Effect of populism on the internationalization of emerging market firms

Abstract

Purpose

This paper aims to determine whether policy uncertainty caused by institutional decay in countries with populist rulers influences the internationalization decision of emerging market firms (EMFs).

Design/methodology/approach

The study used binary logit analysis on firms from Latin American countries undertaking cross-border greenfield investment projects.

Findings

The results suggest that internationalization decision is demotivated by policy uncertainty generated by populist chief executives and promoted by that of political parties.

Originality/value

This study uses populist rhetoric to describe policy uncertainty due to chief executives and ruling parties, which influences internationalization decision by increasing anticipated transaction costs. This inquiry identifies populism as a variable that influences EMFs to internationalize, while empirically testing the claim of theoretical scholarship that populism reconfigured the sociopolitical and institutional forces that shape the world’s business. This study further advances institutional theory by offering a fresh perspective on the influence of home instead of host-country institutions on the internationalization motivation of firms due to institutional decay caused by populist regimes.

Citation

Alcaraz, J., Martinez-Suarez, J. and Montoya, M.A. (2024), "Effect of populism on the internationalization of emerging market firms", European Business Review, Vol. 36 No. 1, pp. 12-31. https://doi.org/10.1108/EBR-01-2023-0025

Introduction

The political environment entails diverse key variables that provide an in-depth understanding of the impact of the institutional environment on global strategy (Devinney and Hartwell, 2020). The political environment of home countries, which is more unstable in developing than developed economies (Bengtson et al., 2013), leads to repercussions on firm strategies (Earl and Rose, 2022). Various disciplines took this outside-in approach to analyze political issues such as populism (Casson, 2021); however, international business (IB) scholars displayed interest in political events only after the last financial crisis (2008–2009) due to increased uncertainty related to populist leaders and parties worldwide (Aguilera et al., 2019; Amore and Corina, 2021).

Populism (a specific set of ideas characterized by a Manichean distinction), between good (the people) and evil (the elite; Mudde, 2004), presents important implications for political environments. The reason is that its anti-establishment ideology changes and challenges the status quo, which sets the context for a new reality that deserves more attention from the IB literature to comprehend its implications on firm strategies (Ghauri et al., 2021). This notion focuses on the effect of the institutional context of emerging markets on the internationalization decision of local firms (Losada-Otálora and Andonova, 2022), because such markets may provide a fertile ground for the interventionism of the state in the economy. This aspect is also related to the increase of protectionist policies pervasive to integration among countries and, consequently, increased global uncertainty and costs for cross-border transactions (Ciravegna and Michailova, 2022).

From a broad perspective, the populist ideology challenges the legitimacy of establishments and intermediate power structures that regulate business–state relationships, which fosters an environment of permanent political uncertainty (Devinney and Hartwell, 2020). Furthermore, as Hartwell and Devinney (2021) highlight, “the current populist moment may outlive its present protagonists and continue to influence the global business environment for decades to come.”

Recent studies cited that institutional deterioration in countries ruled by populists may shape the nonmarket strategies of firms (Blake et al., 2022; Panibratov et al., 2022), affect entrepreneurial activity (Bennett et al., 2022) and influence the allocation of foreign direct investment (FDI) (Carballo Perez and Corina, 2023; de Sousa et al., 2021). Thus, the potential effect of this deterioration on the internationalization motivation of emerging market firms (EMFs) remains under-explored. This topic is not trivial, because home-country institutions can influence internalization decision due to increased government support and encouraging policies or as a consequence of poor market-based institutions and legal frameworks (Boisot and Meyer, 2008; Xu et al., 2021). Accordingly, we aim to determine if uncertainty due to institutional decay in countries with populist rulers can influence the internationalization decision of EMFs.

Our theoretical framework is anchored on institutional economics with a focus on transaction costs (Dorobantu et al., 2017). We empirically test the influence of institutional decay in countries with populist rulers on the internationalization decision of firms. We build on recent work that uses populist rhetoric to describe how chief executives and ruling parties create policy uncertainty (Bennett et al., 2022; Carballo Perez and Corina, 2023), which influences the decision of the latter to internationalize by increasing anticipated transaction costs faced by EMFs. We performed binary logit analysis on 1,108 EMFs from 13 Latin American countries (henceforth, Multilatinas) across 15 years (2004–2018). These Multilatinas conducted 2,492 cross-border greenfield projects across 94 destination countries. The main results suggest that the internationalization decision of Multilatinas is demotivated by the populist rhetoric of chief executives and promoted by the populist rhetoric of political parties.

We contribute to the literature by identifying populism as a variable that encourages Multilatinas to internationalize due to policy uncertainty as a result of the institutional decline of the home country. We also empirically test the claim of theoretical scholarship that populism reconfigured the sociopolitical and institutional forces that shape global business (Blake et al., 2022; Devinney and Hartwell, 2020; Hartwell and Devinney, 2021). We advance institutional theory by demonstrating that the internationalization decision of domestic firms is partially influenced by changes in the rules of the game (North, 1990), which occur under populist regimes. Lastly, the study serves as a guide for scholars and practitioners in illustrating the effect of political environments, in specific populism, on the global strategy decisions of firms (Casson, 2021).

