Political Connectedness and Financial Performance of SMEs

Abstract

The extant literature on the association of political connectedness and performance of large firms has led to controversial results, while the context of micro-, small- and medium-sized enterprises (SMEs) has largely been overlooked in relevant studies. To resolve these gaps, the objective of this paper is to study the link between the political connections of firm board members and financial performance in the Estonian SME population. Using a wide selection of financial performance and political connectedness variables, the composed regressions indicated that firms with politically connected boards underperform their unconnected counterparts. The findings remained robust not only through different measures of dependent and independent variables, but also periods studied.

1. Introduction

Corporate political activeness is a frequent non-market strategy to gain influence over political actors to shape government policies for favoring corporate interests. Corporate political activeness encompasses a wide array of activities such as lobbying, contributing to the political campaigns, and direct political cooperation with political actors (Hillman et al. 2004; Hadani 2012), while one of the most distinct forms of corporate political activeness is political connectedness. Political connectedness is mainly revealed by a (former) politician being directly involved in a firm’s corporate governance (Faccio 2010). Political connectedness may grant many advantages to firms over other non-connected firms by providing preferential treatment by government-owned institutions, smoother access to debt financing, tax relief, winning government procurements, or lower regulatory requirements and monitoring (Faccio 2006; Niessen and Ruenzi 2010).

Prior research has often associated political activeness to large companies or multinational enterprises (e.g., see Sun et al. 2021 for a literature review), as these entities have resources and capabilities to direct developments in the business environment to achieve success (Luiz and Stewart 2014). Therefore, large enterprises seem to have dominance in this research domain due to their power relations with politicians, influence, and role in the economy. However, past studies have largely overlooked the fact that in a multi-layered economy, SMEs (micro-, small- and medium-sized enterprises) would also be capable of and have incentives to adopt non-market strategies towards building political and social ties (Jackowicz and Kozłowski 2019). Compared to large firms, SMEs would have peculiar interests in establishing and maintaining political links due to very SME-specific constraints, such as high dependence on personal relationships and informality, being often owner-managed and independent in decision-making, having strong local business roots (Spence 1999), and relying on internal financing for growth (Lepoutre and Heene 2006; Russo and Perrini 2010). This reliance on internal financing resulting from limited access to finance is considered to be the biggest obstacle to their operations (Beck and Demirguc-Kunt 2006). In relation to the latter, banks and other financial institutions may be reluctant to provide loans due to high default risk, which is often associated with firm age and size, the relative cost of appraisals and monitoring activities, and the burden of liquidizing the assets of a firm in case of default (Arráiz et al. 2014; Kersten et al. 2017).

When these SME specific factors are aligned together, political connectedness based on interpersonal relationships and informality may be a very desirable non-market strategy for firms that operate largely in a local business environment, for example for accessing debt financing or seeking preferential treatment from local governments. Furthermore, SMEs are considered to be an important employment-generating and productive segment of the global economy, which offers extra motivation to study their respective associations. According to OECD (2021), SMEs are the backbone of the Estonian economy, which provides the context for this paper: SMEs constitute 78% of employment and 76% of value added in the country. Hence, SMEs continue to be a productive and dominant segment in Estonia compared to the OECD average for the latter variables, having respective values of 69% and 59% (OECD 2021). Therefore, this paper will focus on the underexplored SME segment by investigating how various financial performance indicators of a firm are associated with corporate board’s political connectedness.

This paper adds several novelties to the corporate political connectedness literature by diverting the focus on SMEs and their business performance rather than focusing on the relatively exhausted context of large and listed firms. Previously found positive effects of political connectedness on firm performance were mainly supported by explanations from the resource-based view. These positive associations could be contested by the explanation from the corporate governance literature, which explains how busy and overworking directors might deteriorate firm performance (e.g., Gray and Nowland 2018).

In addition, even though the corporate political connectedness literature has been continuously getting richer with new contributions, the contradictory effects of political connections on firm performance still require further attention in different country and sample contexts. While one stream of studies has shown positive effects of political connectedness on financial performance (e.g., Greiner and Lee 2020; Wu et al. 2018), another stream of studies has emphasized that political connectedness has adverse consequences on it (e.g., Okhmatovskiy 2010), especially over the long-term in emerging economies (e.g., Sun et al. 2010). Given these two competing perspectives in the literature about the consequences of political connectedness on financial performance and the lack of research in the SME context from the perspective of former socialist countries, this paper aimed to explore the link between political connectedness and firm financial performance on the whole population of firms in Estonia.

The connection between these two phenomena could be dependent on how firm performance (Hersch et al. 2008; Hadani et al. 2017) and political connectedness (Wang and Yu 2022) are specified. Thus, to avoid a bias from the use of a single measure, we captured firm financial performance with a wide array of known measures from the relevant literature. Political connectedness, unlike in papers on large and listed firms, was customized in this study for the SME context by considering passive and active connections, operationalized in terms of party membership and party members running in elections. Thus, by relying on a rich set of measures, this study aims to provide factual population-level evidence on the relationship between financial performance and political connectedness in the understudied SME segment. This approach also enables to determine which of the extant schools prevails in their assessments of the association between these two phenomena in the SME segment.

The rest of the paper is structured as follows. The next section clarifies the theoretical background for the two competing hypotheses that guided the research. This is followed by an explanation of the chosen study design, including the Estonian context and the dataset, variables, and methods used. Then, the main results are presented (Appendix A), after which, the Discussion section outlines the scientific implications of the paper. The paper ends with a standard conclusion that includes main limitations, practical implications, and multiple ideas for future studies.

2. Literature Review and Hypotheses

2.1. SMEs and Political Connectedness

Scholars argue that firms intensify their non-market strategies in the form of corporate political activeness (e.g., lobbying, contributions to political campaigns) as a complementary or alternative strategy to market strategies (Bonardi 2008; Parnell 2018) for the purpose of gaining access to the public policy process (Hillman et al. 1999) in fiercely competitive business environments (Baron 1995).

However, large and multinational enterprises have dominated the literature so far, and surprisingly, less attention has been paid to SMEs. Compared to large and multinational enterprises, studying political connectedness from the SME perspective is motivated by several SME-specific features and theories. First, SMEs deserve special attention for studying the interconnection of political connectedness and firm performance, as SMEs are considered to be the engine of economic growth (Gherghina et al. 2020) and employ a significantly large proportion of the European and Estonian productive workforce (OECD 2021). SMEs’ further growth aspirations might be influenced by political connections, as these connections may alleviate SME constraints.

