Effective Transparency and Institutional Trust in Honduran Municipal Governments

Abstract

This article investigates trust in Honduran Municipal Governments and shows that transparency and municipal services quality impact on trust. We posit the Theory of Effective Transparency: relevant information should be reported; transparency should be based on unbiased information; and municipal data should be comparable with other municipal governments. Municipalities where one political party has an overall majority are trusted more, and audited financial statements generate a positive impact on trust. Engagement and participation also affect trust. As a practical implication, municipal governments should increase trust by involving citizens in municipal issues. Finally, quality of democracy is connected with greater trust.

Keywords

trust transparency local government corruption municipal services quality engagement

Introduction

Trust is considered a cornerstone for democracy, as it enhances the legitimacy, validity, and sustainability of governments by connecting citizens with public institutions (Godefroidt, Langer, & Meuleman, 2017).

At the local level, academics have offered empirical evidence that trust is an important component of the perceived legitimacy of a political system. On one hand, it reinforces compliance with government regulations and taxes, thus preserving the rule of law. On the other, it can contribute to democratic stability in new democracies (Córdova & Layton, 2016).

Although trust is considered as a critical component of good governance (Wu, Ma, & Yu, 2017), public trust in governments has decreased significantly in recent decades (Gordon, 2000; M. J. Park, Choi, Kim, & Rho, 2015; Welch, Hinnant, & Moon, 2005). According to Bouckaert and van de Walle (2003), the most frequently expressed concern by politicians, journalists, and the general public is the decrease in confidence in the government and the harmful effects that this generates for it.

Despite the importance of institutional trust and the crucial role played by local governments in the preservation of economic development and democratic governance, theories about which factors explain citizens’ confidence in local governments remain underdeveloped (Córdova & Layton, 2016). Hence, there is a lack of research on government trust in Central America and, specifically, on municipal government trust.

This article seeks to fill this research gap, and our research question is to what extent transparency and other factors impact on institutional trust. Transparency is essential for the proper functioning of the governments. Public sector transparency is required by state legislation to guarantee the understanding of public policies, to reduce uncertainty, and to promote citizen participation and democracy (Lee, 2017).

A transparent fiscal system encourages politicians to adopt better policies that consolidate a sustainable fiscal system (Arbatli & Escolano, 2015) while also improving public debate. In Latin America, recent legislation has sought greater disclosure and transparency. Nevertheless, the legacy of authoritarianism and opacity still pervades in most Central American countries (Michener, 2015). Sometimes, nominal transparency is far from effective as there may be a gap between what is reported and what is happening in reality (Heald, 2006). This article focuses on transparency as an effective tool to both enhance trust in municipal governments and hold them accountable.

From the Latin American Public Opinion Project (LAPOP) survey, we take a sample of 1,561 responses from 51 Honduran municipalities in 2014. We chose this country and the year 2014 for two reasons. First, Honduras is the only country where the LAPOP questionnaire includes the question about “transparency in municipal government,” and, second, 2014 is the most recent year for which municipal financial data are available in Honduras. These financial data appear as control variables in our models.

Our data show that transparency significantly impacts on trust, but that corruption does not. In addition, the quality of municipal services strongly impacts on trust. Finally, both citizens’ engagement and participation impact on trust.

The article is organized as follows. “Literature Review on Trust in Government” section discusses the literature. “Local Governments in Honduras” section describes the Honduran municipal sector. “Method” section addresses the methodology. “Results and Discussion” section offers our results and the discussion. Finally, “Conclusions and Further Research” section summarizes the conclusions and proposes further research.

Literature Review on Trust in Government

Introduction

Theories of collective action point to the importance of the rules of reciprocity, reputation, and trust, where individual action is influenced, in a certain way, by expectations about how others will act (Bauhr, 2017). Institutional legitimacy contributes to the organization of effective political groups and government institutions, which in turn promotes citizens’ trust (Marlowe, 2004; Maxwell, 2010).

The literature has identified several determinants of institutional trust. First, transparency shows a mixed impact on trust, meaning that if governments become more transparent, this does not ensure, per se, higher trust from citizens. Second, corruption undermines institutional trust. Here, we must bear in mind that the impact differs when comparing different contexts. Thus, in developed countries, corruption erodes trust in the government, but it does not destroy the democratic system. However, corruption in developing countries has deeply shaken these new democracies. Third, citizens’ satisfaction with municipal services is clearly identified by the literature as a determinant of trust in municipal governments. Finally, the literature has reported additional determinants of institutional trust, which are summarized in Table 1.