Literature review and hypothesis

Domestic institutions and internationalization

The institutional economics perspective (North, 1990) completes transaction cost and internalization theories that explain the factors motivating firms to internationalize (Peng et al., 2008). This approach is relevant, because institutions can shape firm strategies, and such capacity is increasingly evident in emerging economies (Wright et al., 2005). We focus on formal institutions, such as those referring to codified laws (e.g. property rights and contract enforcement), central to setting the parameters required to guide all actors in countless transactions within a given society (Kim and Aguilera, 2016). In this regard, domestic institutions constitute features that determine the international strategies of local firms (Tallman and Pedersen, 2015).

Former empirical evidence suggests that the institutional environment in emerging countries influences the internationalization decision of domestic firms (Cuervo-Cazurra, 2008). This notion holds for the Chinese overseas investment scale, which is influenced by domestic institutional hardship (regulatory restriction, bureaucracy, local protectionism, informal payment, bribery and corruption and nontransparent and inadequate policies and services) (Luo and Wang, 2012). In addition, for Chinese firms, institutional fragility (business–government interfaces, development of private firms, development of product markets, development of facto markets and development of market and legal intermediaries) pushes their internationalization decision (Shi et al., 2017). This behavior remains the same when home governments implement discriminatory policies that favor state-owned enterprises (Wei et al., 2015) from specific industries (Enderwick, 2017). Deng and Zhang (2018) identified that low institutional quality in China promotes the decisions of small- and medium-sized enterprises to internationalize.

From the historical perspective, Barnard and Luiz (2018) in South Africa propose that the internationalization decision of local firms is due to increased misalignment and societal contestation (institutional deterioration and contained contestation, periods of societal instability and/or inadequate institutional reforms and pervasive societal instability and/or fundamental changes in institutions); authors use measures for institutional rights (property and political rights) and social stability. Stoian and Mohr (2016) analyzed 29 emerging countries in Europe, Latin America, Asia, Africa and the Middle East from 1995 to 2011 and suggested that regulative voids (protectionism and corruption) influence outward FDI (OFDI).

In Latin America, Aguilera et al. (2017) analyzed unique features to better understand the influence of the home country on firm behavior and their internationalization. The study concludes that the idiosyncrasies of domestic markets, such as suboptimal regulation or protectionism through tariff and subsidy, potentially encourage the international diversification of firms. Similarly, Finchelstein (2017) examined the influence of state actions, such as subsidies and regulatory support, on the internationalization process of large firms in Argentina, Brazil and Chile. The findings demonstrate that “different sets of public policies have consequences on the sectorial diversification, pace, and number of large internationalized firms.” In a recent study, Alcaraz et al. (2023) identified that Latin American firms’ internationalization decisions are influenced by the weak institutional conditions in their home country.

Populism is a political force that reshapes the institutional frameworks of countries and influences the internationalization motivation of firms by imposing policies (macro and micro) to attain redistribution. This scenario occurs in populist regimes in Latin America (Edwards, 2019) or in the promotion of protectionist and isolationist populist policies, as in Central and Eastern Europe (van der Waal and de Koster, 2017). The current study focuses on this thus-far-missed relationship.

Populism and institutional environment

In the political science literature, three approaches are considered the most influential in examining the populist phenomenon (Rovira Kaltwasser et al., 2017), namely, political–strategic (Weyland, 2001), sociocultural (Ostiguy, 2009) and ideational (Mudde and Rovira-Kaltwasser, 2018). This study adopts the ideational approach, which recently attracted scholarly attention. It defines populism as a thin ideology characterized by a Manichean struggle between good (i.e. the people) and evil (i.e. the elite), which privileges the general will of the people.

To determine how populism alters an institutional environment, identifying the specific features of the populist phenomenon that potentially influences the set of laws, regulations, administrative procedures and policies formally sanctioned by the government is necessary. We posit that deinstitutionalization and short-term policy bias (Blake et al., 2022) promote an environment of political uncertainty in populist regimes.

Deinstitutionalization is a process through which populists steadily undermine the institutional safeguards and procedures of modern democratic governance. By claiming that democratic institutions are unnecessary intermediaries that serve evil elites in their ascendance to power (Hawkins, 2009), populists openly engage in the erosion of established political channels. Thus, they can implement other tools of direct democracy (Barr, 2009) to fulfill the will of the people. Therefore, government decisions may not be taken and enforced in accordance with traditional rules and regulations. Instead, alternative institutional channels may be used, such as plebiscites and referenda, which the populist regime considers legitimate.

Regarding short-term policy bias, populists may adopt policies that deliver immediate benefits, which are outweighed by much larger costs in the long run (Edwards, 2019). Populist incumbents magnify:

[…] short-term policy bias by (a) shortening the temporal gap between policy adoption and the perception of benefits received by the populist’s supporters and (b) further increasing the size of future policy costs, i.e., lowering the net present value of a given policy’s economic impact (Blake et al., 2022, p. 6).

In summary, populism deinstitutionalization and short-term policy bias critically exacerbate political uncertainty due to an unclear indication that institutions will continue to support market exchange and protect business and private property rights, which leads to an anticipated increase in transaction costs.