Second, various peculiarities cause SMEs to diverge from their larger counterparts in seeking and maintaining political connections. Among them, the rule of a single individual through different corporate governance layers, being reliant on the personal network of that owner-manager and local business infrastructure, can be among the most important drivers of establishing political connections (Spence 1999).

Another motivation emerges from SME-specific constraints, which motivate them to build political connections. SMEs are reliant on internal and external financing options to maintain the long-term financing they need for realizing high growth aspirations (Dowling et al. 2019). Financial institutions may be reluctant to provide loans to SMEs due to risks and uncertainties that may be rooted in the creditworthiness of SMEs and lack of full information disclosure (Jackowicz and Kozłowski 2019). Because of the latter, SMEs might make use of political connections to seek preferential treatment from financial institutions to overshadow the risks and uncertainties related to them. For instance, Fu et al. (2017) provided evidence that politically connected Indonesian SMEs have a higher likelihood of getting loan approvals and receiving larger loans from state-owned banks through their personal connections to politicians.

Consequently, in the SME context, political connectedness is primarily used as a mechanism to overcome the inherited weaknesses of these firms and pursue higher growth goals. However, large firms would prefer to establish political connections not due to SME-specific weaknesses, but in an effort to turn various external factors in their favor in the market, e.g., to increase their competitive position. Therefore, focusing on the largely overlooked SME context would widen our knowledge regarding the consequences of political connectedness on firm performance.

2.2. Development of Hypotheses

Having established the relevance of political connectedness for SMEs, we have further investigated the possible consequences of political connectedness on SME financial performance. Past studies have indicated that political connections and firm performance are linked, but theoretical discussion and empirical evidence on whether the effect is positive or negative is ongoing. This section outlines the main arguments behind these two competing views. Based on the relevant literature, we develop two testable hypotheses to examine signs of links between political connections and firm performance.

One stream of research lends support to the possible positive effect of political connections on firm performance by using the resource-based view (RBV). According to RBV, firms might seek resources from the external environment by using the firm’s top official’s connections to capture opportunities and enable them to maintain competitive advantage (Lu et al. 2010). In the SME context, the latter explanation should be modified in a way that stresses the primary motivations of SMEs. Namely, firms would instead take advantage of political connections to increase the probability of finding and receiving funding, winning procurements in local municipalities, and eventually boosting firm growth.

Numerous empirical studies have revealed that there is a positive impact of political connections on firm performance, lending support to the position that politically connected firms gain more benefits than non-connected firms (e.g., Wu et al. 2018; Ferguson and Voth 2008). Consequently, political connections may create favorable conditions for SMEs, and thus, they may enable SMEs to achieve higher growth and financial performance than non-connected firms. Drawing on this position, we propose the following hypothesis:

H1. 

Political connectedness is positively associated with SMEs’ financial performance.

In contrast, another stream of research holds that political connections might have adverse consequences on firm performance. Specifically, previous explanations have focused on the extorting nature of the relationship and the unforeseeable costs of political connectedness. For example, political embeddedness may constrain the future strategic options of a firm (Okhmatovskiy 2010), escalate costs for firms, and damage firm productivity in the long-term (Domadenik et al. 2016). Politicians could extract rents from firms, and firm officials may spend valuable time to maintain their relationships instead of firm-related activities. Thus, several empirical examples exist in case of which politically connected firms demonstrated poorer financial performance than non-connected firms (e.g., Khwaja and Mian 2005; Fan et al. 2007; Boubakri et al. 2008; Wong and Hooy 2018).

In the SME context, one theoretical explanation originates from the corporate governance literature, in which the adverse effect of the director’s greater busyness and workload on firm performance has been theoretically explained and empirically proven (e.g., López Iturriaga and Morrós Rodríguez 2014; Liu and Paul 2015). The implication from the busyness literature for this study is that directors who are involved in politics must distribute their workload between different activities, and this, ceteris paribus, reduces the quality of management. From a more theoretical perspective, such directors might lack attention (see Occasio 1997). Based on this position, we set the following hypothesis:

H2. 

Political connectedness is negatively associated with SMEs’ financial performance.

The rejection of both of these hypotheses would mean that political connectedness is not associated with financial performance, which could be caused by two different phenomena. In the first case, the positive and negative consequences may both be present, but their equal magnitude results in one negating the other (Brockman et al. 2013; Tihanyi et al. 2019). In the second case, which is likely a more applicable explanation for SMEs in practice, actually no consequences exist at all.

Several scholars have emphasized the importance of institutions in the business and politics nexus. For example, firms may develop corporate political strategies based on dependency relationships between firms and politicians due to institutional weaknesses (Liedong et al. 2020). Hence, the positive or negative consequences of political connections on firm performance could be dependent on the country’s institutional quality (Brockman et al. 2013). In countries with underdeveloped institutions such as lack of rule of law and democracy, low accountability and transparency, high corruption, firms may prefer to establish and maintain political connections for the purpose of reducing uncertainties in the business environment and enhancing firm productivity. In contrast, countries with developed institutions may create comfortable business environments where firms follow market strategies and may not be in need of political connections to improve firm performance. In connection with the latter, Estonia holds high positions in several international rankings for institutional quality, having received, e.g., an 89.25 percentile rank out of 100 in one study (Kaufmann et al. 2011), and for business environment, having received 18th place out of 190 in the Ease of Doing Business Index (year 2019) by the World Bank (2022). Thus, the results of this study should not be impacted by poor institutional quality, which might not be the case for many other former socialist countries that hold much lower positions.

3. Study Design

3.1. Data and Method

This study incorporated different datasets from several sources. General, financial, and board information of firms originated from the Bureau van Dijk Amadeus database and the Estonian Business Register (EBR), while tax arrears information was sourced from the Estonian Tax and Customs Board. The lists of party members were downloaded from EBR, as parties are non-profit legal entities. Party member lists were in turn merged with the lists of candidates in different elections, publicized by the State Electoral Office. Both the firm- and individual-level data cover the whole population, and thus, are free from sampling biases.

