Dynamic Spillovers of Economic Policy Uncertainty: A TVP-VAR Analysis of Latin American and Global EPU Indices

Abstract

This study examines the dynamic interconnectedness of economic policy uncertainty (EPU) among Latin American economies—Brazil, Chile, Colombia, and Mexico—and significant international regions, including the United States, Europe, and Japan, as well as a global EPU index. Using a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model with monthly data, this study reveals the evolving spillover effects and dependencies capturing how uncertainty in one market can transmit across others on both regional and global scales. The findings highlight the significant impact of external EPU, particularly from the U.S. and global EPU sources on Latin America, positioning it as a primary recipient of international uncertainty. These results underscore the need for Latin American economies to adopt resilience strategies—such as trade diversification and regional cooperation—to mitigate vulnerabilities to global shocks. This study offers valuable insights into the mechanisms of economic uncertainty transmission, guiding policymakers in developing coordinated responses to reduce the effects of external volatility and foster regional economic stability.

1. Introduction

In recent years, the rapid advance of globalization has strengthened the connections between national economies, institutions, and policies worldwide. This increased integration means that economic policy changes in one country can have ripple effects on other countries, even those not directly affected by the original policy decisions (Nyakurukwa & Seetharam, 2023). Major global events over the last few decades, including the 2008 Global Financial Crisis, the 2018 U.S. trade war, the COVID-19 pandemic in 2020, and the 2022 Russian–Ukrainian conflict, have intensified economic policy uncertainty (EPU). Policymakers faced with the challenge of stabilizing economies amid such crises often employ untested strategies, thereby introducing additional layers of unpredictability. The impact of EPU spans various economic and financial domains, such as stock markets, international trade, the cryptocurrency market, industrial production, and employment, as reported in studies conducted by Abaidoo (2019), Paule-Vianez et al. (2021), and Roma et al. (2021).

Theoretical and empirical research has examined the channels through which EPU influences broader macroeconomic and financial variables. Song et al. (2022) highlight two principal mechanisms underlying this relationship: the supply–demand market sentiment channels. Increased uncertainty within the supply–demand channel can dampen production incentives, leading to financial market volatility and macroeconomic indicator instability. Empirical evidence shows that EPU can reduce investment (Gulen & Ion, 2015; Baker et al., 2016; Kong et al., 2022), play an increasingly crucial role in driving fluctuations in oil demand (F. Liu et al., 2024; Degiannakis et al., 2018), affect stock market liquidity (Paule-Vianez et al., 2020; Mishra et al., 2024; Raza et al., 2023; Xiao et al., 2024), and heighten exchange rate volatility (Huang et al., 2024; Chen et al., 2020; Bush & Noria, 2021; Rúa & Marín-Rodríguez, 2024). Such volatility often causes short-term shifts in commodity demand, disrupting the balance of supply and demand over the long term and triggering broader economic imbalances.

These spillover effects and their further explanation through global feedback loops have attracted substantial scholarly interest as researchers strive to understand the multiplier effects of EPU on economic performance. Notably, Antonakakis et al. (2018), Zhou et al. (2022), K.-H. Wang et al. (2023), and Kayani et al. (2024) have demonstrated that domestic uncertainty is shaped by both internal and external uncertainty spillovers, underscoring the importance of distinguishing between a country or region’s internal and external policy uncertainty transmission processes. By capturing both direct and indirect effects, it becomes possible to assess the broad impacts of uncertainty shocks, which are often prolonged by international spillovers. This emphasizes the need for policymakers to establish “early-warning systems” (Diebold & Yilmaz, 2012) that enable them to mitigate potential adverse effects by adjusting policy instruments. Moreover, identifying which forms of uncertainty exert the most influence is crucial, as some types may have more substantial economic implications than others (Mumtaz & Surico, 2018).

Most studies have traditionally focused on country-level uncertainty propagation mechanisms. However, recent advancements, such as the work by Gabauer and Gupta (2018), have introduced a more granular perspective by analyzing categorical economic policy uncertainty spillovers between nations, particularly between the U.S. and Japan, across categories like monetary, fiscal, trade, and currency policy uncertainties. This approach offers insights into both internal and external transmission mechanisms, revealing the amplification effects of international feedback and the persistence of uncertainty shocks within interconnected economies. The availability of detailed EPU indices, including those for Greece (Hardouvelis et al., 2018), facilitates a deeper examination of uncertainty dynamics across specific regions.

Zhou et al. (2022) also analyzed EPU spillovers across 19 developed and emerging economies, including some Latin American countries, finding significant cross-border effects that intensified during major events like COVID-19. Their study mapped EPU transmission paths and highlighted increased spillover impacts from pandemic-related restrictions; however, it did not explore the specific effects on Latin America. Furthermore, while the study provided valuable insights into EPU spillover dynamics, it lacked a distinction between internal (regional) and external (global and major economies) spillovers.

Building on this foundation, the present study addresses this gap by examining the complex and interconnected nature of economic policy uncertainty (EPU) across global and Latin American economies using a Time-Varying Parameter Vector Autoregression (TVP-VAR) model. This approach provides essential insights for policymakers and investors, offering a framework to anticipate the varied impacts of EPU on emerging markets. With a focus on Latin America’s vulnerability and resilience to global economic fluctuations, this study supports the development of more effective policy responses, ultimately promoting economic stability in the region.

The primary objective of this study is to investigate the interconnected dynamics of economic policy uncertainty (EPU) across global and Latin American economies, utilizing a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model. To achieve this, this study formulates and addresses the following research hypotheses (RHs), each aimed at deepening our understanding of EPU spillover effects and their implications:

• RH1: external EPU significantly influences Latin American economies more than internal EPU, making these economies primarily recipients of global uncertainty spillovers.

This hypothesis stems from prior studies, such as Antonakakis et al. (2018) and Zhou et al. (2022), which highlight the dominant role of external EPU in shaping domestic uncertainty in interconnected markets, particularly in emerging economies.

• RH2: within Latin America, economies like Brazil and Mexico act as regional EPU transmitters, but their role is secondary to their dependence on global economic interactions.

This hypothesis draws on Zhou et al. (2022) and Kayani et al. (2024), examining how Brazil and Mexico simultaneously serve as receivers of external EPU and limited regional transmitters within Latin America.

• RH3: the intensity of external EPU impacts varies across Latin American economies, influenced by their degree of economic integration with global markets.

Inspired by K.-H. Wang et al. (2023), this hypothesis examines how economic linkages, such as Mexico’s close ties to the U.S., shape the heterogeneity of EPU spillover effects within the region.

• RH4: strengthened institutional frameworks can reduce Latin America’s vulnerability to external EPU, highlighting the role of governance and policy stability.

This hypothesis emphasizes the insights of K.-H. Wang et al. (2023), regarding the importance of institutional quality in mitigating adverse effects from external uncertainties.

• RH5: regional cooperation through frameworks such as the Pacific Alliance and Mercosur can alleviate shared EPU risks, fostering economic stability in Latin America.

Reflecting on studies like Gabauer and Gupta (2018), this hypothesis investigates how collaborative economic policies can reduce systemic spillover impacts in interconnected markets.

By analyzing the interconnectedness between regional and global EPU indices, this study captures the influence of external uncertainties on Latin American countries, which share common macroeconomic structures and exposure to international economic forces. This approach provides a framework for assessing cross-regional spillover effects and potential systemic impacts. By addressing these hypotheses, this study aims to offer actionable insights for policymakers, financial regulators, and investors, facilitating more effective strategies to mitigate the impacts of economic policy uncertainty in Latin America. The hypotheses are systematically explored in the Section 5 and Section 6, ensuring a clear linkage between findings and this study’s objectives and providing valuable insights into the mechanisms of uncertainty transmission and its implications for global macroeconomic stability.

The structure of this study is as follows: Section 2 provides a literature review, offering an overview of key literature clusters on economic policy uncertainty and a historiographic analysis of EPU and TVP-VAR research, focusing on their foundations, evolution, and emerging trends. Section 3 describes the empirical data and methodology, detailing the dataset and the TVP-VAR model. Section 4 presents empirical findings, focusing on the general, internal, and external connectedness results, while Section 5 discusses the implications of these findings. Finally, Section 6 concludes this study with final remarks.