Effect of populism on firms’ internationalization

Institutions provide the context for the arena in which firms play. We state that populism is a part of this context given its broad implications to countries that adopt this stream. Populists represent a source of political uncertainty, which “disrupt(s) a country’s political and economic institutions by challenging, weakening, undermining, or eliminating pre-existing rules of the game previously set by the ruling elite” (Carballo Perez and Corina, 2023, p. 4). Therefore, institutional decay in countries ruled by populists offers a key landscape for analyzing the effect of the former on the internationalization decision of Multilatinas. This view is related to the pivotal role of home-country institutions in the internationalization of firms (Aguilera et al., 2017; Peng et al., 2008).

To examine the influence of institutional decay and resultant policy uncertainty on the internationalization decision of Multilatinas, we use the populist rhetoric of leaders and ruling parties to describe the creation of policy uncertainty by political actors (Bennett et al., 2022; Carballo Perez and Corina, 2023). These types of rhetoric are deeply interlinked, which may render differentiating the effects on the internationalization of firms difficult. A sensible approach for analyzing the rhetoric of populist leaders and political parties is to conduct the analysis separately to better visualize if such actors influence the internationalization motivation of firms.

Contemporary populism belongs to strong political actors that occasionally receive direct mandates from the people to govern, as expressed by voting results. The pressure is high because political leaders, at least at the beginning of their presidential period, frequently hold high shares of supporters. For example, populist leaders in Latin America won the presidency with more than 50% of the total votes from 2003 to 2018 (Grzymala-Busse and McFaul, 2020). Thus, expecting that the populist rhetoric of leaders reflects their intention to undertake immediate actions to fulfill the will of the people is favorable.

For example, the “democratic and cultural revolution” proclaimed by Evo Morales changed Bolivia’s constitution and directly threatened the old business elites by promising to redistribute land to the poor (Wolff, 2016). Ecuador’s Rafael Correa, alternatively, declared himself as the leader of a citizen’s revolution that would reshape the country’s institutional architecture by pursuing an anticorporatist agenda that cut the channels of communication with business associations (Goeury, 2021). Early in his presidency, Andrés Manuel López Obrador fulfilled his campaign promise of canceling the construction of Mexico City’s new international airport due to its alleged harm to the environment and corruption practices. As such, he sent investors the message that previous long-term commitments between past administrations and private firms may no longer be honored (Panibratov et al., 2022).

The populist rhetoric of leaders in developed economies also signals a potential increase in policy uncertainty. For instance, Donald Trump’s statements in support of a closed and aggressive trade policy for protecting domestic firms and jobs in the US led to the renegotiation of the North America Free Trade Agreement (NAFTA, currently the United States-Mexico-Canada Agreement). He claimed that other countries took advantage of the US through “unfair” trade deals (Lester and Manak, 2018). NAFTA’s renegotiation created a climate of uncertainty for firms in the trading bloc, because Trump’s view was based on restrictive rules of origin for reducing the number of tariff-free usable materials, which resulted in higher tariff for inputs or final goods.

Thus, the rhetoric of populist leaders encourages private firms to engage in internationalization due to the threat posed to the established institutional structure and related costs. Thus, the study presents the following hypotheses (Figure 1):

H1.

The populist rhetoric of political leaders from the home countries of Multilatinas is likely to promote their internationalization decision.

Although presidential systems grant the executive with power to control other institutions (legislature and judiciary; (Pérez-Liñán et al., 2019)), highlighting the relevance of political parties to political systems is necessary. Political parties possess longer-term goals than do politicians (Levitsky, 2018), and the capacity of presidents to influence law and policy formulation depends on the relationship between their constitutional and partisan powers (Mainwaring and Shugar, 2002).

Populist parties, such as leaders, are political outsiders prone to adopt radical positions related to policymaking due to their interpretation of politics through the good-versus-evil lens of morality. de Sousa et al. (2021, p. 7) state that, “[…] they may also pursue ill-conceived, kneejerk and sound-bite solutions to complex problems, without taking into consideration the financial sustainability and externalities of their decisions.” In addition, the policymaking of populist parties presents procedural features that constitute new challenges for business. For example, policymaking in nonpopulist democracies is heavily influenced by area-specific technocrats, whose expertise provides firms with a view of the potential effects of a given policy on economic and sociopolitical environments. Conversely and by definition, populist policymaking is against expert positions, because it is responsive to the majoritarian preferences of the people (Caiani and Graziano, 2022).

For example, when asked which economists supported Brexit, UK cabinet minister Michael Gove responded that “people in this country have had enough of experts” (Mance, 2016). Similarly, Mario Delgado, leader of the National Regeneration Movement (Morena), lashed out at a political opposition who criticized the revocation of mandate consultation by twitting “Conservatives are terrified that the most important decisions in the country are no longer made by an elite, but by the people of Mexico” (Newsroom, 2022).

Consequently, the populist rhetoric of parties could signal an inclination to fulfill the will of the people by circumventing established institutional channels, while unconditionally passing laws, programs, projects and budget allocations in the senate or congress. This tendency translates into immediate benefits for constituencies, which ignores the potential long-term adverse effects, such as the loss of foreign human capital, shortfall in budget, inflation and currency devaluation, of such initiatives on macroeconomic and sociopolitical stability. Therefore, we present the following hypothesis:

H2.