As we intended to find out whether politically connected firms perform better, worse, or no differently than their non-connected counterparts, we applied linear regression (OLS) with the following general format (FFP standing for firm financial performance and PC for political connectedness):

$$FFP=\alpha +{\beta }_{1}PC+{{\displaystyle \sum }}_{i=2}^N{\beta }_{i}Controls+\epsilon$$

(1)

As the focus of the study was to disentangle the controversy whether politically connected firms perform better than their counterparts or not, the chosen method exactly served the purpose. It should be emphasized that the study was not focused on causal relationships, as firm performance could be influenced by various different factors, of which internal ones, especially those focused on employees, have been reported to be more prominent than external ones (Hansen and Wernerfelt 1989).

Following the previous literature on firm financial performance (e.g., Altman et al. 2017; Süsi and Lukason 2019), we included several control variables in the estimations. The control variables encompass firm age (in years), size (the natural logarithm of firm’s total assets), firm location (1 for the capital city of Tallinn, 0 for other), mean of the biological age of individuals on the board (in years), women on the board (1 for at least one woman on the board), and mean of the time spent on the board (in years). These controls have been frequently used in the empirical studies on the relationship between political connectedness and firm performance, with recent examples of board member age in Liang et al. (2022) and Xia et al. (2019), board member gender and firm region in Wu et al. (2018), board tenure in Kuzman et al. (2018), and firm age and size in He et al. (2019) and Liu et al. (2021). We also included industry dummy variables to control for industry fixed effects (referred to as “Industry FE”).

To exclude the possibility of yearly biases, in addition to running regressions with financial performance variables calculated from the base year of 2019, we conducted analysis with the same dependent variables for two previous years (2017 and 2018) as well, keeping the independent political connectedness variables the same as for the base year. Such an econometric approach is more feasible than the use of yearly dummies, as it enables the direct comparison of the coefficient signs and p-values of the same independent and control variables across different years. The selected three years are characterized by stable economic growth in Estonia without substantial external developments (e.g., the COVID-19 virus outbreak, the war in Ukraine, the availability and price fluctuations of raw materials), which can substantially impact firm performance. Therefore, in the selected period of study, the average company was not likely to have witnessed abnormal performance fluctuations.

3.2. The Concepts of Financial Performance and Political Connectedness

3.2.1. Financial Performance

The concept of performance has been widely discussed in the academic literature, with various definitions being available (Neely et al. 1995). Out of the possible conceptualizations of performance, financial performance has long been considered as the narrowest and least debated (Venkatraman and Ramanujam 1986). Nonetheless, a large variety of accounting-based financial performance indicators exist, of which financial ratios are the most prominently used, while numerous measures based on other publicly available financial information exist as well (Iwanicz-Drozdowska et al. 2016; Veganzones and Severin 2020). Financial ratios have distinct theoretical properties, which enable them to portray both good and bad performance (Lukason and Laitinen 2019). Still, their benefits can be limited, as the values of financial ratios can be volatile over time and context (e.g., sector, economic cycle) specific (Tan et al. 1997), leading to the need to apply a complex performance indicator that is less vulnerable to such deficiencies. A family of such indicators has long been in use in the form of failure prediction models, which provide firm failure probabilities mostly through the use of financial ratios (Altman et al. 2017). Such probabilities have been long indicated to serve well as proxies of performance (Dimitras et al. 1996). Single or complex indicators such as these enable the ordering of firms by declining performance. In addition, by relying on the theoretical foundations of the firm failure process (e.g., Weitzel and Jonsson 1989), payment defaults offer a measure for portraying the factual temporary poor performance of still-vital firms (Lukason and Andresson 2019). Thus, it was important to consider the concept of financial performance broadly in order to obtain holistic and robust results in the empirical portion of this study.

In Equation (1), FFP indicates firm financial performance, which we portray with different single and complex measures. For the complex measure, we applied the robust failure prediction model that was composed in Altman et al. (2017). This 2017 study presented the latest all-European SME-specific extension of the historic model introduced in Altman (1968). The model was built with millions of SME observations and has demonstrated high classification accuracy in many European countries, including Estonia. The “Model 2” from Altman et al. (2017) was applied in this study, providing a firm failure probability from 0 to 1 (coded as Altman Score). For the different single performance measures, we applied the four individual financial ratios that were applied in the Altman et al. (2017) model. These ratios portray firm liquidity (working capital to total assets, i.e., WCTA), annual profitability (earnings before interest and taxes to total assets, i.e., EBITTA), accumulated profitability (retained earnings to total assets, i.e., RETA), and leverage (total equity to total debt, i.e., BVETD). Since the first multivariate failure prediction model by Altman (1968), the latter variables have been very common indicators of firm performance (Dimitras et al. 1996; Lukason and Laitinen 2019). As financial ratios can have extreme values, they were winsorized before application in the regression analysis.

Aside from the indicators described above, we also applied one factual indicator of bad performance. This indicator reflects the presence of tax arrears (i.e., defaults on paying taxes), and based on earlier research (e.g., Lukason and Andresson 2019), we applied it as the number of month ends during the fiscal year for which a firm had tax arrears (coded as Tax Arrears). Thus, the minimum value was 0, and the maximum was 12. As tax arrears of different magnitude could be inherent to firms of different sizes, it was beneficial to use a payment default proxy, which could be free from the size context. In addition, the binary usage of tax arrears presence at least once per financial year would have been less useful, as it would not have enabled to consider the persistency of temporary payment defaults. As the latter variable could be calculated for firms liable to pay taxes, the population of firms applied in respective regressions was smaller than for other performance indicators.

The last financial indicator to be implemented in this study reflects firm export propensity. Namely, the ratio of export sales to total sales (coded as Export Share) was implemented. Exporters have generally been considered to be financially more well-off compared to their counterparts with only domestic orientation (e.g., Wagner 2007). The latter have been associated with a different set of factors, such as the exhaustion of home-market growth opportunities due to quick expansion, the availability of slack resources to implement international enlargement, and greater managerial capabilities in running the company. Thus, the share of export turnover from a firm’s total sales revenue can serve as an alternative complex financial performance indicator. Due to the absence of export information for many SMEs as a result of simpler reporting standards, the population of firms in respective regressions is substantially reduced.