2. Literature Review

2.1. Literature Clusters

Figure 1 presents the interconnectedness and relationships between prominent research terms associated with EPU indices and TVP-VAR methodologies. Using VOSviewer software version 1.6.20 (van Eck & Waltman, 2017) to map data from Scopus and Web of Science, the network illustrates clusters of keywords related to various themes, such as spillovers, volatility, geopolitical risks, and the influence of economic policy uncertainty across global contexts. This network serves as a foundation for analyzing the dynamic connectedness and spillover effects of EPU indices using a TVP-VAR approach, a focal point of this study. The analysis reveals four primary clusters: red, green, blue, and yellow, which will be examined in detail below.

2.1.1. Cluster 1 (Red): Drivers and Transmission Channels of Global EPU

Cluster 1 centers on the dynamic and interconnected nature of EPU indices and its spillover effects across both Latin American and global economies, focusing on how policy-related uncertainties, particularly those linked to monetary and economic policies, influence financial stability and decision making on an international scale. This study employs a TVP-VAR model to capture the evolving nature of these spillovers, with the extended framework by Antonakakis et al. (2020) providing insights into cross-country EPU transmission, as demonstrated in studies of US–Japan and US–China economic interactions (Gabauer & Gupta, 2018; Jiang et al., 2019).

Keywords such as trade, investment, exchange rates, and stock markets emphasize the economic dimensions most affected by EPU, with studies indicating that these areas are particularly sensitive to policy-induced shocks. For instance, trade tensions and exchange rate fluctuations between major economies impact uncertainty transmission, underscoring the need for coordinated responses to mitigate adverse effects (Antonakakis et al., 2019). Additionally, macroprudential policies are suggested as a stabilizing measure, especially in markets like real estate and finance, which are significantly impacted by macroeconomic uncertainties within the United States (Gabauer & Gupta, 2020).

Specific economic shocks, such as those from oil and gold markets, further underscore the complex pathways through which EPU affects market volatility. Studies highlight that oil supply shocks predominantly drive volatility spillovers to the gold market, influencing hedging strategies and demonstrating the interconnectedness of commodity and financial markets under conditions of heightened uncertainty (Mokni et al., 2020). Geopolitical risk and global health crises, notably the COVID-19 pandemic, have also intensified the transmission of EPU across regions, with significant impacts on international trade, investment, and stock market stability (Youssef et al., 2021).

This cluster underscores the need for coordinated policy responses to manage the cascading effects of EPU shocks across interconnected economies and the relevance of understanding EPU mechanisms in stabilizing global economic and financial systems. The analysis highlights the role of international collaboration in addressing volatility and implementing robust economic stability measures, informed by empirical insights into EPU’s impact on financial and commodity markets. Importantly, this cluster also reveals that while studies have explored global dynamics extensively, the Latin American context remains underexplored, highlighting a critical gap that this study aims to address.

2.1.2. Cluster 2 (Green): Financial Market Reactions and Risk Management Under EPU

Cluster 2 focuses on critical financial concepts and market responses to EPU indices across global and Latin American markets. Central themes include volatility, spillovers, risk, and returns, highlighting how EPU-induced uncertainty intensifies market interconnectedness and influences investment outcomes in both developed and emerging economies (Gabauer & Gupta, 2018; Jiang et al., 2019; Antonakakis et al., 2019).

Keywords like “investor sentiment”, “hedge”, and “gold” reflect typical investor behaviors amid uncertainty. Investor sentiment denotes the collective outlook of market participants, which can drive either increased volatility or stability based on economic signals. Safe-haven assets such as gold and hedging strategies serve as protective measures against market fluctuations and are prevalent risk management responses to EPU (Mokni et al., 2020; Youssef et al., 2021). The literature on safe-haven assets underscores how economic uncertainty drives behavioral shifts among investors seeking stability through traditional assets (Gabauer & Gupta, 2020).

Moreover, terms like “oil price”, “stock”, and “markets” point to specific assets sensitive to policy changes, where forecasting and pricing models are essential for managing the risk that these fluctuations entail. Studies exploring EPU’s impact on assets such as oil and stocks reveal that EPU amplifies volatility and affects predictability, necessitating robust risk management and forecasting models to navigate uncertainty in market returns (Christou et al., 2020; Apostolakis et al., 2021). This cluster emphasizes the interconnected nature of financial markets under EPU and the value of hedging and predictive strategies in managing risk and maintaining market stability. Despite this extensive body of literature, there is limited focus on Latin America’s financial markets and their unique vulnerabilities to EPU-induced spillovers, a gap that this study addresses by examining these dynamics within the region.

2.1.3. Cluster 3 (Blue): Methodologies and Transmission Mechanisms of EPU Spillovers

Cluster 3 focuses on essential methodological and analytical tools for examining the spillovers and transmission mechanisms of EPU indices across regions. Key terms, including causality, determinants, dynamic connectedness, and transmission, strongly emphasize identifying the drivers and causal pathways linking EPU across economies. The cluster highlights econometric techniques like impulse response analysis and unit root tests, which capture the impact of shocks from one market on others over time, especially under conditions of macroeconomic volatility. These methods are crucial in disentangling complex causal relationships and revealing how uncertainty propagates across interconnected markets.

For example, research on the U.S. and China highlights the role of bilateral trade and exchange rate policies in affecting cross-country spillovers (Jiang et al., 2019). Studies focused on Greek and European markets also underscore how internal and external EPU shocks impact fiscal and monetary policies over time (Antonakakis et al., 2019).

Thus, the U.S. plays a particularly significant role in this dynamic, as its macroeconomic conditions strongly influence global and Latin American economies, often generating broad-reaching spillover effects. This underscores the need for dynamic connectedness methods like TVP-VAR, which provide insights into the intricate causal pathways through which uncertainty spreads. The inclusion of these econometric tools, along with the focus on macroeconomic volatility and efficient testing, reflects this study’s aim to understand how policy shocks in major economies can trigger ripple effects across emerging markets, carrying implications for investors and policymakers alike. However, the application of these methodologies to Latin American economies remains scarce in the literature, and this study contributes to filling this gap by providing region-specific insights.

2.1.4. Cluster 4 (Yellow): Nonlinear Dynamics and Specific Shocks in EPU

Cluster 4, which includes nonlinear causality and oil price shocks, underscores the importance of exploring specific types of shocks and their nonlinear effects on economic policy uncertainty (Mokni et al., 2020; Elsherif, 2024). Nonlinear causality highlights the complex relationships between variables, acknowledging that economic systems often exhibit nonlinear responses to sudden or extreme shocks. This complexity is especially relevant for events like oil price shocks, which can cause disproportionate and unpredictable impacts across both energy and broader financial markets, potentially destabilizing economic stability.

Studies conducted by Z. Liu et al. (2023) reveal that oil price shocks, whether from supply, demand, or risk factors, impact various markets in asymmetric and nonlinear ways, intensifying economic policy uncertainty and spreading volatility across financial systems (Mokni et al., 2020). These findings highlight the critical need for policy frameworks considering nonlinear dynamics, as shocks in one area can propagate unpredictably through interconnected economic and financial networks, underscoring the necessity of robust, adaptive response mechanisms. Despite these insights, existing studies do not adequately address nonlinear dynamics within the Latin American context, which is a significant gap this study seeks to fill by examining specific shocks and their implications for this region.

In conclusion, the current body of literature reveals a gap in understanding the intricate and evolving relationships among economic policy uncertainties, specifically due to the absence of studies focused on the Latin American context. This lack of research leaves critical questions unanswered about how economic policy uncertainties propagate within and across Latin American economies and their interactions with global indices. This proposed study aims to address these gaps by applying advanced dynamic modeling to quantify the spillover effects of economic policy uncertainty, a necessary step to provide robust evidence on the region’s vulnerability and its reliance on external shocks. Understanding these relationships is crucial for identifying potential areas of economic fragility and resilience. Additionally, the empirical insights generated will offer policymakers essential guidance in formulating resilient, proactive strategies to mitigate the destabilizing effects of economic policy uncertainty, ultimately contributing to better-informed decision making that supports economic stability and development in Latin America.

2.2. Historiographic Analysis of EPU and TVP-VAR Research: Foundations, Evolution, and Emerging Trends

As Garfield (2004) described, the historiographic map visually represents a chronological network of key citations within a bibliographic collection, constructing a matrix that traces the historical connections between foundational works. This approach provides insights into temporal patterns and scholarly impact, as Aria and Cuccurullo (2017) outlined. According to Marín-Rodríguez et al. (2023) and Vogel et al. (2021), such maps reveal the progression of core studies, showing the transmission of ideas across research.

