1. Introduction

The initial response to the COVID-19 pandemic compelled states worldwide to implement public health measures such as social isolation, mask mandates, and lockdown orders. Some of these included more unusual restrictions, such as flight bans and, ultimately, airport closures. In addition to public health interventions, various economic policies were enacted globally to address the pandemic's fallout. These policies encompassed a diverse array of measures, including cash transfers, childcare support, social pensions, in-kind food vouchers, school feeding programs, public employment initiatives, paid leave, social security contributions, wage subsidies, and labor regulations (Gentilini et al., 2020). This type of response is often interpreted as indicative of politicians' commitment to social welfare and equity during uncertain times (Thomas, 1998). Similarly, the implementation of such emergency policies has the potential to enhance political trust in public institutions. While initially conceptualized as diffuse support independent of government performance (Easton, 1975), political trust has rapidly turned to a requirement for effective governmental action. The emphasis on trust during the COVID-19 pandemic is underscored by recent evidence indicating that higher levels of public trust lead to greater compliance with public health measures, including social isolation, mobility restrictions, mask usage, and vaccination (Ayalon, 2021; Kreps & Kriner, 2020; Mallapaty & Callaway, 2021).

In El Salvador, emergency economic measures were enacted as early as March 2020, including restrictions to non-essential jobs, a freeze on layoffs, a moratorium on basic service payments, and a $300 monetary transfer to 1.5 million households (Ayala Durán, 2021b). According to the Salvadoran executive branch, this unconditional transfer was projected to benefit 75% of households nationwide, with an estimated total cost of $450 million (Lemus, 2020). This initiative was intended to support vulnerable households already enrolled in the national subsidy registry (Ayala Durán, 2021b; Martínez & Sánchez, 2022).

While the emergency transfer may have yielded positive effects, Ayala Durán (2021b) identified several critical shortcomings, particularly associated with its design, allocation, implementation, and evaluation. Some of these flaws were corroborated by the Salvadoran Court of Audits, which pointed to the existence of lax controls and unverified beneficiaries (Campos, 2020). Additionally, other emergency policies enacted during the pandemic in El Salvador were later marred by political clientelism and potential corruption (Ayala Durán, 2021b; Labrador, 2021; Palifka & Balán, 2024). Even when this unconditional transfer exhibited significant shortcomings, public support for president Bukele’s handling of the pandemic remained robust. Studies indicated a positive citizen appraisal of both the president and the health minister’s performance (Ayala Durán, 2021c, 2022a; Ayala Durán & Mendez, 2021; Centro de Estudios Ciudadanos, 2020).

The literature on emergency support during COVID-19 frequently assesses whether public interventions yield higher levels of trust (Ayala Durán, 2021c, 2022a; Ayala Durán & Mendez, 2021; Ayalon, 2021; Gozgor, 2021; Poma & Pistoresi, 2024). However, studies—either explicitly (Ayala Durán, 2021c; Sosa-Villagarcia & Hurtado, 2021) or implicitly (Ayala Durán, 2022a; Gozgor, 2021; Poma & Pistoresi, 2024), tend to focus on a single branch of government, notably the executive. This focus can be problematic, as health provision involves shared responsibilities among various levels of government within countries (Rodrigues et al., 2021; Smith, 2020; Woody, 2005) and government branches globally have often been in conflict when it comes to the pandemic management (Arévalo, 2021; Jung & Lee, 2023; Ringe et al., 2022; Smith, 2020; Zulianello, 2023).

While trust in various institutions has been widely examined, only few studies have explicitly focused on different levels of public administration. Liu et al. (2022) compared trust in the president, state authorities, and health experts, while Poma and Pistoresi (2024) centered on trust in the police, the European Union, and the healthcare system. However, to the authors’ knowledge, the following text represents a novel attempt to measure trust in parliament, particularly in comparison to presidential trust within the pandemic context.

Including legislative and executive branches might be particularly relevant in El Salvador, where president Bukele emerged as the main and polarizing figure leading the pandemic management despite initially having minimal parliamentary support following his 2019 election, holding only 11.9% of parliamentary seats. This dynamic remained unchanged until 2021, when his coalition secured a supermajority of 70% of parliamentary seats (among others, see Nilsson, 2022; Zemmouche, 2021). Adopting this broader approach helps move beyond the narrow tendency to equate government solely with the executive branch, thereby mitigating any misleading interpretations of trust in public branches.