3.2.2. Political Connectedness

The concept of political connectedness has been expressed in a multitude of ways in the literature (e.g., Faccio 2006; Sharma et al. 2020). First, studies diverge with respect to whether the research object itself is or was somehow politically involved (e.g., Wu et al. 2018) or is rather linked to politics or politicians due to ties of some kind (e.g., Faccio 2010). Studies focusing on direct connection seem to prevail, as most individuals could plausibly have at least some indirect political connections. Second, despite the variance in firm-level positions that have been studied, board members and directors seem to prevail over others, such as major shareholders. Third, studies that focus on the connections of managers exhibit a wide spectrum of connection modes. Most studies use a connection criterion of current or former activity as a member of parliament, minister, president, or other top official (e.g., Boubakri et al. 2008; Faccio 2010), but some studies also include participation in elections or party membership as a connection criterion (Khwaja and Mian 2005; Wu et al. 2018). Former top political officials are characteristic to studies focusing on listed multinational or government-owned companies and might not be applicable in the SME segment.

On the basis of these trends in the literature, in this section, we focus on describing the individuals that are responsible for governing a firm; furthermore, as this study investigated the whole population of firms of a single country, we considered it appropriate to look at connectedness broadly, by means of party membership and participation in elections, of which the latter can portray active connections, and the former, passive ones.

In Equation (1), PC denotes political connectedness. We addressed political connectedness via two dimensions. The first dimension is passive political connection, which is reflected through party membership and is constructed based on whether at least one board member of a firm at the time of compulsory annual report submission date (being the 30th of October 2020 for the fiscal year 2019) was a member of any of the 14 parties listed in Estonia at that time (coded as Passive PC). In Estonia, the average registered firm is a micro firm wherein the sole owner is also the only member of the board (Lukason and Camacho-Miñano 2020). It should be noted that for most of the included firms, the corporate governance system is simple and customized for the SME context; in such cases, the board (usually also overlapping with managers) is subordinate directly to owners (Süsi and Lukason 2019). For the second dimension, we further analyzed political connectedness by observing whether these party members were actively involved in politics. Specifically, we considered these party-connected board members “active” when at least one of them had recently participated in the general elections for Parliament in 2019 or the local elections in 2017 in Estonia (coded as Active PC).

4. Results and Discussion

4.1. Empirical Findings

Table 1. Descriptive statistics of dependent (D), independent (I), and control (C) variables.

Variable	Mean	Std. Deviation	Median	N
WCTA (D)	0.49	0.48	0.60	124,248
EBITTA (D)	0.09	0.36	0.03	124,248
RETA (D)	0.36	0.50	0.45	124,248
BVETD (D)	3.14	2.10	5.00	124,248
Altman Score (D)	0.27	0.25	0.25	121,933
Tax Arrears (D)	0.64	2.20	0.00	53,690
Export Share (D)	0.43	0.42	0.26	14,825
Passive PC (I)	0.08	0.27	0.00	124,248
Active PC (I)	0.02	0.14	0.00	123,840
Firm age (C)	10.38	7.27	8.45	124,248
Firm size (C)	10.32	2.27	10.17	124,248
Location (C)	0.47	0.50	0.00	124,248
Average board age (C)	46.43	11.72	45.82	124,248
Women on board (C)	0.36	0.48	0.00	124,248
Average time on board (C)	8.14	5.80	6.58	124,248

Note: statistics and frequencies for the year 2019.

presents the descriptive statistics of the population. In terms of the independent variables, 8% of the firms had party members on the board, while the share of those companies with politically connected members that also participated in the elections was around four times lower. In the whole population of adults, around 5% were party members (based on dividing the number of party members with the Estonian adult population); thus, politically connected individuals generally seem to be more entrepreneurial.

The average firm in the population was an old micro-firm with acceptable liquidity, low leverage, and profitable in the short- to long-run. Around half of the firms were situated in the capital. The boards were most frequently populated by middle-aged men, although around a third had a female board member as well. Characteristic for SMEs, members had long tenures on boards, especially when comparing firm age and average time on the board. The infrequent changes in the boards of Estonian SMEs have been noted in earlier studies as well (see, e.g., Süsi and Lukason 2021). The correlation matrix provided in Appendix A 

Table A5. Correlation matrix.

	WCTA	RETA	EBITTA	BVETD	Altman Score	Tax Arrears	Export Share	Passive PC	Active PC	Firm Age	Firm Size	Location	Average Board Age	Women on Board	Average Time on Board
WCTA	1.00	0.30	0.26	0.70	−0.64	−0.12	0.08	−0.03	−0.02	−0.05	−0.26	0.05	−0.03	0.07	−0.01
RETA		1.00	0.32	0.46	−0.59	−0.11	0.03	0.01	0.01	0.24	0.52	−0.01	0.12	−0.05	0.22
EBITTA			1.00	0.23	−0.50	−0.04	0.08	−0.01	−0.01	−0.12	0.10	−0.01	−0.09	0.01	−0.10
BVETD				1.00	−0.72	−0.19	0.04	−0.01	−0.01	0.06	−0.21	−0.04	0.06	0.08	0.08
Altman Score					1.00	0.12	−0.07	0.01	0.01	−0.03	−0.03	−0.01	−0.01	−0.02	−0.05
Tax Arrears						1.00	−0.08	0.03	0.01	−0.05	0.01	−0.02	−0.05	−0.04	−0.05
Export Share							1.00	−0.06	−0.04	−0.23	0.00	0.09	−0.07	−0.01	−0.22
Passive PC								1.00	0.46	0.05	0.02	−0.06	0.07	0.01	0.04
Active PC									1.00	0.04	0.02	−0.06	0.06	−0.01	0.03
Firm Age										1.00	0.34	−0.04	0.45	−0.02	0.76
Firm Size											1.00	0.04	0.16	−0.15	0.23
Location												1.00	−0.03	−0.03	−0.06
Average Board Age													1.00	0.03	0.43
Women on Board														1.00	−0.04
Average time on Board															1.00

Notes: Pearson correlations, which all have p < 0.05 except for BVETD × Location, Export Share × Firm Size, Export Share × Women on Board.

does not indicate any multicollinearity issues between the independent and control variables.

OLS results are presented in

Table 2. OLS regressions with seven different dependent variables (from year 2019) and Passive PC as the independent variable.