The populist rhetoric of the political party over which a political leader presides is likely to promote the internationalization decision of Multilatinas.

Methodology and data

The study used various datasets to rigorously test the hypotheses and evaluate the influence of the populist discourse of presidents and the populism of political parties on the internationalization decision of firms. We obtained the sample from fDi Markets, a database developed by the Financial Times group, which tracks and records greenfield projects from 2003 to date. It also registers project date, firm name, source and destination countries, industry, capital invested (Capex) and jobs created. This database excludes the financial identifiers of companies (e.g. Ticker and PermID). It is a limitation to the retrieval of firm-level information in studies that use many firms from diverse countries (Loncan, 2021). Nevertheless, it is not a constraint, as demonstrated by scholars (Duanmu, 2014; Wu et al., 2022).

The sample is integrated by 1,108 parent companies from 13 countries in Latin America. These firms performed 2,492 greenfield projects from 2004 to 2018 across 94 destination countries. Appendix provides a list of the countries. Given the data limitations, atypical data from the pandemic, and potential biases to the results, we considered 2018 as the last year of study. We focused on Latin America, because support for populist mobilizations regained unparalleled traction over the past two decades, and leaders considered to be at least somewhat populist rule a number of the largest economies in the region (Lewis et al., 2019).

The study used greenfield investment projects as the unit of analysis. Firms from emerging markets frequently undertake high-risk and high-control investment entry modes, such as greenfield investments, to expand operations abroad (Luo and Tung, 2007). Thus, increased attention is required for studies on greenfield investment (Jiang et al., 2021) to unveil the internationalization strategies of EMFs.

Dependent variable

Following Luo et al. (2019) and Yang (2018), the dependent variable is dichotomous and is coded 1 if a firm decides to internationalize, and 0 otherwise. This dummy variable captures the internationalization decision of firms as influenced by the populist discourse of presidents and the populism of ruling political parties from the firms’ home country.

Independent variable

The key independent variables cover the populist discourse of presidents and the populism of political parties. To measure the first, we utilized the Global Populism Database (GPD), which is part of the New Populism Project (affiliated with the Guardian Media Group). The second and most recent version of the data set covers 241 chief executives from 74 countries and includes 324 government terms and 1,240 speeches between 2000 and 2022 (Hawkins et al., 2019). GPD measures the level of populist discourse from political leaders (presidents and prime ministers) through textual analyses of speeches, coded as 0 (uses few if any populist elements), 1 (uses strong populist elements but also non-populist elements) and 2 (includes extremely populist elements). GPD calculates the final scores for political leaders during their term as the average of speeches delivered throughout their tenure. We used the method of Hawkins (2009) to measure populism as the discourse of political leaders, which is consistent with recent studies such as Bennett et al. (2022) and Carballo Perez and Corina (2023). Measuring political discourse is an appropriate approach given the strong connection between populism and leaders governing populist movements (Rode and Revuelta, 2015). Focusing on the chief political executive, the study intends to effectively capture the essence of populism and its association with the actions and rhetoric of leaders.

Regarding the populism of political parties, we use the Global Party Survey, a research project based at Harvard University and published in 2020. Experts (1,861 scholars of parties and elections) from 163 countries answered the questionnaire, which comprises 21 core items rated using a 0–10-point continuous scale that estimates the ideological values, issue positions and populist rhetoric of 1,043 political parties. We used item 8, which is the central measure for populist rhetoric (Norris, 2020a, p. 10). It states:

Parties can also be classified by their current use of POPULIST OR PLURALIST rhetoric. POPULIST language typically challenges the legitimacy of established political institutions and emphasizes that the will of the people should prevail. By contrast, PLURALIST rhetoric rejects these ideas, believing that elected leaders should govern, constrained by minority rights, bargaining, and compromise, as well as checks and balances on executive power. Where would you place each party on the following scale? 0 Strongly favors pluralist rhetoric…10 Strongly favors populist rhetoric.

The measure of populist rhetoric is recognized as a reliable and valid indicator of the populism position and ideological values of political parties (Norris, 2020b). Recent studies employed this approach to investigate populism in various contexts (de Jong et al., 2022; de Reguero and Jakobson, 2023), which further supports its applicability and significance in contemporary research.

We aim to explore the impact of populism on the investment decisions of Multilatinas and focus on the influence of political leaders and their political parties. Thus, we introduce a variable that measures the level of populism attributed to political leaders, which enables the assessment of its effect on internationalization decision. Additionally, we incorporate a variable for evaluating the further influence of populism associated with the political party led by the president on internationalization decision. This research intends to shed light on the complex relationship between populism and corporate decision-making in Latin America.

Control variables

Given the nature of the data set, we control for the home country conditions of firms and the environments of host countries. In terms of home-country variables, we measure institutional conditions using the World Governance Indicator (WGI) of the World Bank (WB) (Kaufmann et al., 2010). Given the multidimensionality of institutional conditions (Shi et al., 2017) we use an aggregated variable established using a simple average of the six dimensions (Fon et al., 2021) of the WGI. This process is correct, because the dimensions present high correlations among themselves (Slangen and Beugelsdijk, 2010). Following Cuervo-Cazurra and Genc (2008), we modified the original scale (−2.5 to 2.5) by adding 2.5 to each grade to obtain a scale ranging from 0 to 5, which eases interpretation.