In summary, looking to address the gap between public trust, COVID-19 emergency policies, and government branches, the present study examines the impact of the monetary transfer on public trust in two Salvadoran government branches: the presidency and the parliament. To achieve this, a probabilistic sample collected during the early pandemic (April 2020) is used. Considering that the transfer was not implemented using random sampling but rather allegedly targeted the most vulnerable households, a matching quasi-experimental design is employed (Mahalanobis nearest neighbor matching). The central hypothesis guiding this text is that economic measures and public support generally lead to increased trust in political institutions. Consequently, higher levels of trust would be expected following the receipt of monetary transfers in both government branches. This is supported by recent evidence on different scales of government action (Liu et al., 2022; Poma & Pistoresi, 2024).

2. Political trust: a traditional framework increasingly used in the COVID-19 pandemic

Trust constitutes a well stablished research area in political science and public administration. For most of the twentieth century, seminal work by David Easton (Easton, 1965, 1975, 1976) envisioned trust as a form of diffuse support, namely a “reserve of support that enables a system to weather the many storms when outputs cannot be balanced against inputs of demands” (Easton, 1965, p. 273). Broadly speaking, it would entail a backup legitimation mechanism when results are not delivered by the authorities, regime or community (Easton, 1965, p. 274). This approach was followed by a myriad of empirical studies, particularly in western democracies as the United States and the United Kingdom (Citrin, 1974; Devine, 2024). As the number of democratic regimes increased throughout the late twentieth century, studies on public trust began to proliferate in the developing world as well. This expansion was linked to the notion that higher levels of public trust were associated with greater support for democratic governance (Morales, 2008).

Based on this framework, development scholars and international organizations have carried out applied research on public trust in the developing world and some studies have moved beyond the original disconnection between performance and trust, transitioning towards an apparent relationship between the two. This shift is plausible, considering that confidence is further theorized as a cognitive, rational behavior that is informed by past experiences, individual assessments, and available information (Güemes & Brugué, 2017; Organisation for Economic Co-operation and Development, 2014). In Latin America, Criado and Herreros (2007) observed that citizens are more likely to trust institutions that effectively deliver results. Similarly, Segovia (2006) conceptualizes political trust as a specific type of support wherein citizens believe that institutions will act appropriately and achieve outcomes, even in the face of limited oversight.

Previous evidence indicate that higher levels of trust imply specific behaviors by citizens, including greater voluntary compliance with the law (Marien & Hooghe, 2011; Scholz & Lubell, 1998). Such behaviors are fundamental in crisis situations, such as natural disasters or public health emergencies (Organisation for Economic Co-operation and Development, 2013). In the context of the growing uncertainty derived from the COVID-19 emergency, public trust has shifted from traditional domains of social science to become an essential asset in managing the health crisis. Higher levels of trust, along with corresponding voluntary compliance with laws, facilitate adherence to various health recommendations, including physical distancing, home isolation, regular use of medical devices, and vaccination against the virus. In this regard, Lucia, Kelekar and Afonso (2020) demonstrate that trust in public health experts can be decisive for communities to adopt specific health measures. Similarly, multi-country empirical evidence highlights the importance of public trust in promoting voluntary COVID-19 vaccination (Gozgor, 2021; Vergara et al., 2021). Conversely, Mallapaty and Callaway (2021) identified the relation between distrust and the lack of transparency in the development of the AstraZeneca vaccine in Brazil, South Africa, and the United States.

