.

Construct	1	2	3	4
1. Innovation performance	1
2. Human capital investment	0.27 *	1
3. International experience	0.08 *	0.03 *	1
4. Political connection	0.14 *	0.04 *	0.07 *	1
Cronbach’s alpha	0.70	0.83	0.88	0.75
Composite reliability	0.81	0.92	0.94	0.82
rho_A	0.70	0.83	0.95	0.73
AVE	0.52	0.86	0.89	0.69
SQRT(AVE)	0.72	0.93	0.94	0.83
HTMT (<0.85)	Yes	Yes	Yes	Yes

Notes: roh_A = Dijkstra and Henseler’s composite reliability, AVE = average variance extracted, SQRT = square rooted, * p < 0.05.

presents Cronbach’s alpha, composite reliability, rho_A (Dijkstra and Henseler’s composite reliability), AVE (average variance extracted), square root values of AVE, HTMT (heterotrait-monotrait ratio of correlations), and correlations between the constructs. Following Hair, Black, Babin and Anderson [98], Cronbach’s alpha, composite reliability, and rho_A were all higher than 0.7, confirming internal consistency and convergent validity. We estimated AVEs of all latent variables to be higher than the recommended level of 0.5, while the square roots of AVEs were higher than the correlation between latent variables. Calculating HTMT using Smart PLS 4, and alternative discriminant validity as proposed by Henseler, et al. [104], we found that all latent variables used in this study were lower than 0.85, indicating that there is discriminant validity [98,105,106].

3.5. Common Method Bias (CMB) Test

We tested CMB in the following three ways. First, our constructs showed an explanatory power of 24.84%, which is less than the threshold suggested by Harman’s single-factor test using principal component analysis [107,108]. Second, based on the maximum variance inflation factor (VIF) of 3.3 recommended by Kock [109], our maximum VIF was 1.007 [110]. Third, when the marker variable technique (MVT) suggested by Lindell and Whitney [111] is introduced, the existing construct should have a low correlation with the marker variable (M1), an instrumental variable, and should be insignificant. We constructed the import experience of ESMEs as M1 and confirmed that the correlation was less than 0.3 and insignificant.

Table 4. Sample demographic.

Variables [Chi2, p-Value]	Respondents		N	%
	Early	Late
Village [11.28, 0.59]
Van Phuc	4	6	10	5.21
Co Nhue	5	3	8	4.17
Tuong Giang	6	4	10	5.21
Phung Xa	27	20	47	24.48
Tien Phong	24	7	31	16.15
Van Tu	7	7	14	7.29
Phuong La	19	10	29	15.10
Hoa Hau	8	6	14	7.29
Duong Xa	1	1	2	1.04
Tan Trieu	5	4	9	4.69
Duong Noi	2	3	5	2.60
ai Quoc	2	0	2	1.04
Thien Phien	1	0	1	0.52
Thanh Ha	8	2	10	5.21
New product development [0.02, 0.89]
No	64	40	104	54.17
Yes	55	33	88	45.83
Number of employees [2.31, 0.51]
≤10	37	24	61	31.77
11–20	20	16	36	18.75
21–50	38	24	62	32.29
>50	24	9	33	17.19
Sales (unit: million VND) [3.79, 0.29]
≤5000	33	27	60	31.25
5001–10,000	16	13	29	15.10
10,001–50,000	41	17	58	30.21
>50,000	29	16	45	23.44
Formal bank account [0.92, 0.34]
No	9	3	12	6.25
Yes	110	70	180	93.75
Total	119	73	192	100

Note: no significance between early and late respondents using Chi² analysis and t-tests.

shows the demographic statistics for Vietnamese SMEs and the verification results for no-response bias. Of the total sample of 192, about 40% were concentrated in Phung Xa (N = 27) and Tien Phong (N = 24). Since the samples were all SMEs, more than half the sample had no more than 50 employees (N = 159). The sales of the selected Vietnamese SMEs were evenly balanced. By testing the difference between early and late responses for demographic items by Chi² test using the samples that underwent third-round, we found that all of them were statistically indifferent, showing that our sample was free from the problem of non-response bias [112].

4. Empirical Results

4.1. Structural Assessment

Figure 2 exhibits the standardized path coefficient and statistical significance of each construct. The explanatory power (R²) of this model is about 18%, and the predictive relevance (Q²) is approximately 16%. Control variables were added to reduce the variance between the independent and dependent variables: the firm’s age (β = 0.05, p > 0.05), number of employees (β = −0.08, p > 0.05), formal bank account (β = −0.05, p > 0.05), and professional membership (β = 0.15, p < 0.05). The result supports H1, as we observe a positive effect of human capital investment on innovation performance (β = 0.27, p < 0.001). On the other hand, the coefficient of international experience is insignificant, rejecting H2 (β = 0.01, p > 0.05). When ESMEs such as those in Vietnam have political connections, H3, an expectation of positive effects on innovation performance, is verified as significant (β = 0.13, p < 0.05). Only international experience (H2) does not influence the innovation performance by ESMEs among the three internal intangible resources.