Table 1.

Control Variables.

Determinant of trust		Reference	Methodology	Findings
Political factors	Partisanship-ideology (?)	Córdova and Layton (2016)	Data set collected in 2010 and 2011 in El Salvador	Political partisanship has a weak effect on trust.
		T. E. Cook and Gronke (2005)	National telephone survey in 2002 in the United States.	Democrats are not inclined to trust government less or to express active distrust. Conservatives are presumably less favorably inclined to “government” but may be more so toward the political institutions they are in charge of.
		Rudolph and Evans (2005)	2000 National Annenberg Election Survey (USA).	Effects of political trust are significantly more pronounced among conservatives than among liberals. Ideology conditions the effects of political trust on attitudes toward both distributive and redistributive spending.
		Stoyan, Niedzwiecki, Morgan, Hartlyn, and Espinal (2016)	2008, 2010, and 2012 LAPOP. Haiti and Dominican Republic.	Right-leaning respondents in both countries are more trusting of government than those on the left. Some support to the view that democratic values are positively associated with trust.
		Wilkes (2015)	Data from American National Election Studies, USA (1958-2012).	It is also clear that the magnitude of the effects of partisanship on trust also vary by race. Black Democrats are less trusting than White Democrats.
		Wang (2016)	2005-2008 Asia Barometer survey (Japan, South Korea, and Taiwan)	People who identify with the party in government are more likely to place political trust than their counterparts.
	Mayor gender (−)	Araujo and Tejedo-Romero (2016)	100 largest Spanish municipalities in 2008	Women’s political representation in municipalities has a positive influence on trust.
Local government financial indicators (+)		Stoyan et al. (2016)	2008, 2010, and 2012 LAPOP. Haiti and Dominican Republic.	Economic evaluations have significant positive effects on institutional trust in each survey except Haiti 2012.
		Porumbescu (2016)	This study uses data from a 2012 online survey of Seoul citizens.	Relationship between propensity to trust and trust in government is completely mediated by perceptions of public sector performance.
		Intawan and Nicholson (2018)	U.S. respondents in 2015, American National Election Studies data	The positive and significant coefficient for explicit trust suggests that higher trust in government leads to greater support for increasing spending.
		Espinal, Hartlyn, and Kelly (2006)	Three surveys conducted over 1994 to 2001. Dominican Republic.	Trust in government institutions is shaped primarily by perceptions of economic performance by government.
		Whiteley, Clarke, Sanders, and Stewart (2016)	British electorate surveys conducted each month from June 1997.	The economic performance and trust in government are strongly related.
Socioeconomic factors	Education (+)	Sønderskov and Dinesen (2016)	Two Danish panel surveys: Danish Data Archive 2002, 2009, and Danish Social and Political Attitudes Survey	Education does not have a significant impact on trust given the limited within-individual variation in a panel survey of adults.
		Seligson (2002)	Interviews: El Salvador; Nicaragua; Bolivia, and Paraguay.	Higher education is linked to lower trust in Paraguay and Bolivia.
		Corbacho, Philipp, and Ruiz-Vega (2015)	2007 World Gallup Survey. Latin American countries.	Education shows a negative impact on trust.
		Espinal et al. (2006)	Three surveys conducted over 1994 to 2001. Dominican Republic.	Increased level of education is positively, but weakly, associated with greater institutional trust.
		Montalvo (2010)	2008 LAPOP. 23 Latin American nations.	Citizens with more years of completed education express less trust in municipal governments than those with less formal education.
		Morris and Klesner (2010)	2004 LAPOP: Mexico.	Education is inversely related to trust in political institutions.
Socioeconomic factors	Age (+)	Espinal et al. (2006)	Three surveys conducted over 1994 to 2001. Dominican Republic.	Age has a nonlinear effect as older generations, who experienced authoritarianism as children, are more trusting of democratic institutions.
		Montalvo (2010)	2008 LAPOP. 23 Latin American nations.	As people grow older, the degree of trust tends to increase relative to younger individuals.