Gross national income (GNI) enables the measurement of the economic performance of home countries (Shi et al., 2017); the larger the economic performance, the more the chances that firms from these countries obtain the required resources for internationalization. We measure infrastructure using the number of fixed-line and mobile telephones per 100 inhabitants using data from the WB (Cuervo-Cazurra and Genc, 2008). Firms from emerging economies operate under poor infrastructure conditions, which undermines and potentially hinders performance.

Host-county conditions also influence internationalization decision; thus, we include variables to control for this aspect. We control for trade openness in host countries, which could be considered part of pro-market reforms and could reduce transaction costs, provide better governance conditions and favor market performance, which can motivate the internationalization of firms from emerging countries (Borda Reyes et al., 2019). We use trade freedom of the Heritage Foundation to measure trade openness, which estimates freedom in trade policy (absence of tariff and non-tariff barriers affecting foreign trade of goods and services). We measure FDI openness using the investment freedom of the Heritage Foundation (Cuervo-Cazurra, 2006), whose rationale is similar to that for trade openness. This variable identifies FDI freedom and capital flow; firms can move resources without restriction in countries without constraints on FDI. Host-country economic development, which indicates economy size, is an appealing factor for firms (Buckley et al., 2007). We measured it using the GNI of the countries; data were retrieved from the WB.

We also control for variables that capture differences between home and host countries that influence the internationalization decision of firms such as common colony, common language and distance (Alcaraz et al., 2020). We use a dummy variable to measure if two countries share colonial ties (coded as 1, and 0 otherwise). Common language is a dummy variable that measures whether people from two countries share the same language (coded as 1, and 0 otherwise). The fact that individuals from two countries share the same language decreases transaction costs and facilitates IB activities. Distance (in kilometers) denotes the geographical distance between two countries; the larger the distance, the lower the chances of cross-national business operations. We obtained data for these variables from Centre d’Études Prospectives et d’Informations Internationales, a French research center dedicated to international economics.

Additionally, we control for firm-level variables because they can influence internationalization decision. First, we control for the internationalization experience of firms. Based on the availability of the dataset, we build a proxy variable using the cumulated number of projects by a firm (Wu et al., 2022). Second, we control for firm-owned resources by building a dummy variable (Loncan, 2021) using the median of Capex on projects (Duanmu, 2014); large- and small-scale projects take values of 1 and 0, respectively. Third, using the two-digit industry codes of the North American Industry Classification System, we control for firm industry (Dau et al., 2020) given that the internationalization decision of firms could vary across industries (Luo et al., 2019).

Statistical methods

Given the dependent variable nature in the panel data set, we use a binary logit model as the estimation method to test the hypotheses. The previous literature indicates that logistic models are helpful for estimating the probability of the international investment decisions of firms (Kong et al., 2020; Saikia et al., 2020; Valentino et al., 2019). This probability considers diverse conditions, such as firm, home and host-country features (Clegg et al., 2018), and bilateral variables that reflect differences between source and destination countries (Thede and Karpaty, 2023). The logit model uses the maximum likelihood method for estimation, which avoids the assumption of multivariate normality and equal covariance matrixes (Fischer, 1973; Trueck and Rachev, 2009). To adequately capture variability in the yearly observations of firms and shifts over time, we use random effects in the analysis to control for the characteristics of unobserved firms (Cuervo-Cazurra and Dau, 2009). To avoid heteroskedasticity, we use robust standard errors (Belderbos et al., 2020; Yang, 2018) clustered on parent companies (instead of investing companies) that perform greenfield investment projects. Furthermore, we lag by one-year time-varying explanatory variables, which is a broadly accepted practice, to control and mitigate potential issues related to endogeneity (Alcaraz and Salamanca, 2018; Clegg et al., 2018; Mazouz et al., 2021; Saikia et al., 2020). We present the logit regression model as follows:

Pr(OFDIijt=1|X)=β0+β1PopPolLeadjt−1+β2PopPolParjt−1+δX+εijt

where i refers to the firm of interest, j denotes home country and t represents the time of occurrence of OFDI, that is, firm decision to perform a greenfield international investment project. PopPolLead_jt−1 denotes the populist discourse of political leaders, and PopPolPar_jt−1 represents the populism of political parties; both are derived from the home countries of firms. H1 and H2 predict that the coefficients of populist discourse from political leaders (β₁) and of the populism of political parties (β₂) is positive in the abovementioned base logit regression model. X is a vector of regressors and includes the control variables. Regarding home-country control variables, PolEnv_{jt − 1} denotes the political environment, EcPerf_{jt − 1} stands for the economic performance of countries, and Inf_{jt − 1} represents infrastructure. ComCol_jk, ComLang_jk and Dist_jk are control variables that capture differences between home and destination countries and represent common colony, common language and distance, respectively. For host-country control variables, TradOp_{kt − 1} denotes trade openness, FDIOp_{kt − 1} represents FDI openness and EcSize_{kt − 1} pertains to economic size. Finally, for firm control variables, InteExp_{it − 1} refers to firm international experience, Capex_{it − 1} represents the size of the project and Ind_i is the firm industry.