The evaluation of the moderating effect of access to valuable government information is as follows. First, H4a is rejected because the effect on innovation performance of the interaction between SMEs investing in human capital and access to government information was insignificant (β = 0.01, p > 0.05). Next, when ESMEs with international experience have access to valuable government information, the effect of this interaction on innovation performance is positive and significant, indicating that H4b is accepted (β = 0.37, p < 0.05). Lastly, H4c is also accepted because ESMEs with political relations are better at innovation performance because of external information from the government (β = 0.46, p < 0.05). Although ESMEs should construct more efficient resources than their competitors, this can be interpreted as the need to access vital government information through resource expansion using surrounding network resources.

4.2. Robustness Tests

To validate the robustness of our findings, we investigated the differences in path coefficients based on the sample’s unobserved heterogeneity level [113,114]. We selected two groups of SMEs based on the highest value of the consistent Akaike information criterion (CAIC₂ = 575.531), a normalized entropy statistic (EN = 0.61), and normalized the entropy criterion (NEC = 0.58). We employed multigroup analysis to assess the compositional invariance and confirmed no difference between segments [115]. In addition, to dispel concerns about the endogeneity of PLS-SEM, we adopted Hult, et al.’s [116] 2SLS method. We verified the overidentifying restriction using the equality of board directors as an instrumental variable (IV); there was no difference from the existing model, indicating that our IV constraint was appropriate [116].

We employed OLS regression analysis for additional robustness verification in

Table 5. Results of OLS regression.

Variables	Model 1	Model 2	Model 3	Model 4	Model 5
Firm’s age	0.00	0.00	0.00	0.00	−0.00
	(0.01)	(0.01)	(0.01)	(0.01)	(0.01)
Formal bank account	−0.13	−0.10	−0.08	−0.08	−0.02
	(0.28)	(0.29)	(0.28)	(0.28)	(0.28)
Number of employees (reference: ≤10)
11–20	0.24	0.24	0.24	0.17	0.18
	(0.20)	(0.21)	(0.20)	(0.21)	(0.20)
21–50	−0.34	−0.34	−0.35 *	−0.32	−0.35 *
	(0.17)	(0.18)	(0.17)	(0.17)	(0.17)
>50	0.09	0.07	0.15	0.08	0.17
	(0.24)	(0.24)	(0.24)	(0.24)	(0.24)
Professional memberships	0.26 *	0.26 *	0.26 *	0.26 *	0.27 **
	(0.10)	(0.11)	(0.10)	(0.10)	(0.10)
ET	0.25 ***	0.28 *	0.26 ***	0.24 ***	0.32 **
	(0.07)	(0.11)	(0.07)	(0.07)	(0.11)
IE	0.08	0.07	−0.11	0.07	−0.15
	(0.08)	(0.08)	(0.13)	(0.08)	(0.13)
CON	0.10 *	0.09 *	0.10 *	0.22 *	0.24 *
	(0.07)	(0.07)	(0.07)	(0.17)	(0.17)
HCI $\times$ GIA		0.07			0.09
		(0.15)			(0.15)
IE $\times$ GIA			0.27 *		0.31 *
			(0.14)		(0.14)
PC $\times$ GIA				0.38 *	0.42 *
				(0.19)	(0.19)
Constant	0.01	0.02	−0.02	−0.03	−0.08
	(0.32)	(0.32)	(0.32)	(0.32)	(0.32)
N	192	192	192	192	192
R2	0.17	0.16	0.18	0.18	0.20
adj. R2	0.12	0.11	0.13	0.13	0.15
p-value	0.00	0.00	0.00	0.00	0.00

Notes: standard errors are in parentheses; HCI = human capital investment, IE = international experience, PC = political connection, GIA = government information access; * p < 0.05, ** p < 0.01, *** p < 0.001.

. The firm’s age, formal bank account, number of employees, and number of professional memberships were included as control variables that could affect the innovation performance of ESMEs. Latent variables were applied to the independent and dependent variables used in the structural equation model. Model 1 is a baseline in which only control variables and independent variables are inserted. Model 1 is a baseline in which only the control variables and independent variables are included, and the significant and insignificant hypotheses are consistent with the structural equation results (${\beta }_{ET}$ = 0.25, p < 0.001; ${\beta }_{PC}$ = 0.10, p < 0.05; ${\beta }_{IE}$ = 0.08, p > 0.05). In Models 2, 3, and 4, the moderating effect was tested by including an interaction term between government information access and each independent variable; these were also almost identical to the result of the structural equation (${\beta }_{ET\times GIA}$ = 0.17, p > 0.05; ${\beta }_{PC\times GIA}$ = 0.38, p < 0.05; ${\beta }_{IE\times GIA}$ = 0.27, p < 0.05). In Model 5, a full model in which all variables are included, the significance of the details is mostly consistent.