Figure 2 illustrates a single cohesive cluster, indicating a concentrated lineage of influence within the field. Thus, this presents a historiograph of the academic development of EPU and Time-Varying Parameter Vector Autoregressive (TVP-VAR) research, with four clusters differentiated by color (red, blue, purple, and green) to indicate thematic or methodological groupings. Then, each color represents a distinct subfield, such as foundational works (red), regional applications (blue), sector-specific studies (purple), and emerging themes (green). This visualization captures the evolution and diversification of research, highlighting foundational studies and their subsequent influence on specific subfields. The leading studies are centered around key influential studies, notably Gabauer and Gupta (2018) in red, which acts as a critical node for the entire network. This study explores internal and external EPU spillovers between the U.S. and Japan, pioneering the disaggregation of monetary, fiscal, currency, and trade policy uncertainties. Also, this highlighted the Fukushima Daiichi accident as a substantial negative trade shock, demonstrating how discrete events can propagate international spillovers, particularly in monetary policy.

In this way, the red cluster extends to several studies between 2019 and 2024, such as Apostolakis et al. (2021), Mokni et al. (2020), and Youssef et al. (2021), and represents studies utilizing TVP-VAR and related methodologies to analyze EPU spillovers in financial and commodity markets. Significant connections suggest that this methodological framework has become a standard approach within the field, with subsequent research building on it to examine specific financial variables, macroeconomic indicators, and spillover effects. Additionally, studies conducted by Zou et al. (2024) and Ren et al. (2024) represent the most recent research. These studies are likely applying or extending the TVP-VAR methodology to analyze EPU impacts in even newer or more complex domains, possibly influenced by global events like the COVID-19 pandemic. The high connectivity among these nodes suggests that recent events have spurred a surge in research focusing on EPU spillovers in response to contemporary global challenges, leading to fresh insights within the field.

This network’s branching indicates both a buildup of existing methodologies and an exploration into new applications or geographic contexts over time. Each color in the historiograph corresponds to a distinct thematic or methodological grouping, allowing for a visual understanding of the diverse research trajectories within EPU and TVP-VAR literature.

On the other hand, the blue cluster is positioned mainly in the upper section. This includes studies such as Degiannakis et al. (2018), E.-Z. Wang and Lee (2020), and Y. Wang et al. (2022). This grouping focused research on regional or country-specific applications of EPU, examining how policy uncertainty affects localized economies or specific markets (e.g., China or other Asian markets). The peripheral placement of this cluster, still connected to the central red cluster, suggests that these studies are influenced by Gabauer and Gupta (2018) foundational methodologies but apply them in geographically distinct contexts.

The purple cluster, which includes studies conducted by Assaf et al. (2021) and Christou et al. (2020), suggests an orientation toward specific sectors or nuanced areas of finance, potentially examining the impact of EPU on non-traditional markets or alternative assets. This group may focus on topics such as real estate, tourism, or niche investment vehicles that experience unique impacts from policy uncertainty. The placement of the purple nodes implies that this line of inquiry is more specialized and has evolved somewhat independently from the central, finance-focused research.

Finally, the green cluster, represented by studies conducted by L. Liu et al. (2020), seems smaller and less interconnected with the main network, potentially indicating newer or less mainstream applications of EPU analysis. This could include emerging areas such as environmental finance, green bonds, or sustainability-linked financial instruments, where the field is beginning to establish connections with EPU research. The relatively isolated placement suggests that this area is still developing within the broader EPU framework.

This historiographic map provides a well-organized representation of the progression of EPU research, with each color-coded cluster capturing a unique subfield or methodological approach. The red cluster denotes the foundational and widely adopted TVP-VAR methodology, influencing a broad range of applications in finance and economics. Furthermore, it illustrates the field’s current focus on contemporary challenges. The blue, purple, and green clusters represent extensions of this work into regional, sectoral, and emerging areas. This structure reflects a dynamic research area that, while rooted in foundational methodologies, continues to branch into new and evolving topics.

Finally, the historiographic mapping of EPU and TVP-VAR research highlights a well-documented evolution within the field, yet it also reveals a notable research gap concerning EPU spillovers specific to Latin America. While foundational studies have explored EPU’s cross-national impacts, particularly between major economies like the U.S. and Japan, the focus has remained mainly on developed nations or singular economic events. As a result, the unique dynamics of EPU within Latin American economies—and their interactions with global uncertainties—remain insufficiently examined. This gap underscores the need for a dedicated analysis of how EPU spillovers influence global and Latin American markets, especially given the region’s economic interdependence and vulnerability to external shocks. Addressing these issues will contribute to a complete understanding of EPU’s role in emerging markets and provide valuable insights for more tailored economic policies in Latin America.

3. Empirical Data and Methodological Approach

3.1. The Dataset and Research Framework

The data selection for the chosen Latin American countries—Brazil, Chile, Colombia, and Mexico—was based on their shared macroeconomic frameworks, specifically their similar exchange rate regimes (floating and free-floating) and their adoption of inflation-targeting monetary policies. Additionally, the availability of consistent and reliable data for these countries was a fundamental factor in their inclusion, as they represent the major economies in Latin America. Other Latin American countries were excluded due to the lack of such data, which would have compromised the robustness and comparability of the analysis. Furthermore, the selected countries are the largest and most influential economies in the region, accounting for a significant proportion of Latin America’s GDP. This selection ensured that the analysis would be robust and comparable across all regions examined, providing meaningful insights into the dynamics of economic policy uncertainty in the region.

These common characteristics imply that these countries may experience similar impacts on their macroeconomic variables in response to shocks in international variables. To complement this analysis, the EPU indices of the selected Latin American countries will be contrasted with those of the U.S., Europe, Japan, and a global EPU index. This comparative approach fosters a deeper understanding of how economic policy uncertainty in Latin American economies aligns with global market trends.

Table 1. List of EPU index returns analyzed.

Variable	Label	Description
EPU Index for Brazil	EPU_BRA	Developed by Baker et al. (2016) using articles from the journal Folha de São Paulo.
EPU Index for Chile	EPU_CHI	Developed by Cerda et al. (2016) using articles from the journals El Mercurio and La Segunda, following Baker et al. (2016).
EPU Index for Colombia	EPU_COL	Developed by y Gil León and Silva Pinzón (2019) using data from the journal El Tiempo, based on Baker et al. (2016).
EPU Index for Mexico	EPU_MEX	Constructed by Baker et al. (2016) using articles from the journals El Norte and Reforma.
Global EPU Index	GEPU	Constructed by Baker et al. (2016), the global EPU (GEPU) index re-normalizes national indices and imputes missing data to reflect GDP-weighted policy uncertainty across 21 countries, covering around 71% of global output.
EPU Index for Mexico	EPU_MEX	Baker et al. (2016) created a U.S. EPU index by analyzing the ten major U.S. newspapers: USA Today, Miami Herald, Chicago Tribune, Washington Post, Los Angeles Times, Boston Globe, San Francisco Chronicle, Dallas Morning News, Houston Chronicle, and Wall Street Journal.
EPU Index for Europe	EPU_EUR	Baker et al. (2016) developed a European EPU index based on five European countries: France (Le Monde, Le Figaro), Germany (Handelsblatt, Frankfurter Allgemeine Zeitung), Italy (Corriere Della Sera, La Stampa), Spain (El Mundo, El Pais), and the United Kingdom (The Times of London, Financial Times).
EPU Index for Japan	EPU_JPY	Baker et al. (2016) introduced a Japan EPU index, constructed by counting articles in four major Japanese newspapers: Yomiuri, Asahi, Mainichi, and Nikkei.

Source: compiled by the authors using data from Baker et al. (2016), Cerda et al. (2016), and Gil León and Silva Pinzón (2019).

shows the variables employed in this study, all available monthly, as shown in Table 1. The sample comprises a set of monthly series from January 2010 to May 2022, based on the availability of the EPU index for Colombia, resulting in 148 observations.

Additionally, the EPU index returns employed in this study, which are shown in Table 1, were developed by Baker et al. (2016), who constructed the indices for Brazil, Mexico, the United States, Europe, Japan, and global EPU indices by identifying uncertainty-related keywords in major national newspapers, thereby generating daily or monthly indices for these countries. Similarly, the EPU indices for Chile and Colombia were developed by Cerda et al. (2016) and Gil León and Silva Pinzón (2019), respectively, following the methodology developed by Baker et al. (2016).