Part of the global COVID-19 response included economic measures and stimuli to offset the consequences of the public health emergency (Gentilini et al., 2020). These policies encompassed a wide array of initiatives, including cash transfers, childcare support, social pensions, in-kind food vouchers, school feeding programs, public employment, paid leave, social security contributions, wage subsidies, and labor regulations (Gentilini et al., 2020). Broadly speaking, these initiatives aimed to enable citizens to comply with physical mobility restrictions while ensuring adequate resources for household consumption (Ravallion, 2020; Robalino, 2020). Evidence suggests that economic policies and benefits designed to mitigate the adverse effects of the public health emergency can also enhance trust by demonstrating politicians’ commitment to social welfare and equality (Thomas, 1998). Recent findings on COVID-19 and trust support this rationale. For instance, Poma and Pistoresi (2024) highlight that government support is positively associated with institutional trust in European countries. Their study also finds that economic distress diminishes citizens’ perception of government support (Poma & Pistoresi, 2024, p. 371). Rieger and Wang (2022) report similar results, noting that perceived insufficient responses to the COVID-19 pandemic consistently reduce government trust. To the same extent, Liu et al. (2022) analyze data from 178 countries and report that public action, particularly economic support measures, yields positive results in government trust. Jørgensen et al. (2021) indicate that interpersonal trust is a strong predictor of support for governmental responses in eight Western democracies. These individual studies seem to have correspondence with the systematic review by Devine (2024), supporting the assertion that trust has an overall moderate effect on COVID-19 related behaviors and attitudes.

Trust is particularly relevant in the context of healthcare provision, which constitutes a shared responsibility across different levels of government in several countries—especially in federal systems like Brazil and the United States (Rodrigues et al., 2021; Smith, 2020; Woody, 2005). However, studies assessing trust and governmental responses frequently focus solely on the executive branch. Whether explicitly (Ayala Durán, 2021c; Sosa-Villagarcia & Hurtado, 2021) or implicitly (Ayala Durán, 2022a; Ayala Durán & Mendez, 2021; Gozgor, 2021; Poma & Pistoresi, 2024), existing literature often neglects trust across different government branches. This oversight can lead to misleading interpretations and creates a gap in the literature, as i) citizens' perceptions of government trust may differ across branches, and ii) international public responses to COVID-19 have exhibited clashes among executive, legislative, and judicial branches.

Firstly, while executive branches typically bear heightened responsibilities in delivering health services during pandemics or emergencies, effective public action necessitates the collaboration and shared action across several governmental branches and levels (Rodrigues et al., 2021; Woody, 2005). Generally, the parliament is responsible for funding allocation, while the executive implements the plans approved by the parliament. Additionally, the judiciary often oversees scrutiny and accountability of funds. Secondly, conflicts among government branches became evident globally in 2020 (Ádám & Csaba, 2022; Arévalo, 2021; Jørgensen et al., 2021) including in the Americas, where countries such as the United States, El Salvador, and Brazil experienced clashes among their executive, legislative, and judicial branches regarding the pandemic management (Arévalo, 2021; Jung & Lee, 2023; Renteria & Arellano-Gault, 2021; Smith, 2020). In the United States and Brazil specifically, the heads of executive branches (Presidents) often conflicted with state and local governments over lockdown implementation and social distancing policies (Jung & Lee, 2023; Rodrigues et al., 2021; Smith, 2020). Despite this inter independence, studies examining trust during the pandemic overly focus on national governments, notably executive government, relegating legislative or judicial branches. Even studies that assess multiple levels of the administration, such as Poma and Pistoresi (2024), implicitly frame their inquiry as “trust in your country government”, namely the executive branch. In time, this may conflate trust in different institutions for trust in the executive government.

To help address this gap in the literature, this study examined trust in two government branches: the presidency and the parliament. To achieve this, the following strategy was implemented.

3. Materials and Methods

This study assesses the effect of receiving the single $300 monetary transfer on trust in the government's handling of the COVID-19 pandemic. To this end, a probabilistic nationwide sample collected in April 2020 is utilized. To the best of the authors’ knowledge, this represents the first data collection exercise since the enactment of lockdown measures in March 2020 in El Salvador and one of the earliest in the region. The database was provided by the Centro de Estudios Ciudadanos [Center for Citizen Studies] (2020), a research institution within Francisco Gavidia University. This survey was conducted with a 95% confidence level and a sampling error of 2.5%, including a total of 1,222 individuals. Data gathering was completed between April 23^rd – 27^th (Centro de Estudios Ciudadanos, 2020), while the monetary transfer was announced April 6^th., 2020. For the implementation of this study, the following questions were used as dependent variables:

Aiming to assess the effect of the transfer, Mahalanobis nearest neighbor score matching was employed (Rubin, 1978) in STATA 17. This represents a quasi-experimental matching method widely utilized in political science and public policy, as it reduces imbalance when matching observations and assessing impact. Evidence suggests that Mahalanobis matching outperforms other methods while producing smaller standard errors (King & Nielsen, 2019; Zhao, 2004, p. 97). Additionally, matching offers a more flexible alternative to OLS regression by relaxing key assumptions, such as the orthogonality of the regression error and the linearity of the relationships with covariates, though it may lead to higher variance (Black, 2015). Additionally, matching allows for the estimation of both the average treatment effect (ATE) and the average treatment effect on the treated (ATET). A robustness check using the OLS estimator produced similar results; however, due to its ability to improve covariate balance, to reduce model dependence and the possibility to calculate both ATE and ATET, nearest neighbor matching is employed. The control variables used to match policy effects on public trust are described in Table 1. Variables such as age, income, and education are commonly used in similar studies. Although relevant to political trust, the original survey did not include any political variables or past voting behavior that could serve as controls, representing a limitation to this text. The central hypothesis of this text is that economic policies and public support typically enhance trust in political institutions (Liu et al., 2022; Poma & Pistoresi, 2024). Consequently, higher levels of trust should be linked to the receipt of monetary transfers.

4. Results and discussion

Table 2 presents the descriptive statistics. Notably, the population exhibits an overly positive assessment of presidential trust, approaching a score of nine points (on a 1-10 scale). In contrast, trust in the parliament is significantly lower, with a score of nearly four points. Additionally, the sample can be broadly characterized as relatively young, with a slight majority of women. Nearly fifty percent have completed high school education (45.50%), while more than half possess a family monthly income of $500 or less (62.42%). Only a quarter of respondents were beneficiaries of the monetary transfer in El Salvador. Similar grants with varying durations were implemented in other South American countries such as Brazil, Colombia, and Ecuador (Barrios et al., 2022; de Leon et al., 2023; Jara et al., 2022) as well as in Central American nations like Costa Rica and Guatemala (Martínez & Sánchez, 2022). In South America, the implementation of transfers appears to have followed allocation mechanisms targeting lower-income and/or rural households (de Leon et al., 2023, p. 144; Jara et al., 2022, p. 2792). However, in El Salvador, the selection process for this grant did not adhere to clear criteria, and often lower-income households were not included as beneficiaries (Ayala Durán, 2021b; Ross, 2022). The absence of technical guidelines on public subsidy allocation in El Salvador is not unprecedented, extending to the pre-COVID-19 period (Ayala Durán, 2021a, 2022b).

Mahalanobis nearest neighbor matching allowed for the pairing of observations, resulting in an overall mean difference in match scores of -0.0255 (not shown in text). This indicates virtually no variation between the control and treatment units. As indicated in previous literature, this method reduces imbalance and decreases standard errors (King & Nielsen, 2019; Rubin, 1978; Zhao, 2004). The density plots for individual variables with the highest mean average treatment effect variation (presidential trust) are presented in Figure 1. Visually, matching substantially reduced the differences between treated and control units1.

Figure 1Density plot for raw and matched cases across recipient status. Source: own elaboration.

Results on the impact of the monetary transfer on presidential and parliamentary trust in managing the pandemic are presented in Figure 2. For presidential trust, the average treatment effect peaked at 0.319. This indicates that if the entire sample had been assigned the monetary transfer, the pandemic handling trust score would have increased by 0.319 points (on a 1-10 scale). A robustness check using OLS, not included in the main text, yields an equivalent measure (β = 0.319, p = 0.004, 95% CI = [0.102, 0.536]). Similarly, those who were actual beneficiaries (average treatment effect on the treated) only exhibited a higher score of 0.227 (on a 1-10 scale). Both coefficients were statistically significant, although their magnitudes remained relatively low. The monetary transfer resulted in only a 3.19% (ATE) and 2.27% (ATET) increase in trust on a 1-10 scale. Such increases in trust levels are likely to result in greater compliance with COVID-19 restrictions (Devine, 2024; Güemes & Brugué, 2017; Kreps & Kriner, 2020), which represented a relevant objective during the early outbreak of COVID-19 in El Salvador. These findings also align with previous evidence on COVID-19 emergency policies and increased trust globally (Ayala Durán, 2021c; Liu et al., 2022; Poma & Pistoresi, 2024; Rieger & Wang, 2022). Notably, Liu and Hoque (2022, p. 586) report that one standard deviation increase in economic support measures yields a 0.367% increase in public trust.