4.3. Additional Model

Another objective of this study was to test the hypothesis that access to national information (GIA) may play a mediating role, compared to a moderating role, in strengthening existing and expanding resources. If a firm has political connections at the governance level, it may naturally be incentivized to seek access to important state information. To address these research objectives, we further analyzed the mediating effect to ensure robustness. In Figure 3 and

Table 6. Significance testing of mediation with bootstrap.

Statistics	HCI→GIA→IP	IE→GIA→IP	PC→GIA→IP
Indirect effect	0.02	0.01	0.02
Standard error	0.01	0.01	0.01
z-statistic	1.27	0.55	1.62
p-value	0.21	0.58	0.10
BCCI	(−0.01, 0.04)	(−0.01, 0.05)	(−0.00, 0.05)

Notes: (1) 5000 iterations for bootstrapping, (2) confidence level of 95%, (3) BCCI = bias-corrected confidence interval, (4) HCI = human capital investment, IE = international experience, PC = political connection, GIA = government information access.

, no mediating effect of GIA was confirmed for HCI and PC, and it was not significant for IE. However, the partial significance of each variable suggests that there might be a mediation relation between variables if future research could be extended to a broader range of firm-level studies.

5. Discussion and Conclusions

To better understand the dynamics of rapid industrialization and economic growth in emerging and transitioning economies such as Vietnam, it is crucial to investigate the factors of SMEs’ innovation performance. This topic is particularly important because, while SMEs are vital to fostering innovation and economic development, they also face many obstacles in an emerging market context. Among the many barriers to innovation for ESMEs and transition economies, we focused on the role of intangible resources and the moderating role of key government information through political activity. Lack of open information and SMEs’ inability to acquire, understand, process, integrate, and employ information and turn it into knowledge for innovation persistently pose a challenge to ESMEs. For this reason, we aimed to reiterate the importance of intangible resources from the RBV and ERBV perspectives.

Our analysis provides insight into the factors that influence SMEs’ innovation. First, based on the intangible resource-based view, companies that invest in human capital, especially through employee training, are significantly more likely to innovate. However, as for the moderating effect, this positive association is not affected even if SMEs have access to critical government information. Second, there is no correlation between international trade and innovation performance. However, when internationalized SMEs have access to government information, the likelihood of innovation increases. This suggests that simply participating in international markets is not sufficient to promote innovation, but additional corporate activities are necessary. Third, political connections within SMEs are positively correlated to innovation, and the positive effect of political ties are further enhanced by SMEs’ access to information from the government. In other words, SMEs’ ability to leverage external information resources is critical to maximizing the innovation potential of these intangible assets. This study provides useful insights for policymakers and business leaders seeking to promote innovation in emerging markets.

5.1. Theoretical Implications

Our results indicate noteworthy insights into innovation performance in ESMEs by incorporating RBV, ERBV, and corporate political activity. First, our findings increase our understanding of what is vital as an intangible resource for ESMEs from an RBV perspective. Of the three intangible resources selected in this research, human capital investment in the form of employee training has the largest effect on innovation, followed by political connections. Nonetheless, firms with export experience as organizational learning do not exhibit better innovation performance than non-exporters. One potential explanation for this rejection of the hypothesis is that the exporters included in our sample are not necessarily highly productive and competitive. According to anecdotal cases from interviews with exporters, many of these exporting firms do not actively seek to export because they do not know much about the trade process. Rather, international buyers visit the villages and seek cheap products. Contrary to to our hypothesis, Oura, Zilber and Lopes [103] found that the innovation capacity of Brazilian SMEs is conducive to export performance. To reinterpret this, it is difficult for ESMEs to experience internationalization without significant luck or external support; conversely, sufficient innovation capability means that export is possible because ESMEs can compete with foreign firms in global markets. The fact that SMEs are competitive enough to produce innovative products in emerging markets can be interpreted as having a higher level of R&D investment, marketing, manufacturing, and resource allocation factors than their competitors [56,88,102]. Summarizing the theoretical implications of our study from the RBV point of view, we found that it may be desirable for ESMEs to create innovative performance by investing in continuous innovation capability enhancement for internal employees rather than international experience. This implication is in close accord with El Shoubaki, Laguir and Den Besten’s [49] verification, which emphasized that SMEs need “general human capital” with comprehensive capabilities and “specific human capital” specialized for particular businesses or functions. In a hyper-competitive society, where the domestic market of emerging economies is gradually being eroded by globalization, EMSEs may need to prepare detailed strategies for human capital and innovation [34,49,102,117].