Finally, it is important to note that the European economic policy uncertainty (EPU) index used in this study is a composite index that aggregates data from five major countries: France, Germany, Italy, Spain, and the United Kingdom. This approach reflects the methodology employed by Baker et al. (2016), which uses data from two prominent newspapers per country, standardized and averaged to create both country-level and region-wide indices. For the purposes of this study, the European EPU index represents an average of these five countries, rather than the entire EU-27 or geographical Europe. The decision to use Europe as a single block was driven by the availability of the composite index developed by Baker et al. (2016), as individual country-level EPU indices for all European nations are not available. This limitation necessitated the use of a regional index to ensure consistency and comparability in the analysis.

Figure 3 illustrates the dynamic behavior of EPU price and returns across the analyzed regions over time. Notably, Brazil and Mexico exhibit pronounced fluctuations, with prominent peaks around 2011 and 2016, indicating sharp increases in uncertainty likely tied to regional economic or political events. In contrast, Japan and Europe show lower amplitudes throughout the period, suggesting more stable policy environments. A particularly high peak was observed in Brazil in 2011, while Mexico experienced notable surges in 2016. Conversely, lower returns are observed across most regions around 2013 and 2020, reflecting periods of relative stability in policy uncertainty. These high-frequency oscillations highlight substantial variability across regions, with Latin American countries, especially Brazil and Mexico, showing more extreme movements. This variability underscores the heterogeneity in economic responses to policy uncertainty, where some regions experience episodic spikes while others maintain regular trends. We transformed the EPU index series into returns to ensure stationarity across all series, as shown in Figure 3b.

Table 2. Descriptive statistics of EPU index returns.

Statistic/Metric	REPU_BRA	REPU_CHI	REPU_COL	REPU_MEX	RGEPU	REPU_USA	REPU_EUR	REPU_JPY
MEAN	0.124 **	0.052 **	0.057 *	0.095 **	0.024	0.002	0.006	0.000
	(0.013)	(0.043)	(0.068)	(0.020)	(0.147)	(0.926)	(0.746)	(0.989)
VARIANCE	0.361	0.095	0.140	0.242	0.040	0.086	0.045	0.036
SKEWNESS	2.095 ***	0.899 ***	1.512 ***	1.010 ***	1.271 ***	0.043	0.291	−0.049
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.824)	(0.138)	(0.799)
EX. KURTOSIS	6.973 ***	1.505 ***	3.352 ***	1.167 **	3.070 ***	0.559	1.158 **	0.803 *
	(0.000)	(0.006)	(0.000)	(0.019)	(0.000)	(0.143)	(0.019)	(0.063)
JB	408.093 ***	33.894 ***	125.642 ***	33.554 ***	97.967 ***	1.974	10.362 ***	4.038
	(0.000)	(0.000)	(0.000)	(0.000)	(0.000)	(0.373)	(0.006)	(0.133)
ERS	−5.872	−1.469	−3.570	−1.563	−5.569	−3.690	−5.708	−6.530
	(0.000)	(0.144)	(0.000)	(0.120)	(0.000)	(0.000)	(0.000)	(0.000)
Q(20)	14.940	26.555 ***	26.529 ***	29.874 ***	18.166 **	22.567 ***	24.644 ***	18.620 **
	(0.124)	(0.001)	(0.001)	(0.000)	(0.037)	(0.006)	(0.002)	(0.031)
Q2(20)	18.669 **	4.569	10.249	12.757	17.690 **	8.116	11.543	6.538
	(0.030)	(0.971)	(0.478)	(0.249)	(0.045)	(0.713)	(0.350)	(0.863)
KENDALL CORRELATIONS
REPU_BRA	1.000 ***	0.099	0.119 **	0.029	0.254 ***	0.147 ***	0.138 **	0.056
REPU_CHI	0.099	1.000 ***	0.130 **	0.096	0.202 ***	0.197 ***	0.111 **	0.105
REPU_COL	0.119 **	0.130 **	1.000 ***	0.222 ***	0.266 ***	0.255 ***	0.206 ***	0.105
REPU_MEX	0.029	0.096	0.222 ***	1.000 ***	0.170 ***	0.139 **	0.140 **	0.126 **
RGEPU	0.254 ***	0.202 ***	0.266 ***	0.170 ***	1.000 ***	0.579 ***	0.509 ***	0.235 ***
REPU_USA	0.147 ***	0.197 ***	0.255 ***	0.139 **	0.579 ***	1.000 ***	0.328 ***	0.153 ***
REPU_EUR	0.138 **	0.111 **	0.206 ***	0.140 **	0.509 ***	0.328 ***	1.000 ***	0.178 ***
REPU_JPY	0.056	0.105	0.105	0.126 **	0.235 ***	0.153 ***	0.178 ***	1.000 ***

Source: authors’ own research. Notes: * significant at the 10% level; ** significant at the 5% level; *** significant at the 1% level.

presents the descriptive statistics of the analyzed return indices. The transformation of the EPU index series into returns, calculated as the natural logarithmic difference in consecutive monthly values, ensures stationarity by stabilizing variance and removing trends inherent in the original index values. This transformation is essential for the validity of econometric models such as the Time-Varying Parameter Vector Autoregressive (TVP-VAR) model (Gabauer & Gupta, 2018, 2020). Without this transformation, the presence of non-stationary data could lead to biased estimates and unreliable statistical inferences, violating the assumptions of the TVP-VAR model and potentially resulting in issues such as spurious relationships or overestimated spillover effects.

The analysis of EPU return indices across eight regions—Brazil, Chile, Colombia, Mexico, the global economy, the United States, Europe, and Japan—reveals distinct distributional characteristics and interdependencies. For example, Brazil and Mexico exhibit the highest mean values and variance, indicating significant and frequent fluctuations in policy uncertainty, while Japan and Europe show much lower levels, suggesting more stable policy environments. Positive skewness and high kurtosis in Latin America, especially Brazil, point to episodic surges in uncertainty, likely driven by political or economic instability. Additionally, the Jarque–Bera test results confirm that these distributions deviate from normality in both Latin America and the global index, reinforcing the perception of volatility in these markets.

Kendall correlation values indicate strong interdependencies between the global EPU indices and major economies like the U.S., Europe, and Japan. This implies that changes in policy uncertainty in the U.S. and Europe could have far-reaching effects, influencing global trends and investor sentiment. While Latin American indices are moderately correlated with each other and with the global index, their weaker ties to Japan suggest some degree of regional independence. These correlations underscore the potential for policy shocks to propagate across regions, particularly among highly interconnected economies, necessitating thoughtful policy coordination and consideration of cross-border effects.

The ERS (Elliott–Rothenberg–Stock) unit root test, Q(20) (autocorrelation within the series up to 20 lags) test, and Q2(20) (autocorrelation in the squared series up to 20 lags) test add further insight into the persistence, predictability, and volatility clustering of EPU index returns. Significant ERS values for Brazil, Colombia, the U.S., and other regions indicate stationarity, suggesting that policy uncertainty shocks tend to be short-lived. However, non-significant ERS values for Chile and Mexico imply persistent uncertainty shocks. Additionally, significant Q(20) values in Chile, Colombia, Mexico, the U.S., and Europe reveal short-term autocorrelation, suggesting a predictable pattern of policy uncertainty over time, while Brazil’s non-significant Q(20) reflects greater unpredictability. Q2(20) results highlight volatility clustering in Brazil, Chile, and the global index, indicating prolonged market reactions to policy changes in these areas. Thus, the findings shown in Table 2 suggest a complex and regionally distinct landscape of economic policy uncertainty. Emerging markets like Latin America may require adaptive policies to manage heightened volatility, while stable environments in developed economies enhance resilience to global shocks. Understanding these regional dynamics and correlations is essential for policymakers and investors to navigate and mitigate the interconnected risks in a globally integrated financial landscape.

This study leverages monthly data to monitor uncertainty spillovers effectively, providing a balanced perspective on fluctuations and enabling consistent tracking of responses across Latin American markets. Monthly data are well suited for capturing regional economic trends and allow for a more nuanced analysis of longer-term spillovers. This data frequency is particularly beneficial for Latin American policymakers, as it minimizes the noise of daily fluctuations and supports more strategic decision making to stabilize domestic economies in response to external or internal shocks. This approach is especially relevant for Latin America, where interconnected trade relationships and reliance on external markets make the region highly sensitive to sustained economic uncertainties.