Independent & Control Variables	Dependent Variables
	WCTA	EBITTA	BVETD	RETA	Altman Score	Tax Arrears	Export Share
Passive PC	−0.0373 ***	−0.00913 *	−0.128 ***	−0.0189 ***	0.0113 ***	0.155 ***	−0.0694 ***
	(−7.72)	(−2.52)	(−6.08)	(−4.32)	(4.29)	(4.65)	(−5.32)
Firm Age	0.00101 ***	−0.00738 ***	0.0211 ***	−0.00325 ***	0.000583 ***	−0.00586 **	−0.0112 ***
	(3.34)	(−32.67)	(16.02)	(−11.93)	(3.55)	(−3.11)	(−17.83)
Firm Size	−0.0549 ***	0.0272 ***	−0.215 ***	0.116 ***	−0.00427 ***	0.00790	0.0180 ***
	(−86.16)	(56.99)	(−77.60)	(201.86)	(−12.36)	(1.53)	(9.26)
Location	0.0451 ***	−0.0177 ***	0.0298 *	−0.0302 ***	0.000342	−0.0638 **	0.0553 ***
	(16.62)	(−8.71)	(2.51)	(−12.30)	(0.23)	(−3.24)	(7.63)
Average Board Age	−0.000423 **	−0.00122 ***	−0.000321	−0.000300 *	0.000221 **	−0.00714 ***	0.00242 ***
	(−3.27)	(−12.59)	(−0.57)	(−2.56)	(3.14)	(−7.04)	(6.75)
Women on Board	0.0255 ***	0.0197 ***	0.0701 ***	0.0256 ***	−0.0103 ***	−0.0929 ***	−0.00770
	(9.05)	(9.36)	(5.71)	(10.08)	(−6.72)	(−4.42)	(−0.98)
Average Time on Board	0.00403 ***	−0.000317	0.0219 ***	0.0128 ***	−0.00299 ***	−0.000466	−0.00770 ***
	(11.14)	(−1.17)	(13.86)	(39.13)	(−15.22)	(−0.21)	(−10.28)
Constant	1.034 ***	−0.0352 ***	5.153 ***	−0.853 ***	0.311 ***	0.961 ***	0.334 ***
	(127.23)	(−5.78)	(145.41)	(−116.08)	(70.50)	(13.30)	(12.26)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	124,248	124,248	124,248	124,248	121,933	53,690	14,825
R2	0.083	0.044	0.069	0.299	0.010	0.006	0.087

t statistics in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001.

and

Table 3. OLS regressions with seven different dependent variables (from year 2019) and Active PC as the independent variable.

Independent & Control Variables	Dependent Variables
	WCTA	EBITTA	BVETD	RETA	Altman Score	Tax Arrears	Export Share
Active PC	−0.0518 ***	−0.00798	−0.129 **	−0.0337 ***	0.0159 **	0.246 ***	−0.0643 *
	(−4.90)	(−1.01)	(−2.80)	(−3.53)	(2.78)	(3.52)	(−2.15)
Firm Age	0.000998 ***	−0.00739 ***	0.0210 ***	−0.00327 ***	0.000583 ***	−0.00594 **	−0.0113 ***
	(3.30)	(−32.61)	(15.91)	(−11.95)	(3.55)	(−3.15)	(−17.82)
Firm Size	−0.0549 ***	0.0272 ***	−0.215 ***	0.116 ***	−0.00427 ***	0.00784	0.0182 ***
	(−86.04)	(56.98)	(−77.40)	(201.59)	(−12.34)	(1.52)	(9.33)
Location	0.0454 ***	−0.0177 ***	0.0305 *	−0.0302 ***	0.000325	−0.0641 **	0.0571 ***
	(16.68)	(−8.67)	(2.57)	(−12.28)	(0.22)	(−3.24)	(7.85)
Average Board Age	−0.000448 ***	−0.00123 ***	−0.000454	−0.000314 **	0.000229 **	−0.00717 ***	0.00239 ***
	(−3.45)	(−12.66)	(−0.80)	(−2.68)	(3.25)	(−7.05)	(6.67)
Women on Board	0.0250 ***	0.0195 ***	0.0682 ***	0.0254 ***	−0.0101 ***	−0.0911 ***	−0.00894
	(8.85)	(9.24)	(5.54)	(9.97)	(−6.60)	(−4.32)	(−1.14)
Average Time on Board	0.00403 ***	−0.000331	0.0219 ***	0.0128 ***	−0.00298 ***	−0.000447	−0.00772 ***
	(11.12)	(−1.22)	(13.86)	(39.04)	(−15.13)	(−0.21)	(−10.28)
Constant	1.034 ***	−0.0356 ***	5.152 ***	−0.854 ***	0.311 ***	0.972 ***	0.329 ***
	(126.99)	(−5.83)	(145.15)	(−115.98)	(70.40)	(13.43)	(12.06)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	123,840	123,840	123,840	123,840	121,529	53,477	14,783
R2	0.083	0.044	0.069	0.299	0.010	0.006	0.086

t statistics in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001.

, specifically for passive and active connectedness. Firms with politically connected boards are performing worse than their non-connected counterparts, leading to the acceptance of Hypothesis 2 and the rejection of Hypothesis 1. Out of the specific performance measures, all except annual profitability (EBITTA) indicated clearly significant associations, with p-value < 0.001 in most regressions. Politically connected firms exhibit lower liquidity, accumulated profitability and export share, higher leverage and failure risk, and more tax arrears. Hypothesis 2 holds in the case of both passive and active connections, with the significances being slightly weaker for the latter independent variable. Annual profitability, which exhibited the least significant results, might not be a good measure of performance, as profit can substantially fluctuate in time. The robustness checks with the two prior years (2018 and 2017) clearly support the findings from the year 2019, as Passive PC and Active PC remained significant at least for the p-value threshold of <0.05 (see Appendix A 

Table A1. OLS regressions with seven different dependent variables (year 2018) and Passive PC as the independent variable.