Results

Table 1 presents the correlation matrix and descriptive statistics. We calculated the variance inflation factor (VIF) to ensure that no multicollinearity exists among the independent variables. All values are below the cutoff of 10 (Kutner et al., 2004); the largest value was 2.51, and a mean VIF of 1.57. Thus, the results suggest that multicollinearity is not a major concern.

Table 2 presents the logit estimations. Model 1 is the base model and reports the results of the control variables. In Model 2, we introduced the populist discourse of political leaders. Contrary to expectations, the result is negative and statistically significant (β = −0.356, p < 0.01); thus, H1 is not supported.

In Model 3, we added the populism of the party of the political leader. As expected, the result is positive and statistically significant (β = 0.035, p < 0.01). Therefore, H2 is supported. Model 4 simultaneously analyzes the populism of the political leader and party. The results remain the same as those for Models 2 and 3. The results for the estimated control variables were similar in each of the four models.

To assess robustness, we performed additional analyses (available upon request). We employed factor analysis to construct a single measure of the populism of political parties using various indicators from the Global Party Survey (Norris, 2020a): program, populist rhetoric, populist saliency, environment, nationalism, clientelism, people should decide, and strongman rule.

Through factor analysis, we extracted a single-factor score with varimax rotation, which is used in the literature; e.g. (Krug and Falaster, 2023). All factor loadings (range: 0.607 to 0.937) exceed the cutoff (≥ 0.40). The cumulative explained variance of this factor is 0.761. Following Pett et al. (2003), we conducted the Bartlett test of Sphericity and the Kaiser–Meyer–Olkin (KMO) test to assess the adequacy of data in factor analysis. The result of Bartlett’s test was significant at p < 0.000, which indicates that correlations among variables are sufficiently high for building common factors. The KMO measure of sampling adequacy was 0.753. The results are considered satisfactory and suitable (Bartlett’s test: p < 0.05; KMO: superior to 0.6) for factor analysis (Malhotra, 2019). We then use the single-factor score as a variable denoting the populism of political parties. In summary, the results of robustness analyses are consistent with the findings in Table 2.

Discussion

This study builds on the idea that populism can influence the internationalization decision of Multilatinas, because it represents a source of political uncertainty, which increases transaction costs due to the deterioration of the institutional and legal frameworks of the home country. Surprisingly, although populist leaders exert a deterrent effect on the internationalization decision of Multilatinas, populist parties may promote this decision. The following reasons might explain the outcomes.

Populist leaders may discourage the internationalization decision of Multilatinas due to the political instability generated by survival prospects of populists and policy coherence. For the survival of populists, statistical analyses demonstrate that during 1990–2018, populist leaders frequently left office under dramatic circumstances, as only one-third of them finished their constitutional terms. However, a larger number were forced to resign or were impeached, as illustrated by Fernando Lugo (Kyle and Mounk, 2018) in Paraguay and, recently, Pedro Castillo in Peru. Instances occur in which constitutional change enabled populist actors to overstay in power; nonetheless, these cases could constitute outliers in which democratic institutions are particularly weak (Weyland, 2022). The early departure of populist leaders is typically related to democracy backsliding and the “irresponsible, unsustainable policies that frequently end in disaster and hurt most the ordinary people they purportedly aim to help” (Rodrik, 2018).

Policy coherence refers to “the ambiguity, complexity, and dynamism that is introduced by policy change” (Utesch-Xiong et al., 2022) and may influence internationalization decision due to unclear regulations (ambiguity), increased instruments through which policies are implemented (complexity) or increased scope of government intervention (dynamism). In most presidential democracies, presidents are formally authorized to introduce bills, which makes them primary lawmakers (Mainwaring and Shugar, 2002). Bills introduced by populist chief executives may considerably increase the time and resources required by firms to determine the extent to which policy changes (positively or negatively) influence operations. The reason is that any shift in legislation under populism will benefit (harm) companies according to the degree to which the market activities of the former promote the interests of the regime and its “people” (Blake et al., 2022). Consequently, firms may adopt a “wait-and-see” approach before embarking on international diversification.

In terms of populist parties, we contend that Multilatinas are more likely to internationalize because firms perceive such political actors as outsiders (Barr, 2009) that lack appropriate qualifications and work experience for policymaking and for taking important positions in government (Levitsky, 2018). In addition, populist parties are weakly or less institutionalized movements (Goldfajn et al., 2021) associated with the emergence of political neophytes. They are more likely to “respond poorly to crises” and tend to “lack skills in bargaining and negotiation, coalition building, dealing with opponents, handling media, and other areas that are critical to governing and legislating” (Levitsky, 2018). Consequently, they reduce the space for negotiation and compromise with partisan opposition while sidelining technocratic experts that may challenge incumbent populist policies (Blake et al., 2022).

This inquiry identifies the influence of institutional decay under populism on the internationalization decision of Multilatinas. The empirical model is grounded in a political phenomenon that exploits and weaponizes social grievances such as populism. Thus, we contribute to the debate that calls for additional IB scholarship that elucidates the impact of political, economic and social trends on IB (Casson, 2021). Simultaneously, it tests the claim of theoretical research that populism reconfigured the sociopolitical and institutional forces that shape the global business environment (Blake et al., 2022; Devinney and Hartwell, 2020; Hartwell and Devinney, 2021).