		Corbacho et al. (2015)	2007 World Gallup Survey. Latin American countries.	Age shows a positive impact on trust.
		Seligson (2002)	Interviews: El Salvador, Nicaragua, Bolivia, and Paraguay.	Age does not show a clear impact on trust.
	Gender (−)	Seligson (2002)	Interviews: El Salvador, Nicaragua, Bolivia, and Paraguay.	Gender does not show a clear impact on trust.
	Gender (−)	Godefroidt, Langer, and Meuleman (2017)	National Service Scheme Survey	Male students exhibit slightly but significant higher levels of trust in institutions than their female counterparts
	Race (?)	Wilkes (2015)	Data from American National Election Studies (USA).	The magnitude of the effects of partisanship on trust also varies by race. Black Democrats are less trusting than White Democrats.
		Emig, Hesse, and Fisher (1996)	Alabama (USA)	Black were significantly more politically trusting, efficacious, and involved in the community than were Whites.
		Howell and Fagan (1988)	Registered voters (New Orleans, USA).	Very notable racial differences are apparent in the local sample. Over half of the Blacks exhibit high trust in government.
Socioeconomic factors	Communication (+)	M. J. Park, Choi, Kim, and Rho (2015)	Korean sample, people eligible to vote and who used Facebook.	Formation of citizen trust in government can be expanded into patronage intention toward social media and actual behavior.
	Communication (+)	Welch, Hinnant, and Moon (2005)	Council for Excellence in Government (Hart/Teeter 2001, USA).	Trust depends heavily on general government satisfaction variables—involvement, security, efficiency, and information provision.
	Income and labor (+)	Córdova and Layton (2016)	Data set collected in 2010 and 2011 in El Salvador.	Individuals with the lowest socioeconomic status show a lower level of trust in local government than individuals with the highest socioeconomic status.
		Seligson (2002)	Interviews: El Salvador, Nicaragua, Bolivia, and Paraguay.	No clear impact of income on trust.
		Marlowe (2004)	Data from the 1996 General Social Survey (n = 401). Americans.	The personal characteristics examined here (include income) do not seem to play any role in the assessment of trust in public administrators.
		Montalvo (2010)	2008 LAPOP. 23 Latin American nations.	Strong relationship between economic growth and institutional trust.
		Espinal et al. (2006)	Three surveys conducted over 1994 to 2001. Dominican Republic.	Curvilinear effect between income and institutional trust: middle-sector groups less trusting government institutions than either the poor or wealthy
	Participation and engagement (+)	Espinal et al. (2006)	Three surveys conducted over 1994 to 2001. Dominican Republic.	Little evidence of a relationship between civic engagement and institutional trust.
		Bauhr (2017)	Transparency International Global Corruption Barometer 2013	Citizens who have higher levels of institutional trust and who perceive corruption as a problem are more likely to engage against corruption.
		Brehm and Rahn (1997)	General Social Survey cumulative file for 1972-1994 (USA).	Relationship between civic engagement and interpersonal trust, and that this relationship has consequences for confidence in institutions.
		Stoyan et al. (2016)	2008, 2010, and 2012 LAPOP. Haiti and Dominican Republic.	Civic engagement would have greater potential to promote trust in institutions (2008 and 2012 Haiti surveys).
		Montalvo (2010)	2008 LAPOP. 23 Latin American nations.	Individuals who participate more often in community improvement activities tend to manifest higher degrees of trust in their municipalities.
	Democracy (+)	Marlowe (2004)	Data from the 1996 General Social Survey (n = 401). Americans	Respondents who felt that democracy works fine agree that administrators can be trusted more.
	Democracy (+)	Espinal et al. (2006)	Three surveys conducted over 1994 to 2001. Dominican Republic.	No relationship between democratic values and institutional trust.
	Trust in others (+)	Maxwell (2010)	2007 Home Office Citizenship Survey	As individuals’ social and political attitudes are more positive, they will be more likely to have high levels of trust in government.
		Montalvo (2010)	2008 LAPOP. 23 Latin American nations.	Trust in municipal governments improves as interpersonal trust increases.
		Marlowe (2004)	Data from the 1996 General Social Survey (n = 401). Americans	Trust in other positively affects trust in public administrators.