Independent & Control Variables	Dependent Variables
	WCTA	EBITTA	BVETD	RETA	Altman Score	Tax Arrears	Export Share
Passive PC	−0.0446 ***	−0.0126 ***	−0.153 ***	−0.0210 ***	0.0160 ***	0.156 ***	−0.0808 ***
	(−9.66)	(−3.64)	(−7.55)	(−4.99)	(6.36)	(4.57)	(−6.80)
Firm Age	0.00144 ***	−0.00774 ***	0.0227 ***	−0.00204 ***	0.000260	−0.00592 **	−0.0121 ***
	(5.02)	(−36.05)	(18.07)	(−7.81)	(1.67)	(−3.04)	(−20.92)
Firm Size	−0.0629 ***	0.0231 ***	−0.247 ***	0.110 ***	0.000135	−0.0253 ***	0.0228 ***
	(−101.16)	(49.59)	(−90.64)	(193.90)	(0.40)	(−4.72)	(12.95)
Location	0.0475 ***	−0.0139 ***	0.0325 **	−0.0277 ***	−0.00238	−0.0652 **	0.0584 ***
	(18.02)	(−7.06)	(2.81)	(−11.53)	(−1.66)	(−3.21)	(8.86)
Average Board Age	−0.000506 ***	−0.00118 ***	−0.000951	−0.000395 ***	0.000311 ***	−0.00636 ***	0.00218 ***
	(−4.07)	(−12.67)	(−1.74)	(−3.49)	(4.58)	(−6.14)	(6.86)
Women on Board	0.0132 ***	0.0182 ***	0.0411 ***	0.0195 ***	−0.00584 ***	−0.0987 ***	−0.0129
	(4.82)	(8.83)	(3.42)	(7.80)	(−3.93)	(−4.54)	(−1.81)
Average Time on Board	0.00318 ***	0.0000830	0.0196 ***	0.0117 ***	−0.00246 ***	−0.000682	−0.00780 ***
	(9.40)	(0.33)	(13.18)	(37.95)	(−13.35)	(−0.30)	(−11.57)
Constant	1.120 ***	0.0140 *	5.465 ***	−0.796 ***	0.261 ***	1.469 ***	0.316 ***
	(142.54)	(2.37)	(158.52)	(−111.12)	(61.13)	(20.05)	(12.89)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	129,088	129,088	129,087	129,088	126,833	59,830	16,070
R2	0.100	0.039	0.083	0.277	0.009	0.005	0.101

t statistics in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001.

and

Table A2. OLS regressions with seven different dependent variables (year 2017) and Passive PC as the independent variable.

Independent & Control Variables	Dependent Variables
	WCTA	EBITTA	BVETD	RETA	Altman Score	Tax Arrears	Export Share
Passive PC	−0.0381 ***	−0.00551	−0.133 ***	−0.0146 ***	0.00832 ***	0.224 ***	−0.0831 ***
	(−8.19)	(−1.60)	(−6.51)	(−3.50)	(3.32)	(6.72)	(−7.09)
Firm Age	0.00192 ***	−0.00766 ***	0.0260 ***	−0.00131 ***	0.000126	−0.00111	−0.0128 ***
	(6.73)	(−36.21)	(20.66)	(−5.11)	(0.82)	(−0.58)	(−22.23)
Firm Size	−0.0656 ***	0.0229 ***	−0.261 ***	0.107 ***	0.00215 ***	−0.0316 ***	0.0234 ***
	(−103.41)	(48.74)	(−93.63)	(187.98)	(6.28)	(−6.01)	(13.19)
Location	0.0530 ***	−0.0114 ***	0.0623 ***	−0.0202 ***	−0.00901 ***	−0.0728 ***	0.0545 ***
	(19.74)	(−5.74)	(5.28)	(−8.35)	(−6.23)	(−3.68)	(8.35)
Average Board Age	−0.000631 ***	−0.00109 ***	−0.00124 *	−0.000405 ***	0.000308 ***	−0.00576 ***	0.00163 ***
	(−5.03)	(−11.80)	(−2.25)	(−3.59)	(4.54)	(−5.73)	(5.28)
Women on Board	0.00624 *	0.0137 ***	0.0117	0.0108 ***	−0.00114	−0.0919 ***	−0.0138
	(2.23)	(6.59)	(0.95)	(4.27)	(−0.75)	(−4.33)	(−1.94)
Average Time on Board	0.00266 ***	0.0000960	0.0163 ***	0.0109 ***	−0.00211 ***	0.000735	−0.00685 ***
	(8.03)	(0.39)	(11.19)	(36.60)	(−11.82)	(0.34)	(−10.43)
Constant	1.135 ***	0.00913	5.529 ***	−0.785 ***	0.246 ***	1.435 ***	0.348 ***
	(142.31)	(1.55)	(157.68)	(−109.37)	(57.33)	(20.44)	(14.40)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	124,252	124,252	124,251	124,252	122,290	61,825	16,264
R2	0.103	0.036	0.086	0.272	0.008	0.005	0.096

t statistics in parentheses, * p < 0.05, *** p < 0.001.

for Passive PC,

Table A3. OLS regressions with seven different dependent variables (year 2018) and Active PC as the independent variable.

Independent & Control Variables	Dependent Variables
	WCTA	EBITTA	BVETD	RETA	Altman Score	Tax Arrears	Export Share
Active PC	−0.0690 ***	−0.0169 *	−0.177 ***	−0.0419 ***	0.0216 ***	0.162 *	−0.0751 **
	(−6.84)	(−2.24)	(−3.99)	(−4.55)	(3.95)	(2.23)	(−2.68)
Firm Age	0.00141 ***	−0.00776 ***	0.0226 ***	−0.00206 ***	0.000267	−0.00610 **	−0.0122 ***
	(4.92)	(−36.06)	(17.92)	(−7.87)	(1.71)	(−3.12)	(−20.90)
Firm Size	−0.0628 ***	0.0232 ***	−0.247 ***	0.110 ***	0.000116	−0.0251 ***	0.0230 ***
	(−100.90)	(49.62)	(−90.37)	(193.66)	(0.34)	(−4.68)	(13.04)
Location	0.0478 ***	−0.0136 ***	0.0340 **	−0.0277 ***	−0.00254	−0.0669 **	0.0603 ***
	(18.09)	(−6.87)	(2.94)	(−11.51)	(−1.77)	(−3.29)	(9.14)
Average Board Age	−0.000524 ***	−0.00119 ***	−0.00107 *	−0.000404 ***	0.000322 ***	−0.00624 ***	0.00216 ***
	(−4.20)	(−12.78)	(−1.96)	(−3.56)	(4.74)	(−6.01)	(6.76)
Women on Board	0.0128 ***	0.0181 ***	0.0397 ***	0.0193 ***	−0.00576 ***	−0.0958 ***	−0.0152 *
	(4.66)	(8.76)	(3.30)	(7.71)	(−3.86)	(−4.40)	(−2.12)
Average Time on Board	0.00318 ***	0.0000829	0.0196 ***	0.0117 ***	−0.00246 ***	−0.000637	−0.00785 ***
	(9.37)	(0.33)	(13.19)	(37.85)	(−13.32)	(−0.28)	(−11.60)
Constant	1.119 ***	0.0133 *	5.460 ***	−0.797 ***	0.261 ***	1.472 ***	0.310 ***
	(142.12)	(2.26)	(158.13)	(−111.09)	(61.15)	(20.06)	(12.65)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	128,648	128,648	128,647	128,648	126,399	59,597	16,026
R2	0.099	0.039	0.082	0.277	0.009	0.005	0.099

t statistics in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001.