Academic interest in populism and its implications for the strategic responses of firms is increasing (Blake et al., 2022; Feldmann and Morgan, 2022; Hartwell and Devinney, 2021). However, the IB literature focuses on advanced economies due to watershed events that shook the political setting of countries with institutionally robust democracies such as the British vote to leave the European Union (Brexit) and the election of Donald Trump to the US Presidency. We advance knowledge in the field by offering empirical analysis that suggests that uncertainty due to populist regimes in a geographic area, which are characterized by a long history of poor institutional and legal frameworks, seemingly partially influences the internationalization motivation of Multilatinas (Levitsky, 2018).

By incorporating populist rhetoric to describe the creation of policy uncertainty by chief executives and ruling parties (Bennett et al., 2022; Carballo Perez and Corina, 2023), we provide a novel approach for assessing institutional volatility, because many studies evaluate the said variable through indices that measure economic policy uncertainty (Baker et al., 2016), quality of governance (Kaufmann et al., 2010), or the extent to which any political actor or its replacement is constrained by their choice of future policies (Holburn and Zelner, 2010). Nevertheless, no study considers the potential impact of populism on the institutional framework of a country and its corresponding effects on the internationalization process of firms.

In this regard, the IB literature continuously calls for an in-depth understanding of the context related to IB phenomena. We claim that populism is a social and political phenomenon that exemplifies the importance of local contexts when examining the strategic choices of firms. Populism is a thin ideology whose core elements, such as “the people,” “the elite,” and the “will of the people,” are empty vessels that are open to being variably filled by populists according to context (Mudde and Rovira-Kaltwasser, 2013; Rodrik, 2018). Thus, firms need to identify the potential implications of each element in policymaking and to adjust their strategy accordingly. For example, in its left-wing variant, populism pursues redistributive policies and inclusive enfranchisement, because it is mainly driven by class conflict. Alternatively, populism perceives globalization as an alien that intends to debase national culture and society, which facilitates xenophobia, protectionist policy and exclusive enfranchisement (Mudde and Rovira-Kaltwasser, 2018).

Conclusions

The present study further advances institutional theory by offering a fresh perspective on the potential influence of home, instead of host, country institutions on the internationalization decision of firms due to institutional decay caused by populist regimes. Populists disrupt the legacy of political establishment by dismantling the institutional system that has served the interests of “evil elites” (Hawkins, 2009). This notion inevitably implies the need to change the rules of the game, which renders them unstable and, in extreme cases, arbitrary. For example, by appointing loyalists to independent institutional bodies, such as courts and bureaucracies, populists neutralize checks and balances; consequently, firms cannot rely on impartial adjudication if they will contest laws or regulations that harm their assets and returns (Blake et al., 2022).

This inquiry contributes to managers and practitioners. First, it can serve as reference for understanding the implications of populism on the strategic choices of companies due to the abovementioned deterioration. Second, it guides in explaining the potential responses of firms to policy-related uncertainty derived from populist leaders and parties. Finally, it enriches static political indicators (readily available from management consulting agencies) by including populism as a variable that increases policy uncertainty and shapes the internationalization strategy of firms.

This study has its limitations. First, although the definition of populism comprises the necessary and sufficient attributes that distinguish populist from non-populist actors and can be applied consistently to different geographical regions (Bennett et al., 2022), the study does not make a clear-cut distinction between its variants such as left- and right-wing populism. Accordingly, future research could focus on identifying which of the aforementioned variants poses an encouraging or deterring effect on the internationalization motivation of firms. Second, this research is deemed preliminary, because populist regimes after 2018 continued to rule a few of the countries examined such as Nicaragua. Meanwhile, other countries experienced the populist phenomenon immediately at the end of 2018 such as Mexico. Furthermore, the largest economy in the region, namely, Brazil, committed an unmistakable right-wing populist through Jair Bolsonaro as head of state during 2019–2022. Thus, a larger sample could reveal a clear view of the role of populism – and its variants – in the internationalization decision of firms. Similarly, the sample comprises firms operating in countries with presidential systems. Thus, future studies can examine other political systems such as parliamentary ones. Finally, this inquiry analyzes the internationalization decision of firms via greenfield investments; therefore, room exists for examining the effect of populism at each stage of the international diversification process (including location) of firms combined with specific entry modes.

Figures

Figure 1.

Conceptual model

Table 1.