Note. LAPOP = Latin American Public Opinion Project. Expected sign is given in parenthesis.

Transparency

There is no clear impact of transparency on trust. As Cucciniello, Porumbescu, and Grimmelikhuijsen (2017) indicate, the relationships between transparency and trust in government depend on whether the study drew on observational data or experimental data. In fact, these authors find that, from the literature they reviewed, seven articles show a positive impact, three a negative one, four mixed ones, and one finds no effect.

Steves (2001) refers to the Southern Cone of Latin America, and states that democratic development served as the “condition of possibility,” creating institutions which fostered transparency and trust. Transparent institutions enhance their credibility, which is connected with higher trust. In a survey on Latin American Municipalities, Pérez, Bolívar, and Hernández (2010) show that increasing transparency will generate a higher degree of trust. According to Cooper and Yoder (2002), efforts are being made in developing countries to increase transparency, which should improve trust in governments. As these authors state, trust is the ultimate object of transparency. In developing, highly corrupt countries, Baur and Grimes (2014) empirically show that an increase in transparency tends to breed resignation rather than indignation. If corruption is endemic, transparency can imply higher levels of distrust in institutions and provoke public resignation, due to the firm belief that corrupt practices are impossible to eradicate.

In developed countries, some scholars show that transparency increases the degree of trust (H. Park & Blenkinsopp, 2011). López-Arceiz, Bellostas Pérezgrueso, and Rivera Torres (2017) and Zhao and Hu (2017) define transparency as a strategic responsibility that is essential in enhancing trust in relationships (Bauhr & Grimes, 2014; Wasike, 2016). In addition, transparency creates a climate of trust among people and encourages them to believe more in governments, fostering an environment of openness that allows citizens to control institutions and participate in decision making.

For some academics, political distrust is not a consequence of weak government, but rather, the point is that better-informed citizens trust the government more, as they understand clearly how local government policies contribute to public welfare (F. L. Cook, Jacobs, & Kim, 2010; Porumbescu, 2015). Accordingly, disseminating information about government activities is crucial in increasing public confidence (Grimmelikhuijsen & Meijer, 2014).

Aside from these effects of transparency on trust, some authors, such as Buehn, Dell’Anno, and Schneider (2017), consider that the lack of transparency impacts on the loss of confidence, which in turn triggers fiscal illusion. Furthermore, agency theory shows that higher levels of information reduce information asymmetries, which limits potential conflicts of interest between principals and agents, and increases confidence in politicians (Araujo & Tejedo-Romero, 2016).

De Fine Licht (2014) finds no significant effect of transparency in trust. Grimmelikhuijsen, Porumbescu, Hong, and Im (2013) find that transparency has no significant effect on perceived trustworthiness. Similarly, Grimmelikhuijsen and Meijer (2014) show mixed impact of transparency on perceived trustworthiness. Piotrowski and van Ryzin (2007) find a negative relationship between trust in local officials and government transparency. Finally, Worthy (2010) reports that increased transparency is not connected with higher trust in government.

We align our assumption with the literature that finds a positive impact of transparency on trust. Accordingly, we assume the following hypothesis:

Hypothesis 1: Higher transparency will have a positive impact on trust.

Corruption

Canache and Allison (2005) show that corruption undermines the legitimacy of governments in Latin America, which erodes government trust. Similarly, Seligson (2002), in an empirical article on El Salvador, Nicaragua, Bolivia, and Paraguay, shows that those who experience corruption are less likely to believe in the legitimacy of their political system. Espinal, Hartlyn, and Kelly (2006), on a survey in the Dominican Republic, find that respondents who regard corruption as less of a problem are more trusting than those who consider it a very serious problem. Individuals surveyed in 23 Latin American countries who have been victimized by corruption tend to trust their municipalities less (Montalvo, 2010). In a study on Mexico, respondents with stronger perceptions of corruption exhibited a lower level of political trust (Morris & Klesner, 2010). Similarly, people from El Salvador who have been asked for a bribe show lower levels of trust in the local government (Córdova & Layton, 2016).

The drop in institutional quality stemming from corruption can erode social trust in the government (Bauhr & Grimes, 2014). Therefore, policymakers should be aware of how corruption can affect institutional trust (Sønderskov & Dinesen, 2016). The impact of corruption differs when we compare developed with developing countries. As Canache and Allison (2005) indicate, corruption scandals in developed countries have undoubtedly eroded trust and confidence in the government, but they have not destroyed their democratic systems. Conversely, incidents of political corruption in developing countries have shaken these new democracies to their core. The literature summarized so far shows the impact of corruption on trust in developing countries. As Canache and Allison (2005) point out, trust works differently in the developing countries, so the remainder of this section covers literature on developed countries as a point of contrast.