and

Table A4. OLS regressions with seven different dependent variables (year 2017) and Active PC as the independent variable.

Independent & Control Variables	Dependent Variables
	WCTA	EBITTA	BVETD	RETA	Altman Score	Tax Arrears	Export Share
Active PC	−0.0634 ***	−0.00793	−0.131 **	−0.0360 ***	0.0161 **	0.179 *	−0.0955 ***
	(−6.29)	(−1.06)	(−2.96)	(−3.96)	(2.96)	(2.52)	(−3.46)
Firm Age	0.00190 ***	−0.00767 ***	0.0258 ***	−0.00132 ***	0.000132	−0.00111	−0.0129 ***
	(6.64)	(−36.20)	(20.51)	(−5.13)	(0.85)	(−0.58)	(−22.23)
Firm Size	−0.0657 ***	0.0229 ***	−0.261 ***	0.107 ***	0.00215 ***	−0.0318 ***	0.0235 ***
	(−103.29)	(48.66)	(−93.43)	(187.74)	(6.27)	(−6.04)	(13.26)
Location	0.0530 ***	−0.0113 ***	0.0637 ***	−0.0203 ***	−0.00903 ***	−0.0753 ***	0.0559 ***
	(19.73)	(−5.68)	(5.39)	(−8.40)	(−6.23)	(−3.81)	(8.54)
Average Board Age	−0.000630 ***	−0.00110 ***	−0.00135 *	−0.000410 ***	0.000308 ***	−0.00556 ***	0.00163 ***
	(−5.02)	(−11.79)	(−2.45)	(−3.62)	(4.53)	(−5.53)	(5.25)
Women on Board	0.00572 *	0.0137 ***	0.0102	0.0105 ***	−0.00106	−0.0881 ***	−0.0153 *
	(2.04)	(6.57)	(0.83)	(4.17)	(−0.70)	(−4.15)	(−2.15)
Average Time on Board	0.00266 ***	0.0000848	0.0163 ***	0.0109 ***	−0.00211 ***	0.000771	−0.00691 ***
	(8.04)	(0.35)	(11.19)	(36.51)	(−11.78)	(0.35)	(−10.48)
Constant	1.134 ***	0.00896	5.526 ***	−0.786 ***	0.247 ***	1.442 ***	0.343 ***
	(141.99)	(1.52)	(157.38)	(−109.30)	(57.31)	(20.51)	(14.16)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes
N	123,838	123,838	123,837	123,838	121,883	61,586	16,214
R2	0.103	0.036	0.085	0.272	0.008	0.004	0.094

t statistics in parentheses, * p < 0.05, ** p < 0.01, *** p < 0.001.

for Active PC).

The single performance indicators, by means of financial ratios, have clearly better explanatory power when compared with the Altman Score and Tax Arrears. Out of the applied four financial ratios, the model with accumulated profitability (RETA) led to the highest explanatory power (R²). When considering the explanatory power and p-value, factual defaults (Tax Arrears) seemed to be the least distinguishable among the groups of (non-)connected firms. The explanatory power of the model with Export Share exceeded those that included financial ratios, except for RETA. Depending on the specific measure of financial performance, the role of controls can vary, in the context of both signs and significances.

When considering the means of independent variables and their respective coefficients in regressions, then passive or active political involvement had the largest effect on the presence of Tax Arrears. Namely, in respective regressions, political involvement led to a 24% (for Passive PC) or 38% (for Active PC) increase in the occurrence of Tax Arrears in the respective twelve-month period. Given the low occurrence of tax arrears, the mean being 0.64 months over a 12-month period, in practice, the relevant increase cannot be considered remarkable. In the case of the Export Share variable, the respective figures were 16% and 15%. For the other independent variables, the reduction in performance did not exceed 10%. Thus, with respect to the economic implications, political connectedness decreased performance, while subjectively, the scale of that reduction was not substantial.

As an additional robustness test, we deconstructed the population based on the only binary variable (except for industrial dummies), namely, the presence of women on the board. As indicated in Table 1, 36% of boards have at least one woman involved. The respective coefficients (with t statistics in parentheses) and p-values of Passive PC are shown in

Table 4. Coefficients, p-values, and t statistics in the sub-populations of boards with women and boards with no women.

Sample Division	Women on Board			No Women on Board
Dependent Variable	Independent Variable Passive PC
	Coefficient	t Statistic	N	Coefficient	t Statistic	N
EBITTA	−0.00999	(−1.67)	44,422	−0.00788	(−1.68)	79,418
WCTA	−0.0483 ***	(−6.22)	44,422	−0.0303 ***	(−4.77)	79,418
BVETD	−0.138 ***	(−4.04)	44,422	−0.113 ***	(−4.08)	79,418
RETA	−0.0233 **	(−3.28)	44,422	−0.0159 **	(−2.79)	79,418
Altman Score	0.0160 ***	(3.66)	43,851	0.00718 *	(2.13)	77,678
Tax Arrears	0.121 *	(2.20)	16,308	0.184 ***	(4.31)	37,169
Export Share	−0.0464 *	(−2.06)	4088	−0.0783 ***	(−4.79)	10,695

* p < 0.05, ** p < 0.01, *** p < 0.001.