Pairwise correlation matrix

Variables		1	2	3	4	5	6	7	8	9	10	11	12	13	14
1	OFDI Dummy	1
2	Populism political leader	−0.015	1
3	Populism political party	−0.005	0.061	1
4	Institutional env. home	−0.004	−0.091	−0.329	1
5	Economic size home	0.018	−0.221	−0.392	−0.339	1
6	Infrastructure home	−0.023	−0.197	−0.093	0.225	0.238	1
7	Common colony	−0.006	0.013	0.036	−0.046	0.008	0.045	1
8	Common language	−0.005	−0.006	0.235	0.125	−0.464	−0.164	0.269	1
9	Distance	−0.004	−0.088	−0.196	0.014	0.378	0.229	0.168	−0.485	1
10	Trade openness host	0.019	−0.067	0.008	−0.089	0.101	−0.061	0.165	−0.037	0.043	1
11	FDI openness host	0.012	−0.052	0.114	0.032	−0.191	−0.066	−0.048	0.323	−0.255	−0.002	1
12	Economy size host	0.008	−0.079	−0.091	−0.049	0.247	0.117	−0.011	−0.542	0.491	0.162	−0.478	1
13	International experience	0.085	0.046	−0.037	−0.008	−0.011	−0.086	0.028	−0.011	−0.009	−0.007	−0.028	0.064	1
14	Capex scale	0.447	0.005	−0.013	0.003	0.006	−0.043	−0.004	0.003	−0.015	0.008	−0.012	−0.016	0.082	1
	Mean	0.069	0.137	4.591	0.929	27.564	2.969	0.048	0.408	8.399	4.357	0.925	27.573	1.474	0.034
	S.D.	0.253	0.181	1.778	0.163	1.204	0.197	0.213	0.491	0.896	0.128	0.779	1.858	1.437	0.182
	Min.	0	0	2.1	0.439	23.532	1.385	0	0	5.199	3.161	−5.418	21.234	0	0
	Max.	1	1.733	9	1.322	28.771	3.509	1	1	9.868	4.499	6.213	30.705	25	1

Source: Own elaboration

Table 2.

Logit estimates with robust standard errors

Dependent variable: OFDI Dummy
Variables	Model 1		Model 2		Model 3		Model 4
Pop. pol. leader	---		−0.356**	(0.129)	---		−0.329**	(0.128)
Pop. pol. party	---		---		0.035**	(0.015)	0.029*	(0.015)
Inst. env. home	0.353***	(0.104)	0.287**	(0.104)	0.571***	(0.147)	0.481**	(0.152)
Ec. performance home	0.097***	(0.017)	0.081***	(0.017)	0.131***	(0.025)	0.111***	(0.025)
Infrastructure home	−0.517***	(0.093)	−0.549***	(0.095)	−0.593***	(0.101)	−0.615***	(0.105)
Common colony	−0.269**	(0.099)	−0.236*	(0.097)	−0.261**	(0.099)	−0.231**	(0.097)
Common language	0.012	(0.043)	−0.021	(0.045)	0.002	(0.043)	−0.027	(0.044)
Distance	−0.042*	(0.022)	−0.043*	(0.022)	−0.042*	(0.021)	−0.042*	(0.022)
Trade openness host	0.545***	(0.156)	0.518***	(0.157)	0.527***	(0.156)	0.504***	(0.157)
FDI openness host	0.088**	(0.028)	0.089**	(0.028)	0.086**	(0.028)	0.087**	(0.157)
Economic size host	0.013	(0.011)	0.009	(0.011)	0.011	(0.012)	0.008	(0.012)
International Experience	0.153***	(0.013)	0.156***	(0.013)	0.154***	(0.013)	0.156***	(0.013)
Capex scale	−0.161	(0.131)	−0.157	(0.129)	−0.162	(0.131)	−0.159	(0.129)
Industry	Yes		Yes		Yes		Yes
Constant	−6.997***	(0.864)	−6.141***	(0.891)	−7.951***	(0.987)	−7.017***	(1.027)
Observations	29,391		29,391		29,391		29,391
Log pseudolikelihood	−7,432.843		−7,429.452		−7,430.966		−7,428.071
Wald test: χ2	252.91***		267.74***		252.63***		265.97***

Notes:

Significance levels: ***p < 0.001, **p < 0.01, *p < 0.05, ⁺p < 0.1; Robust standard errors (clustered on investing parent companies) are reported in parentheses

Source: Own elaboration

Table A1.

List of countries

Home countries:

Argentina, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guatemala, Mexico, Nicaragua, Panama, Paraguay, Uruguay

Host country:

Algeria, Angola, Argentina, Australia, Austria, Bangladesh, Belgium, Belize, Bolivia, Brazil, Cambodia, Canada, Chile, China, Colombia, Costa Rica, Croatia, Czech Republic, Denmark, Dominican Republic, Ecuador, Egypt, El Salvador, Ethiopia, Finland, France, Germany, Ghana, Greece, Guatemala, Guinea, Guyana, Haiti, Honduras, Hungary, India, Indonesia, Iran, Ireland, Israel, Italy, Jamaica, Japan, Jordan, Kazakhstan, Kenya, Latvia, Lebanon, Libya, Luxembourg, Malawi, Malaysia, Malta, Mexico, Moldova, Morocco, Mozambique, The Netherlands, New Zealand, Nicaragua, Nigeria, Norway, Oman, Pakistan, Panama, Paraguay, Peru, Philippines, Poland, Portugal, Qatar, Romania, Russia, Rwanda, Saudi Arabia, Singapore, Slovakia, South Africa, South Korea, Spain, Sri Lanka, Sweden, Switzerland, Tanzania, Thailand, Trinidad and Tobago, Turkey, UAE, Ukraine, United Kingdom, USA, Uruguay, Vietnam, Zambia

Source: Own elaboration