Empirical evidence from Japan, South Korea, and Taiwan shows that the perception of corruption is negatively associated with trust (Wang, 2016). If politicians have nothing to hide, that is, there are no corruption signals, they are more likely to enjoy people’s trust (Alt, Lassen, & Rose, 2006).

Although the corruption/trust relationship has been shown to be negative, there is a risk of the coexistence of corruption with high institutional trust and, in particular, confidence in the ability of the political system to fight corruption. Thus, there may be a lack of civic commitment of inhabitants in the fight against corruption, assuming that the government can fight corruption. This may happen in developed, low corrupt countries, such as Sweden, where citizens think their governments can curb corruption (Bauhr, 2012).

In accordance with the literature review, we posit the following hypothesis:

Hypothesis 2: Higher corruption will have a negative impact on trust.

Satisfaction With Public Services’ Quality

Public distrust is often attributed to the malfunctioning of public services (van de Walle & Bouckaert, 2003). According to Wu et al. (2017), fair and reliable public services and predictable decision making inspire public confidence. In this respect, management theories connect lack of confidence in government with low performance of government systems (Bouckaert & van de Walle, 2003). Consequently, the open flow of information and communication is crucial for citizens’ confidence in public services (Araujo & Tejedo-Romero, 2016).

In a sample of Latin American countries, Montalvo (2010) finds that the most powerful explanation of trust is the degree of satisfaction with municipal services. Citizens who are satisfied with the services provided by their municipalities tend to express higher trust than those who are dissatisfied. Córdova and Layton (2016) reveal that the perception of citizens from Central America (El Salvador) about the proper functioning of municipal services shapes their trust in the municipal government.

Porumbescu (2016) and Marlowe (2004) find a strong significant relationship between citizens’ perceptions of public sector performance and trust in government. Citizens’ trust means that governments work in the way they prefer (Bouckaert & van de Walle, 2003).

Following the empirical and theoretical literature summarized in this section, we posit the following hypothesis:

Hypothesis 3: Greater satisfaction with public services will have a positive impact on trust.

Control Variables

The literature has identified additional variables that impact on government trust. These variables are summarized in Table 1.

Table 1 also shows the expected signs of the impact of the control variables on government trust. The context of every empirical work has been highlighted, so emerging democracies, into which Honduras fits, can be contrasted with more advanced democracies.

Local Governments in Honduras

The Honduran public sector is organized into three levels: Central State, 18 regional governments, and 298 local governments. The municipality is organized and operates independently of the State. Municipal autonomy allows election of authorities through direct and secret suffrage; power to collect taxes; preparation, approval, and execution of the budget; and municipal public services management, among others. Citizens may hold the municipality accountable for its acts, participate in the management and local affairs, and request the municipal financial statements either in open councils (the law requires at least five open councils per year) or directly from the municipality.

As of 2009, Transparency Citizen Commissions (TCC) were established in each Honduran municipality to enable social audits, that is, the participation of men and women from community, unions, businesses, and all existing social organizations, to guarantee the transparent execution of programs and projects, as well as the provision of public services in an effective and efficient way. TCC are independent organisms of the municipalities, integrated by a minimum of five members elected by the citizens’ assembly, who perform their functions for a period of 3 years, and are eligible for one re-election. First, TCC ensure the transparent execution of municipal programs and projects, as well as the provision of public services in an effective and efficient way. Second, TCC encourage citizens’ participation in the evaluation of municipal services, make sure municipalities meet the transparency law and grant access to public information, and, chiefly, contribute to curbing municipal corruption. Third, TCC seek to increase citizens’ trust in local governments.

Method

Sample and Variables

We employ, among other data, the 2014 Americas Barometer survey conducted by LAPOP to evaluate the determinants of trust in local governments. The sample consists of 1,561 survey responses from 51 Honduran municipalities, representative of all Honduran regions. We briefly anticipated in the introduction why the focus is on Honduras in 2014. In this section, we provide the full details. We took the year 2014 for the study for three reasons. First, the LAPOP questionnaire item “transparency of the municipal government” is only available for 2014 and 2016 in Honduras. Second, the survey is conducted every 2 years for Honduras, so the latest available database corresponds to 2016. Third, the indicators of the financial situation of the Honduran municipalities for 2016 are not available. However, for some specific tests for which financial data are not essential, we will take the whole LAPOP time span (2004-2016).