Second, our findings theoretically extend the understanding of the ERBV perspective and corporate political activity regarding ESMEs’ innovation in the IB literature. The interaction terms of intangible resources and access to government information differ in their significance and magnitude. Firms led by those who worked in the government benefit the most from access to valuable government information. Firms with international experience are also more likely to innovate when government information is given. The reason for such a significant moderating effect of government information is that we asked whether firms could access “valuable and important” information [12]. Without the absorptive capacity formed by intangible resources, firms cannot decide whether a piece of information is valuable and important. The effect of government information is not meaningful for innovation in firms that have invested in human capital. In other words, the usefulness of external knowledge depends on the types of preexisting intangible resources. Hence, to increase the likelihood of innovation, firms’ absorptive capacity must be prioritized, especially for resource-deprived and less-productive ESMEs. Sharma, et al. [118] explored how political connections affect the exports and the export performance of Chinese companies. Given that the country to which SMEs belong is still in the early stages of development and that corporate governance is greatly affected by the state, our findings are consistent with Sharma, Cheng and Leung [118] in that ESMEs are required to preemptively assess to government information through political connection to generate innovation [87]. That is, while ESMEs should accumulate resources through internal investment to prevent their competitors from mimicking innovation, from an ERBV perspective, assembling government bureaucrats or similar political connections would also be an effective strategy [31,32,119].

5.2. Practical Implications

We provide policy and managerial implications for those seeking to innovate SMEs in emerging markets. For policymakers, the findings of this paper suggest that it is crucial to provide important government information fairly and transparently, especially for information-deprived SMEs. Since SMEs struggle to find quality information for their decision-making, the government should supply information via public information platforms. Such avenues for sharing information can help SMEs overcome information shortages and take full advantage of their innovative potential based on their internal and external resources and capacities. In addition, the government can organize or subsidize training programs to help SMEs develop their human capital. SMEs often cannot achieve economies of scale in human capital investment due to their small size. The government can then aggregate workers from similar industries and geographic locations and offer training. By doing so, governments can effectively foster innovation.

For firms, our results can guide them in determining which mix of intangible resources and social capital can be most relevant to innovation for each firm. Since both accruing intangible resources and corporate political activities incur costs and thus compete for resources, firms must set a priority. For example, Efobi, et al. [120] proved that workers bear the cost of bribery in Vietnam because firms cut down on wages to compensate for the bribe payments. Although it is beyond the scope of this paper to consider costs, pursuing all types of knowledge-accumulating activities, such as human capital investment, internationalization, and establishing political ties, may be unaffordable for ESMEs because of high costs. Hence, ESMEs should consider the most cost-effective combinations to achieve innovation. These findings call for managers and owners of ESMEs to accurately identify their endowed internal and extended resources. Once they find their resources’ strengths and weaknesses, firms can make more efficient decisions on which resources to acquire more for innovation. For example, when firms do not have international experience or political connections, instead of investing in creating liaisons with politicians to obtain information, developing knowledge within the firm by means of employee training may increase the likelihood of innovation much more.

5.3. Limitations and Suggestions for Further Research

It must be noted that this research has several limitations in terms of its measurement of variables. First, we relied on self-reporting for all the variables, including innovation. Self-reporting may result in mismeasurement [121]. Second, we do not distinguish the quality of innovation, whether it is radical or incremental. Although most firms in the sample are low-tech and small, the site visits and interviews revealed observable differences in terms of productivity and level of innovation across the firms. Therefore, further studies can refine their measures of innovation to capture the quality of innovation, measured objectively by, for example, productivity or the adoption of relevant technologies. Additionally, we can apply a multi-dimensional approach to measure innovation, such as frequency, embeddedness among workers, or contribution to the sales. Third, the measure of corporate political activity can be further refined. Since it could be a sensitive question, we limited ourselves from asking too many details about political connections. Still, to deepen our understanding about the effect of corporate political activities, more data collection on the structure, frequency, and nature of the connections is necessary [122]. Our sample also consists of micro-, small-, and medium-sized firms owned mainly by families located in Northern Vietnam. The external validity of our paper’s findings should be tested in different political and business contexts, as Vietnam has a unique political setting, being a single-party state. Expanding this study scope to include high-tech industries such as IT and machinery, as well as oligopolistic sectors in which large state-owned enterprises and foreign firms operate, is also suggested.