Finally, Figure 4 depicts a structured flowchart summarizing the research process into five interconnected sections.

Figure 4 begins with the introduction of research context, which outlines the interconnectedness of economic policy uncertainty (EPU) across Latin American and global economies and the theoretical background underpinning spillovers and dependencies. Next, it presents the research hypotheses, focusing on external versus internal EPU influences, the roles of key transmitters like Brazil and Mexico, and the importance of institutional frameworks and regional cooperation. The methodological framework follows, detailing the use of monthly EPU indices (2010–2022) and a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model to analyze spillover dynamics. The empirical steps section highlights the procedural elements, including data preprocessing, estimation of model coefficients, and the analysis of various spillover types (directional, internal, external, and pairwise). Lastly, the findings and interpretations section emphasizes key insights, including Latin America’s dependence on external uncertainty, asymmetrical spillover structures, and policy recommendations for enhancing resilience. This diagram provides a clear and comprehensive outline of this study’s objectives and methods.

3.2. Time-Varying Parameter Vector Autoregressive (TVP-VAR) Model

This study investigates the mechanisms driving the spillover effects of economic uncertainty at both national and international levels through a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model. Initially developed by Antonakakis et al. (2020) and later applied to analyze uncertainty spillovers among developed economies by Antonakakis et al. (2018), this approach extends the connectedness framework introduced by Diebold and Yilmaz (2009) and Diebold and Yilmaz (2012). In this analysis, national-level spillovers encompass Latin American countries—Brazil, Chile, Colombia, and Mexico—while international-level spillovers include the United States, Europe, Japan, and the global EPU index. By incorporating stochastic volatility through a Kalman filter with forgetting factors, the TVP-VAR model enables a dynamic assessment of spillover effects as they evolve over time.

The TVP-VAR model is expressed as follows:

$${\mathit{y}}_t={\mathit{\beta }}_t{\mathit{z}}_{t-1}+{\mathit{ϵ}}_t{\mathit{ϵ}}_t|{\mathit{F}}_{t-1}\sim N\left(\mathbf{0},{\mathit{S}}_t\right)$$

(1)

$$vec\left({\mathit{\beta }}_t\right)=vec\left({\mathit{\beta }}_{t-1}\right)+{\mathit{\nu }}_t{\mathit{\nu }}_t|{\mathit{F}}_{t-1}\sim N\left(\mathbf{0},{\mathit{R}}_t\right)$$

(2)

where ${\mathit{y}}_t$ and ${\mathit{z}}_{t-1}={\left[{\mathit{y}}_{t-1},\dots ,{\mathit{y}}_{t-p}\right]}^{\prime }$ denote vectors of dimensions $N\times 1$ and $Np\times 1$, respectively, while ${\beta }_t$ is the $N\times Np$ matrix of time-varying coefficients. The term ${\mathit{ϵ}}_t$ represents an error vector with an $N\times 1$ dimensional error disturbance vector with an $N\times N$ time-varying covariance matrix ${\mathit{S}}_t$. Similarly, $vec\left({\mathit{\beta }}_t\right)$, $vec\left({\mathit{\beta }}_{t-1}\right)$, and ${\mathit{\nu }}_t$ are vectors of dimensions $N^{2}p\times 1$ with ${\mathit{R}}_t$ as an $N^{2}p\times N^{2}p$ matrix.

To compute the generalized impulse response functions (GIRFs) and generalized forecast error variance decomposition (GFEVD), following Koop et al. (1996) and Pesaran and Shin (1998), the model is transformed into a vector moving average (VMA) representation:

$${\mathit{y}}_t={{\displaystyle \sum }}_{j=0}^{\infty }{\mathit{L}}^{\prime }{\mathit{W}}_t^j\mathit{L}{\mathit{ϵ}}_{t-j}$$

(3)

$${\mathit{y}}_t={{\displaystyle \sum }}_{j=0}^{\infty }{\mathit{A}}_{it}{\mathit{ϵ}}_{t-j}$$

(4)

where $\mathit{L}={\left[{\mathit{I}}_N,\dots ,{\mathbf{0}}_p\right]}^{\prime }$ is an $Np\times N$ matrix, $\mathit{W}=\left[{\mathit{\beta }}_t;{\mathit{I}}_{N\left(p-1\right)},{\mathbf{0}}_{N\left(p-1\right)\times N}\right]$ has an $Np\times Np$ dimensional matrix, and ${\mathit{A}}_{it}$ is an $N\times N$ matrix. The GIRFs measure the response of each variable to a shock in a specific variable $i$, defined as follows:

$$GIR{F}_t\left(J,{\mathit{\delta }}_{j,t},{\mathit{F}}_{t-1}\right)=E({\mathit{Y}}_{t+J}|{\mathit{ϵ}}_{j,t}={\mathit{\delta }}_{j,t},{\mathit{F}}_{t-1})-E({\mathit{Y}}_{t+J}|{\mathit{F}}_{t-1})$$

(5)

$${\mathit{\Psi }}_{j,t}^g\left(J\right)={\displaystyle \frac{{\mathit{A}}_{J,t}{\mathit{S}}_t{\mathit{ϵ}}_{j,t}}{\sqrt{S_{jj,t}}}}{\displaystyle \frac{{\mathit{\delta }}_{j,t}}{\sqrt{S_{jj,t}}}}{\mathit{\delta }}_{j,t}=\sqrt{S_{jj,t}}$$

(6)

$${\mathit{\Psi }}_{j,t}^g\left(J\right)=S_{jj,t}^{-{\displaystyle \frac{1}{2}}}{\mathit{A}}_{J,t}{\mathit{S}}_t{\mathit{ϵ}}_{j,t}$$

(7)

where ${\mathit{\Psi }}_{j,t}^g\left(J\right)$ represents the GIRF of variable $j$, $J$ is the forecast horizon, ${\mathit{\delta }}_{j,t}$ is a selection vector, and ${\mathit{F}}_{t-1}$ is the information set up to $t-1$.

The GFEVD quantifies the share of variance in one variable attributable to others:

$${\tilde{\varphi }}_{ij,t}^g\left(J\right)={\displaystyle \frac{{{\displaystyle \sum }}_{t=1}^{J-1}{\Psi }_{ij,t}^{2,g}}{{{\displaystyle \sum }}_{j=1}^N{{\displaystyle \sum }}_{t=1}^{J-1}{\Psi }_{ij,t}^{2,g}}}$$

(8)

satisfying ${\displaystyle \sum }_{j=1}^N{\tilde{\varphi }}_{ij,t}^g\left(J\right)=1$ and ${\displaystyle \sum }_{i,j=1}^N{\tilde{\varphi }}_{ij,t}^N\left(J\right)=N$. Using the GFEVD, we derive the total connectedness index (TCI) as follows:

$$C_t^g\left(J\right)={\displaystyle \frac{{{\displaystyle \sum }}_{i,j=1,i\ne j}^N{\tilde{\varphi }}_{ij,t}^g\left(J\right)}{{{\displaystyle \sum }}_{i,j=1}^N{\tilde{\varphi }}_{ij,t}^g\left(J\right)}}\times 100$$

(9)

$$C_t^g\left(J\right)={\displaystyle \frac{{{\displaystyle \sum }}_{i,j=1,i\ne j}^N{\tilde{\varphi }}_{ij,t}^g\left(J\right)}{N}}\times 100$$

(10)

This connectedness framework illustrates how a disturbance in one variable extends its effects to other variables. Initially, we examine the scenario where variable $i$ transmits its impact across all other variables $j$, referred to as the total directional connectedness to others, defined as follows:

$$C_{i\to j,t}^g\left(J\right)={\displaystyle \frac{{{\displaystyle \sum }}_{j=1,i\ne j}^N{\tilde{\varphi }}_{ji,t}^g\left(J\right)}{{{\displaystyle \sum }}_{j=1}^N{\tilde{\varphi }}_{ji,t}^g\left(J\right)}}\times 100$$

(11)

Similarly, the directional connectedness variable $i$ receives it from the variable $j$, called the total directional connectedness from others, as follows:

$$C_{i\leftarrow j,t}^g\left(J\right)={\displaystyle \frac{{{\displaystyle \sum }}_{j=1,i\ne j}^N{\tilde{\varphi }}_{ij,t}^g\left(J\right)}{{{\displaystyle \sum }}_{i=1}^N{\tilde{\varphi }}_{ij,t}^g\left(J\right)}}\times 100$$

(12)

The net total directional connectedness for each variable is defined as follows:

$$C_{i,t}^g=C_{i\to j,t}^g\left(J\right)-C_{i\leftarrow j,t}^g\left(J\right)$$

(13)

where ($C_{i,t}^g>0$) suggests that $i$ is a network driver, while ($C_{i,t}^g<0$) suggests $i$ is driven by the network. To analyze bidirectional relationships, we compute the net pairwise directional connectedness (NPDC):

$$NPD{C}_{ij}\left(J\right)={\displaystyle \frac{{\tilde{\varphi }}_{ji,t}^g\left(J\right)-{\tilde{\varphi }}_{ij,t}^g\left(J\right)}{T}}\times 100$$

(14)

This framework thus provides a comprehensive view of how economic shocks in one variable propagate across others, highlighting the dynamics of economic uncertainty spillovers.