Figure 2Monetary transfer impact on presidential and parliamentary pandemic handling. Coefficients with 95% confidence intervals. Source: own elaboration. Legend: ATE (average treatment effect), ATET (average treatment effect on the treated), β (coefficient), p (p-value).

Conversely, the monetary transfer did not yield statistically significant effects on parliamentary trust. This represents a rather surprising finding that the Salvadoran parliament consistently supported the executive branch by approving funds, enacting a national state of emergency, and creating lockdown orders. Notably, the parliament promptly approved the disbursement of up to two billion U.S. dollars in March 2020 (Decree 608, 2020), even when this occurred just days after the first recorded case of COVID-19 in the country. Trust in parliament has historically been lower than trust in the presidency, which may help explain this finding. According to the America’s Barometer survey from 2004 to 2023, the average trust in parliament was 3.90 (on a 1-7 scale, where 1 represents little trust and 7 represents a high level of trust). Meanwhile, trust in the president peaked at 4.19 between 2008-2023. Figure 3 illustrates how trust in parliament has regularly remained lower than trust in the presidency over the past two decades.

Figure 3Trust in the president and parliament (2004-2023) with 95% confidence intervals. Source: own elaboration based on Vanderbilt University (n.d.).

The absence of impact also contradicts previous studies on COVID-19, which found that trust often remained consistent across different government levels. For instance, Poma and Pistoresi (2024, p. 367) report that public trust indicators deteriorated among several European institutions between 2020 and 2021, including central governments, police, healthcare systems, and the European Union. Similarly, Deslatte (2020) found consistently lower levels of trust in American health authorities, state representatives, and the president as communicators of COVID-19 actions.

Although the reasons explaining the differing effects across government branches in El Salvador exceed the survey data and scope of this text, ex post evidence suggests several factors: i) an authoritarian grip by the executive branch, ii) a populist communication approach, and iii) political clientelism in the context of an upcoming parliamentary election.

Despite the Salvadoran executive enacting policies to counter the pandemic as early as March 2020, its response was often improvised and frequently disregarded democratic ruling (Ayala Durán, 2021b). Human rights organizations reported illegal actions by the Salvadoran executive, including unconstitutional mobility restrictions, illegal incarcerations, and systematic human rights violations (Gavarrete, 2020; Procuraduría para la Defensa de los Derechos Humanos, 2020; Vaquerano, 2020). While public appraisal was generally positive, the pandemic allowed the Salvadoran president to harness public discontent and tighten his authoritarian control through a permanent state of constitutional exception, even at the cost of defying the Salvadoran parliament and Supreme Court (Landeros, 2022). This trend aligns with the broader consequences of the COVID-19 crisis in Latin America, as Martí & Alcántara (2021) observed, particularly the increasing dominance of executive branches over other institutions and the growing personalization of politics at the expense of political parties. Outside of Latin America, Russia, Hungary or the Philippines also took advantage of the pandemic to increase their power and undermine the opposition (Ádám & Csaba, 2022; Arévalo & Wenham, 2020; Rennó & Ringe, 2022).

Secondly, the president employed antagonistic and populist rhetoric to undermine the parliamentary and judicial branches during the pandemic (Luna et al., 2020; Mila-Maldonado et al., 2022; Nilsson, 2022; Roque, 2021; Wolf, 2021). This included clear populistic traces as elite critique, anti-pluralism, and the formation of identity politics. The president consistently associated the traditional political class with selfish interests, actively targeting the National Assembly (Mila-Maldonado et al., 2022, p. 29). This strategy extended beyond the legislature, as opposition parties, civil society organizations, and independent media also faced sustained pressure (Martí & Alcántara, 2021; Nilsson, 2022) further weakening dissent. By opposing other institutions, the president galvanized citizen discontent to increase his popularity.