. Based on this robustness check, it can be concluded that Passive PC remained significant and maintained coefficients with the same sign, when compared with the base regression shown in Table 2. Organically, the absolute values of coefficients were to some extent altered, indicating a stronger or weaker association between politically connected boards led by women and different financial performance dependent variables. Namely, in the sub-population of woman-led boards, in case of political connectedness the decrease in financial performance was more pronounced in case of some dependent variables, namely three financial ratios and Altman Score. In turn, the contrary occurred for the Tax Arrears and Export Share dependent variables.

4.2. Scientific Implications

The main contribution of this study to the extant literature is that political connectedness in SME boards relates to the inferior performance of respective firms, therefore providing support to earlier studies indicating a negative nexus (e.g., Okhmatovskiy 2010; Domadenik et al. 2016). The result is stable over a variety of performance and political connectedness proxies, but also in terms of the analyzed timeline. On this basis, perhaps the resource-based theoretical approach behind the mainstream assumption of the positive effect of political connectedness on firm performance is valid only in the case of large or listed companies. Firms such as these might often be in need of high-end political connections that are likely to provide benefits in return. Meanwhile, for SMEs, possibilities to preferentially influence the external environment might be more limited, and thus, even the direct involvement of politicians in a company may fail to enhance its performance.

While the latter presented one potential explanation for the rejection of the positive link, then the argument for the established negative association might be more straightforward. We propose that director busyness (e.g., López Iturriaga and Morrós Rodríguez 2014) could be the main cause behind underperformance. More specifically, as the average board in these firms is populated by a single individual, such managers could be overloaded with different activities, and because of the latter, performance lags are caused by a loss of attention (Occasio 1997). As politically connected firms in the analyzed population are just underperforming in comparison to the opposite group, while still remaining profitable and not in a risk of acute failure, it could also be the case that due to their limited focus, managers have intentionally chosen a less risky business strategy. On average, a business strategy with lower risk propensity fails to outperform other market participants. The agency problem described by Shleifer and Vishny (1994) might not be an issue in this case, as normally there is some (or even full) overlap of the board and owners. Moreover, given Estonia’s remarkable transformation from a socialist state to market economy with high institutional quality and low corruption level, the latter premises of the positive association of political connectedness and financial performance acknowledged in the literature are probably not relevant for this study.

We have shown that politically connected firms perform worse than their non-connected counterparts with almost all performance proxies used. Thus, we can postulate that underperformance might be universal rather than specific. In terms of the extant literature, these results indicate that a variety of measures could be applied for future research, rather than maintaining the use of a few well-known measures. The only possible exception among the performance proxies is the widely used indicator of annual profitability (i.e., ROA), which might be too time-volatile for stable results.

The results indicate that passive and active politicians are characterized by similar performance patterns. This lends support to the use of various (and often quite broad) definitions of connectedness in the literature (e.g., Faccio 2006), as the results might not be definition-dependent. In addition, the results suggest that in the population of politically connected managers, those who become actively involved in politics might be characterized by a relatively random selection with respect to their entrepreneurial abilities.

While the signs and significances of political connectedness variables across different years remain unaltered, the regressions we have provided in the main text and appendixes show that coefficient values could vary. Thus, the magnitude of performance implications of political connectedness could be time-variant. This argument is additionally supported by the facts that the values of control variables in the dataset are relatively time-invariant and that their coefficients in respective models do not change as much as those of political connectedness variables.

5. Conclusions

By relying on the conflicting extant theoretical and empirical evidence, this study aimed to provide population-level evidence about the association of political connectedness and financial performance in the SME segment. The dataset included information about the active and passive political involvement of all Estonian board members, while the dependent variables reflected various measures of firm performance over multiple years. The results explicitly indicate that politically connected firms underperform their unconnected counterparts, thus lending support to the minority of studies postulating a negative nexus. Still, the significances and explanatory power of different independent variables reflecting political connectedness can vary depending on which performance indicator is applied. Firms with political connections more frequently exhibit lower liquidity, accumulated profitability and export propensity, higher leverage and failure risk, and more prevalent tax arrears. A potential explanation of this underperformance could be the greater busyness of politically connected board members, which leads to less attention to corporate activities. In turn, such lower attention can lead to, e.g., a lower willingness to implement riskier business strategies that require more monitoring. Obviously, the latter interpretations do not rule out other potential explanations of underperformance.

A practical implication for corporate shareholders is that, when involving a politically connected individual in the board, their concurrent involvement in politics might not help to improve performance. A potential solution might be a multi-person board with a clear role division in order to enhance focus and reduce the negative effects of potential work overload. An obvious implication for politically involved entrepreneurs is to consider whether balancing political and entrepreneurial activities against other matters (e.g., societal, family matters) is achievable, or instead, clear choices have to be made.

This study is not free from limitations, which in turn provide several avenues for developing the discourse further. Despite the application of a variety of different measures of financial performance and political connectedness that were suitable in the SME context, this research did not aim to study all possible perspectives. The multifaceted nature of financial performance provides opportunities to introduce additional variables; also, the study could be developed further to include non-financial measures. A similar limitation concerns the political connectedness variables that we applied, which could, for instance, be enriched by more precisely investigating political activeness. Such investigation might include, for instance, focusing on members of parliament or municipal councils or high-scoring candidates in various elections. Indeed, such an approach would dramatically narrow the number of politically connected board members in the dataset.

Furthermore, our study made no effort to identify causal relationships, as our objective was to address the extant theoretical and empirical controversies about the association of the two phenomena. Still, in the context of SMEs having only passive political connections, the impact of political connectedness on financial performance could be questionable. Derived from the latter, the obtained results should not be interpreted as causative proof that political connectedness leads to the underperformance of firms, as there could be a variety of other explanations that motivate our findings. Therefore, follow-up studies could aim at deconstructing the causal mechanisms of the (under)performance of politically connected SMEs.

We also note that due to data availability restrictions, political connectedness was defined statically in terms of board member’s party membership as revealed at the compulsory submission time of the year 2019 annual report. Although the most recent available information that was available indicated that changes in political connectedness and boards of SMEs—and especially, both of them together—were too infrequent in the short run to significantly alter the results, it would be beneficial (and also novel) to conduct a dynamic analysis that incorporates various scenarios of board connectedness changes.

Finally, despite the use of a country-wide whole population dataset that spanned multiple years, various country-level factors could still have an impact on how board connectedness could affect performance in the SME segment. Thus, replications of the methodology on other datasets that reflect various political, cultural, historic, and economic backgrounds remain necessary.