This study examines the dynamic interconnectedness of EPU index returns across key Latin American economies—Brazil, Chile, Colombia, and Mexico—as well as major international regions, including the United States, Europe, Japan, and a global EPU index. This study applies a TVP-VAR model using monthly data to capture the evolving spillover effects and dependencies among these regions. This model is particularly suited to analyzing how economic uncertainty in one market can spread to others across both regional and global scales.

4. Empirical Results

4.1. General Connectedness Findings

Table 3. Analysis of cross-regional spillover effects and net transmissions in EPU index returns.

	REPU_BRA	REPU_CHI	REPU_COL	REPU_MEX	RGEPU	REPU_USA	REPU_EUR	REPU_JPY	FROM
REPU_BRA	80.47	0.00	0.00	0.00	0.00	3.54	0.85	1.64	6.03
REPU_CHI	0.00	61.31	4.51	2.30	10.81	5.48	6.18	6.43	35.72
REPU_COL	0.00	3.36	51.37	10.17	9.87	12.39	7.29	4.64	47.72
REPU_MEX	0.00	2.97	11.78	60.59	4.74	7.39	4.02	5.96	36.86
RGEPU	0.00	6.10	7.38	3.26	33.84	19.85	17.13	10.40	64.11
REPU_USA	1.38	3.45	10.54	3.52	25.21	40.64	10.21	5.05	59.36
REPU_EUR	0.41	4.49	6.48	3.16	22.41	10.22	43.35	9.49	56.65
REPU_JPY	0.92	7.48	5.55	7.23	12.10	5.67	8.13	52.93	47.07
TO	2.71	27.84	46.24	29.63	85.15	64.54	53.81	43.60	353.51
Inc.Own	83.17	89.15	97.61	90.22	118.98	105.18	97.15	96.54	cTCI/TCI
NET	−3.32	−7.87	−1.48	−7.23	21.03	5.18	−2.85	−3.46	50.50/44.19
NPT	0.00	2.00	3.00	1.00	6.00	6.00	4.00	2.00

Source: authors’ own research.

and Figure 5 illustrate the interconnectedness between EPU index returns across various regions, including Brazil, Chile, Colombia, Mexico, a global EPU index (RGEPU), the United States, Europe, and Japan. This table shows both the influence of each region’s EPU on others (as percentages in the “TO” row) and the influence received from others (in the “FROM” column).

Each region has a substantial share of its own uncertainty (indicated along the diagonal), such as Brazil, with 80.47%, or Japan, with 52.93%. However, some regions, notably the global EPU and the United States, have strong spillover effects, impacting and being impacted by multiple other regions. For example, the global EPU index contributes significantly to other regions (85.15% in the “TO” row) while also receiving influences from them (64.11% in the “FROM” column), indicating a high degree of interconnectedness.

The “NET” row provides insights into whether each region is a net transmitter or receiver of uncertainty. The positive net values for RGEPU and the USA indicate that these regions are net transmitters of uncertainty, while the negative values for regions like Chile and Mexico show they are net receivers.

Figure 5 illustrates the directional spillovers of EPU index returns among global and Latin American economies, highlighting the intricate interconnections. The arrows in the figure indicate the direction of influence, while their thickness represents the strength of these relationships. The global EPU (RGEPU) and the United States EPU (REPU_USA) emerge as central nodes with thick, multidirectional arrows, underscoring their roles as significant transmitters of EPU to other regions, particularly Latin America. This confirms the findings where RGEPU and REPU_USA are net transmitters of uncertainty.

Latin American economies—Brazil, Mexico, Chile, and Colombia—are shown as primary recipients of EPU index return spillovers, with relatively thinner arrows connecting them to other regions. This pattern indicates that Latin American countries are more vulnerable to external uncertainties than they are influential as transmitters of their own EPU. The influence of the U.S. and global EPU on these economies is substantial, emphasizing Latin America’s dependency on external economic conditions. Inter-regional spillovers within Latin America appear weaker, suggesting a limited capacity for these economies to impact each other’s stability.

The findings underscore an asymmetrical spillover structure, where global and U.S.-based uncertainties significantly influence Latin America while the reverse impact remains minimal. This interconnectedness highlights the need for Latin American policymakers to develop resilience strategies to mitigate the region’s susceptibility to external shocks, particularly from dominant global economies. Regional cooperation could play a key role in bolstering economic stability in the face of ongoing global uncertainties.

4.2. Internal and External Connectedness Findings

The internal connectedness calculations were conducted by grouping EPU index returns across key Latin American economies—Brazil, Chile, Colombia, and Mexico. These regional indices were analyzed in isolation to assess their internal spillovers and interdependencies. In contrast, the external connectedness calculations included a broader set of international regions, comprising the United States, Europe, Japan, and a global EPU index. This division allowed for a focused analysis of spillovers within Latin America (internal) versus those influenced by external global actors, providing a comprehensive view of both intra-regional and cross-regional economic policy uncertainty dynamics.

4.2.1. Dynamic Total Connectedness Index (TCI)

Figure 6 provides a dynamic view of the total connectedness index (TCI) over time, capturing various layers of economic policy uncertainty spillovers across global and Latin American economies. The black shaded area, representing the total TCI, reflects the interconnectedness that combines internal and external spillovers. Consistently high TCI values underscore a persistent interdependence in economic uncertainty across markets, with notable peaks aligning with global economic crises or regional disturbances. This confirms the observations from Figure 5 and Table 3, highlighting significant global and U.S.-based influences.

The red line in Figure 6 indicates the TCI when focusing solely on internal spillovers within specific regions or entities, revealing consistently lower levels than the total TCI. This gap suggests that external influences dominate uncertainty spillovers while regional connections are important. This supports the findings in Figure 5 and Table 3, where Latin American economies appear more as recipients than transmitters of uncertainty, with a stronger reliance on external economic dynamics.

The green line, representing global connectedness, highlights global economic uncertainty’s stable influence on regional and local markets. This consistency reflects the ongoing impact of international economic policy changes and global events on Latin American stability, as illustrated in Figure 5.

Finally, the blue line, which captures connectedness within Latin America, remains the lowest among the indices. This suggests that while Latin American economies have some degree of interconnectedness, they are primarily shaped by global influences rather than by each other. The relatively limited intra-regional connectedness indicates that external shocks are a more significant source of uncertainty than regional linkages.

Overall, Figure 6 reinforces the layered nature of economic policy uncertainty spillovers, confirming that Latin American economies are more vulnerable to global and U.S.-driven uncertainties than internal fluctuations. This pattern underscores the need for Latin American policymakers to focus on resilience strategies that can mitigate the impact of external economic shocks, rather than relying solely on intra-regional cooperation.

4.2.2. Net Total Directional Connectedness

Figure 7 provides a detailed view of net directional connectedness by separating external and internal spillovers for each region, adding depth to the insights from the TCI, which confirms the results obtained in Figure 5 and Figure 6 and Table 3. For Latin American economies—Brazil, Mexico, Chile, and Colombia—external spillovers consistently surpass internal ones, reflecting their strong dependence on global economic conditions. While Brazil and Mexico show some variability in their influence within Latin America, they remain predominantly influenced by external factors, reinforcing findings in earlier figures and tables.

Developed regions like Europe and Japan exhibit high, stable external connectedness, affirming their role as major sources of global EPU spillovers, particularly Europe’s far-reaching impact. The United States stands out with one of the highest levels of external connectedness, underscoring its role as a key transmitter of economic uncertainty, especially affecting Latin American economies like Mexico.