Thirdly, El Salvador was preparing in 2020 for general elections in early 2021. The president and his allies exploited the power dynamics associated with COVID-19 management, adhering to a clientelist logic. This became increasingly evident months after the implementation of the monetary transfer, as other public emergency support policies were executed using the president's political party logistics and propaganda. The exchange of favors for votes in 2021 was central to Bukele’s agenda, especially considering his minimal parliamentary representation before 2021. Electing loyal members to the Legislative Assembly was crucial for implementing his political agenda. The 2021 election would not only provide political leverage until the end of his first term but also enable him to controversially oust all constitutional court judges and the attorney general. This move allowed him to appoint former executive advisors to the highest chambers of the judiciary, including the Supreme Court, effectively curtailing criminal investigations into corruption and abuse of authority involving him and his family (Molinari, 2024, p. 36). According to Nilsson (2022, p. 23), the removal of the Supreme Court’s Constitutional Chamber, along with other judicial appointments (including attorney general), represents a clear indication of clientelism, potentially fostering corruption. These appointments ultimately enabled him to run for reelection, despite constitutional ban.

The monetary transfer significantly increased trust in the executive branch but had no measurable effect on trust in parliament. This contrast highlights the complexities of political trust in emergency situations. Although executive trust can increase due to public actions during emergencies, the effect is not necessarily uniform across government branches, showcasing the challenges of governance and accountability at different State levels. This underscores the need for an expanded research agenda that includes various government branches in studies of public trust during COVID-19 and other similar emergency situations. Such an agenda would prove relevant considering that executive branches have clashed with other levels of the public administration in countries such as the United States, Brazil, Mexico, and El Salvador (Arévalo, 2021; Jung & Lee, 2023; Mila-Maldonado et al., 2022; Renteria & Arellano-Gault, 2021; Smith, 2020). Additional public institutions, such as the armed forces, political parties, or state media, should also be included in this research agenda. In addition to the public arena, expanding trust studies to encompass private institutions during emergencies would also be beneficial. In Latin America alone, such entities might include churches, private media, or civil society organizations. Recent studies highlight the importance of including such private actors (Bandeira & Carranza, 2020; Giménez Béliveau, 2021; von Sinner & Zeferino, 2022) as well as other overlooked public institutions like the armed forces (Gedan, 2021; Siman et al., 2024).

Lastly, the results presented here only reflect the early development of the COVID-19 pandemic in El Salvador. However, they focus solely on this initial stage and represent merely one of the emergency measures aimed at supporting Salvadoran households. Several other policies, such as loan payment freezes, food baskets, and utility moratoriums, followed. Further research is needed to assess the overall impact of these policies on public trust while accounting for broader governmental participation over time.

5. Closing Remarks

This paper examined whether the unconditional monetary transfer at the beginning of the COVID-19 pandemic had significant effects on citizens’ trust in the pandemic management by the Salvadorian president and parliament. This fills a literature gap, as other studies focusing on different levels of public administration often overlook the parliamentary branch. To the authors’ knowledge, this is the first study to explicitly examine trust in both the legislative and executive branches simultaneously during the COVID-19 crisis.

The results indicate a positive effect between the emergency transfer and presidential trust, although its magnitude remains relatively low on a 1-10 scale. This finding aligns with the COVID-19 literature on public trust, highlighting the positive impact of public action on the population’s confidence. Consequently, this is likely to result in higher compliance with COVID-19 measures, such as mobility restrictions, vaccination, and social isolation. Conversely, the effect was not consistent for trust in the parliament, which reported no relationship between the monetary transfer and parliamentary confidence. This null effect can be explained by the increasing political tensions in El Salvador, including a presidential authoritarian grip, a populist approach, and the clientelist logic associated with the upcoming 2021 general elections.

These results underscore the importance of including different government branches when studying public trust in emergency situations. Such an often-overlooked addition would enhance our understanding of public trust, particularly when different branches are at odds, as seen in countries like the United States, Brazil, and El Salvador. An expanded research agenda on trust could also encompass relevant private institutions, notably churches, media, and civil society organizations.

Lastly, in the Salvadorian context, this monetary transfer represented only one of several emergency policies implemented during the pandemic. Other measures included prohibitions on layoffs, loan freezes, and food baskets. Assessing the effects of these policies on public trust remains a vital area for future empirical research.