The global EPU (RGEPU) subplot shows continuous external connectedness, illustrating how global events consistently influence individual markets, echoing the interconnected nature highlighted in the TCI, as shown in Figure 6. Over time, external spillovers dominate across regions, particularly affecting Latin America, which shows limited internal resilience. This pattern underscores Latin America’s vulnerability to uncertainties from advanced economies, emphasizing the need for resilience strategies to mitigate these impacts.

4.2.3. Total Directional Connectedness Received from External and Internal Spillovers

Figure 8 highlights the impact of external and internal EPU index return spillovers on various regions, building on insights from Table 3 and Figure 5, Figure 6 and Figure 7. Latin American economies—Brazil, Mexico, Chile, and Colombia—show consistently high levels of external spillovers, underscoring their reliance on global economic conditions. Internal spillovers remain minimal, confirming these economies’ limited resilience against external influences, with Mexico’s stability closely tied to U.S. policy uncertainties.

Developed regions like Europe, Japan, and the United States show stable external spillover patterns. However, the U.S. stands out with a stronger internal spillover influence, indicating that domestic factors play a larger role in shaping its connectedness, reinforcing its central position in global economic stability.

The global EPU (RGEPU) remains steadily impacted by external uncertainties, confirming the pervasive effect of global economic shocks. Figure 8 reaffirms that external spillovers generally exceed internal ones worldwide, suggesting that global events like trade tensions and the pandemic heighten interdependence across regions.

Then, Figure 8 illustrates the dominant role of external spillovers in Latin America’s economic stability and highlights the resilience of developed economies against global shocks. These findings underscore the need for Latin American countries to strengthen resilience strategies to mitigate the effects of global economic fluctuations.

4.2.4. Total Directional Connectedness Transmitted to Other Economies: External and Internal Spillovers

Figure 9 provides insights into the total directional connectedness transmitted by each economy, differentiating between external and internal spillovers. This figure reinforces previous findings from Table 3 and Figure 5, Figure 6, Figure 7 and Figure 8, highlighting a consistent pattern where Latin American economies—Brazil, Mexico, Chile, and Colombia—exhibit higher external than internal spillovers. While these countries do influence their regional counterparts, their connectedness is primarily shaped by external factors, confirming Latin America’s limited role as a transmitter of economic uncertainty. Mexico, in particular, shows stable levels of external spillovers, reflecting its economic integration with the U.S. and North American trade networks, a finding previously observed in Figure 8.

Developed economies, including the U.S., Europe, and Japan, continue to demonstrate substantial and stable external spillovers, confirming their significant influence on global markets. The U.S., in particular, shows the highest level of external spillovers, underscoring its pivotal role in the global financial system and strong economic ties, especially with Latin America. This finding aligns with the observations in Figure 5 and Figure 7, where the U.S. and global EPU appeared as central drivers of economic policy uncertainty. Europe’s steady external spillovers also reinforce its substantial impact on global economic stability, consistent with results across previous analyses.

The dominance of external spillovers across all regions in Figure 9 reflects the high degree of interdependence among global markets. Latin American economies are more impacted by external uncertainties than they contribute, underscoring their vulnerability to global shifts, especially those originating from the U.S. and global EPU. This asymmetry emphasizes the importance of Latin American countries developing resilience strategies that reduce dependency on external economic conditions, an observation also supported by Figure 6 and Figure 8. Additionally, it highlights the leading role of developed economies in shaping global economic policy uncertainty, affirming the central influence of the U.S. and global EPU in driving interconnectedness within the global financial system.

4.2.5. Internal Net Pairwise Total Directional Connectedness Among Economies

Figure 10 offers a focused analysis of internal net pairwise total directional connectedness, showing how EPU index returns flow between individual economies over time. Each subplot illustrates a unique pairwise relationship, highlighting directional influence through periods of net positive connectedness (black shaded areas for outward influence) and net negative connectedness (red lines for inward influence). This detailed view complements the broader trends in previous figures by capturing the dynamic, bilateral interactions between specific economies.

Brazil and Mexico stand out in Latin America with stronger outgoing connectedness toward other Latin American countries, reflecting their central economic roles within the region. For example, Brazil’s influence on Colombia and Mexico’s on Chile reveal how EPU spillovers propagate regionally, although the intensity of these connections varies over time. Despite their regional impact, Latin American economies generally exhibit weaker pairwise connections with developed economies, indicating that internal dynamics within Latin America are more pronounced than their direct influence from or on global players.

For developed economies, such as the U.S., Europe, and Japan, Figure 10 reveals more consistent bidirectional spillovers. The U.S., in particular, shows outgoing solid influence on Latin American economies like Mexico and Brazil, reinforcing its role as a significant source of EPU for the region. The relationship between the U.S. and Europe displays stable, high levels of mutual influence, reflecting a tightly interconnected dynamic between these major economies. While influential globally, Japan shows lower connectedness with Latin American countries, underscoring its relatively limited direct economic ties to the region.

In conclusion, Figure 10 highlights the layered structure of EPU spillovers, where regional dynamics within Latin America coexist alongside substantial influence from major global economies. Brazil and Mexico emerge as regional transmitters within Latin America, while the U.S. and global EPU play dominant roles in the global network, shaping broader EPU trends. This interconnectedness underscores the need for Latin American economies to actively monitor both regional and global sources of uncertainty to strengthen economic stability.

4.2.6. External Net Pairwise Total Directional Connectedness Among Economies

Figure 11 provides a detailed view of external net pairwise total directional connectedness, illustrating how EPU index returns flow between individual countries or regions. Each subplot captures a specific pairwise relationship, distinguishing between periods of net positive (outward influence) and net negative (inward influence) connectedness. This figure enhances the understanding of bilateral EPU spillovers, revealing which economies act as primary sources or recipients of uncertainty over time, complementing the broader trends observed in previous figures.

In Latin America, Brazil and Mexico consistently demonstrate outgoing influence on other Latin American countries, such as Chile and Colombia, positioning them as key transmitters of policy uncertainty within the region. This role aligns with the findings in Figure 10, where Brazil and Mexico emerged as central players in regional spillovers. However, their interactions with developed regions—particularly the U.S. and global EPU—show these Latin American economies more often as recipients of EPU, highlighting their susceptibility to external shocks from larger economies. This trend confirms earlier observations in Table 3 and Figure 6 through Figure 8, which indicated a dependency of Latin American economies on global uncertainties.

The U.S. and global EPU in developed economies demonstrate high and stable outgoing spillover effects, reinforcing their positions as primary transmitters of policy uncertainty across regions. The U.S., with notable spillovers directed toward Latin American countries like Mexico and Brazil, underscores its influential role in shaping regional EPU dynamics. Europe’s relationships, including with Japan and global aggregates, further demonstrate its substantial impact on both developed and emerging markets. While influential globally, Japan shows comparatively weaker spillover effects on Latin America, suggesting limited direct transmission of its policy uncertainty to the region, consistent with findings in previous figures.

In summary, Figure 11 underscores the complex network of external EPU spillovers, where Brazil and Mexico are notable sources of regional policy uncertainty within Latin America, while the U.S. and global EPU act as major transmitters of EPU on a global scale. These bilateral relationships emphasize Latin America’s vulnerability to external shocks, especially from developed economies, underscoring the need for Latin American policymakers to account for both regional and global uncertainties when crafting strategies to enhance economic resilience and stability.

5. Discussion

Our analysis of EPU index return spillovers in Latin American and global economies aligns with findings from Antonakakis et al. (2018), Zhou et al. (2022), K.-H. Wang et al. (2023), and Kayani et al. (2024), who similarly explore EPU dynamics across developed and emerging economies. Each of these studies provides valuable context for understanding the cross-national transmission of EPU and highlights the varying roles that different economies play within this interconnected system.

By addressing RH1, Antonakakis et al.’s (2018) research on EPU spillovers primarily among developed economies identifies significant transmission from the European Union to the U.S., underscoring external factors’ influence on economic stability. Our findings parallel this dynamic, as Latin American countries show high sensitivity to external spillovers, particularly from the U.S. and global EPU indices. This pattern supports Antonakakis et al.’s observation that external drivers can substantially impact domestic uncertainty, especially in economies with strong ties to developed regions. Both studies highlight the importance of understanding these external influences, particularly over longer time horizons, where their impact becomes more pronounced. For Latin American policymakers, this insight underscores the need for strategic planning to buffer against sustained external uncertainty.

By answering RH2, Zhou et al.’s (2022) study expands on the global EPU network by using spillover indices and block models, finding that EPU spillovers increase significantly during extreme events like COVID-19. Similarly, our analysis shows that Latin American economies, particularly Brazil and Mexico, experience intensified spillovers from global sources during periods of global crisis. Zhou’s block model further categorizes countries into distinct spillover blocks, which highlights the importance of external EPU sources for more vulnerable economies. The findings suggest a similar structure in Latin America, with Brazil and Mexico acting as regional transmitters but remaining dependent on EPU from larger economies. This layered model reflects the need for resilience strategies to counteract external vulnerabilities.

By addressing RH3, the study conducted by Kayani et al. (2024) on BRIC countries reveals that global EPU, particularly from dominant economies, substantially impacts emerging markets, with Brazil showing the lowest gross EPU spillover within the BRIC group. This is consistent with our findings, where Brazil and Mexico function primarily as an EPU receiver within Latin America rather than as a significant transmitter. Kayani also highlights the diversity in directional spillovers among BRIC nations, which aligns with our observations in Latin America. For instance, Mexico’s close economic integration with the U.S. results in more stable external spillovers, while other Latin American countries exhibit different sensitivity levels. Both studies suggest that emerging markets must develop strategies to manage these spillover impacts, emphasizing international cooperation’s importance in addressing EPU vulnerabilities.

By answering RH4, K.-H. Wang et al.’s (2023) study on the relationship between China’s EPU and ASEAN countries’ geopolitical risk (GPR) highlights the influence of external shocks on regional stability. Similar to Wang’s findings, our study shows that Latin American economies are significantly affected by external EPU, particularly during global crises. Wang’s analysis of heterogeneous spillover effects also resonates with our findings. At the same time, external EPU impacts all Latin American economies, and the intensity varies based on economic linkages, as seen with Mexico’s stronger connection to the U.S. Furthermore, Wang suggests that enhancing institutional quality can reduce vulnerability to external EPU. This recommendation is also applicable in Latin America, where strengthening internal frameworks could help mitigate the influence of external uncertainties from dominant global players.

In conclusion, by addressing RH5, our study supports the findings of Antonakakis et al. (2018), Zhou et al. (2022), K.-H. Wang et al. (2023), and Kayani et al. (2024), highlighting the significant role of external EPU in shaping the stability of emerging markets. The interconnected nature of modern economies underscores the importance of resilience strategies, especially for Latin American countries that are heavily influenced by U.S. and global EPU dynamics. Strengthening institutional capacity, fostering regional cooperation, and adopting proactive policy measures are critical steps for Latin American policymakers to manage these external risks effectively. These collective insights reinforce the need for emerging economies to be equipped for increased volatility, particularly during global crises, to maintain economic stability in an interconnected world.

The findings from this study underscore the importance of recognizing interconnected economic uncertainties across Latin America, providing policymakers with insights into strategies for strengthening resilience against external shocks. Thus, by addressing RH4 and RH5, identifying and analyzing the spillover effects of the EPU index returns, Latin American economies can take proactive measures to limit their vulnerability to global fluctuations. Key strategies include trade diversification, which would reduce dependence on specific markets or commodities, and adapting monetary policy to manage exchange rate volatility and inflation in response to global economic shifts. Regional collaboration through frameworks like the Pacific Alliance and Mercosur could also significantly reduce shared risks, as aligned policy responses and cooperative economic integration could foster financial stability across borders. Additionally, Latin American monetary and financial authorities could benefit from developing joint contingency plans and pooling resources to respond to unexpected shocks, especially during periods of heightened global uncertainty.

Latin America’s economic landscape is shaped by unique sources of uncertainty, including high dependence on commodities, exposure to global financial markets, and political volatility. Heavy reliance on commodities means that fluctuations in global commodity prices can significantly impact these economies, heightening their vulnerability during market downturns. Exposure to global financial flows introduces additional risks, as sudden capital outflows or shifts in investor sentiment can lead to economic instability. Political risks, including policy changes tied to electoral cycles, add to this volatility, often increasing regional uncertainty. In contrast to developed economies with stronger institutional frameworks, Latin American countries face the challenge of addressing these uncertainties through region-specific policies emphasizing economic diversification, strengthened governance, and regulatory reforms designed to minimize external impacts. This aligns with RH4 and RH5, reinforcing the need for institutional and regional resilience.

6. Conclusions

This study applies a Time-Varying Parameter Vector Autoregressive (TVP-VAR) model with monthly data to examine the dynamic interconnectedness of economic policy uncertainty (EPU) across Latin American economies—Brazil, Mexico, Chile, and Colombia—as well as major international regions, including the United States, Europe, Japan, and a global EPU index.

The analysis of findings reveals that Latin American economies—Brazil, Mexico, Chile, and Colombia—are predominantly impacted by external EPU, rather than acting as primary sources. These economies highly depend on global factors, mainly influenced by the U.S. and the global EPU index (RGEPU), which aggregate uncertainties from multiple regions. This dependency makes Latin America especially vulnerable to shifts in international policy dynamics. Although Europe also contributes to external EPU spillovers, its influence is secondary compared to the U.S. and global EPU, answering RH1.

The internal versus external dynamics indicate that Brazil and Mexico, while serving as regional transmitters within Latin America, are primarily recipients of uncertainty regarding global interactions. This layered structure of spillovers underscores Latin America’s reliance on external economic conditions, highlighting the importance of resilience strategies that reduce susceptibility to global shocks. Strengthening internal economic frameworks and enhancing regional cooperation could better equip Latin America to mitigate the impacts of global EPU, addressing RH2.

In conclusion, Latin America’s reliance on major economies, particularly the U.S. and global economic trends, underscores the need for coordinated international efforts to manage policy uncertainty, fostering greater stability within interconnected global markets, answering RH3.

This study, while comprehensive, is subject to certain limitations. The analysis primarily depends on available monthly EPU data, which may not fully capture short-term volatility or specific sectoral spillovers that could provide deeper insights into economic linkages. Future research could benefit from exploring EPU spillovers at a sectoral level to understand how uncertainties in industries such as energy, manufacturing, or finance propagate across Latin American economies. Additionally, examining the role of fiscal policy, inflation targeting, and financial regulation could offer valuable perspectives on managing economic uncertainty, addressing RH4. Comparative studies with other emerging regions, such as Southeast Asia or Africa, may further enrich our understanding of the distinct and shared challenges faced by Latin America in an interconnected economic landscape.

This study provides a novel contribution by focusing on the application of a TVP-VAR model to Latin American economies, a region often underrepresented in the EPU literature. By integrating both regional and global indices, it offers a comprehensive view of spillover dynamics, revealing the asymmetrical impact of external versus internal uncertainties. This approach not only bridges a gap in existing research but also provides actionable insights tailored to the specific vulnerabilities and interdependencies of Latin America. The findings highlight the necessity for resilience strategies and international coordination, advancing the understanding of EPU in emerging markets, answering RH5.

The results of this study will be valuable for policymakers, economic advisors, and financial regulators, particularly in Latin America. Policymakers can use these insights to design strategies that strengthen their economies’ resilience against external shocks, such as diversifying trade partnerships and stabilizing monetary policy in response to international economic fluctuations. Economic advisors can benefit from understanding the spillover dynamics of EPU indices to forecast potential impacts on domestic markets better, allowing for more informed investment and risk management recommendations. Financial regulators, meanwhile, can use these findings to enhance their frameworks for financial stability, ensuring that regional financial systems are better prepared to handle volatility stemming from global economic shifts. Additionally, international organizations focused on economic stability and cooperation may find the results helpful in developing coordinated approaches to managing EPU spillovers across interconnected markets.

Future research could expand the scope of this study by applying similar TVP-VAR models to other emerging markets, enabling comparisons of EPU spillover effects across different regions. Examining the role of fiscal policies, such as stabilization mechanisms and inflation targeting, may offer further insights into managing economic uncertainty within Latin America. Research on structural reforms, including strengthening financial regulation, transparency improvements, and governance enhancements, could provide additional strategies for mitigating the impact of uncertainty on economic stability. Additionally, further studies might analyze sector-specific spillovers, assessing how uncertainty in sectors like energy or manufacturing propagates through Latin American economies. Investigating these factors would deepen the understanding of the unique challenges faced by Latin America and support the development of more targeted, effective policy solutions.