In “Literature Review on Trust in Government” section, Table 2 describes the variables considered in the regressions. First is the dependent variable: trust in local governments. As regressors, we take several groups of variables: transparency; corruption; quality of services; political, financial, and personal features; communication and access to news; income and labor; participation and engagement in municipal issues; opinion about quality of the democracy; and trust in other people. Table 3 reports correlations of the main variables (the full table is available upon request to the authors).

Table 2.

Descriptive Statistics.

Variable	Description	Observation	M	SD	Minimum	Maximum
Trust
trust_muni	Trust in municipal government. From 1 = none, till 7 = a lot.^a	1,541	4.223	1.961	1	7
Corruption
num_polit_corrup	To what extent is public officials’ corruption widespread. From 1 = none, till 4 = a lot.^a	1,511	3.029	0.864	1	4
count_prob_corrup	Corruption as the most serious problem faced by the country. 1 = corruption, 0 = otherwise.^a	1,561	0.060	0.237	0	1
corrup_bribe	Public employee has requested bribe in the last year. 0 = no, 1 = yes.^a	1,561	0.058	0.234	0	1
Transparency
tran_muni	Transparency of municipal government. From 1 = not transparent, till 4 = very transparent.^a	1,509	2.374	0.965	1	4
Quality
munserv_qual	Municipal services’ quality. From 1 = very bad, till 5 = very good.^a	1,484	3.212	0.869	1	5
Political
mayo_ideology	Mayor’s political ideology (2013 elections). 0 = left, 1 = right.^b	1,561	0.583	0.493	0	1
majority	Mayor’s party holds majority in municipal council (2013 elections). 0 = no, 1 = yes.^b	1,561	0.417	0.493	0	1
mayor_gender	Mayor’s sex. 0 = woman, 1 = man (2013 elections).^b	1,561	0.940	0.237	0	1
cit_ideology	Political alignment of the respondent. From 1 = left, till 10 = right.^a	1,364	6.175	2.926	1	10
presi_vote	Political ideology of the party the respondent voted (2013 elections). 0 = left, 1 = right.^a	1,017	0.555	0.497	0	1
Financial
r_revenue1pc	Municipal direct tax + indirect tax, per capita (2002 real Lempira).^c	1,561	130.559	118.597	2.459	363.061
r_revenue2pc	Municipal direct tax + indirect tax + fees, per capita (2002 real Lempira).^c	1,561	234.924	235.533	13.636	679.961
r_debtpc	Municipal debt per capita (2002 real Lempira).^c	1,561	64.233	181.183	0	933.957
r_deficitpc1	Municipal deficit per capita (2002 real Lempira).^c	1,561	66.766	83.615	−38.839	336.388
sai_audit	2014 municipal financial statements audited by Supreme Audit Institution (SAI). 0 = no, 1 = yes.^d	1,561	0.307	0.461	0	1
Personal features
Education	Last year of education completed. From 0 = none, till 18 = university.^a	1,559	7.402	4.338	0	18
Age	Age in years.^a	1,554	39.283	14.859	18	88
Gender	Respondent’s sex. 0 = woman, 1 = man.^a	1,561	0.501	0.500	0	1
race_w	Respondent’s race. 1 = White, 0 = otherwise.^a	1,561	0.227	0.419	0	1
race_b	Respondent’s race. 1 = Black, 0 = otherwise.^a	1,561	0.028	0.166	0	1
race_n	Respondent’s race. 1 = native, 0 = otherwise.^a	1,561	0.063	0.244	0	1
race_o	Respondent’s race. 1 = other race, 0 = otherwise.^a	1,561	0.031	0.173	0	1
Communication
often_news	Attention to news (either TV, radio, newspapers, or Internet). From 1 = never, till 4 = daily.^a	1,561	3.464	0.808	1	4
Income and labor
lnr_month_income	Monthly income of the household. (2002 real Lempira, natural logarithm).^a	1,397	7.557	0.775	6.139	8.957
lnr_income	Municipal income per capita (2002 real Lempira, natural logarithm).^e	1,372	6.919	0.460	5.481	7.501
income_evol	Income evolution in the last 2 years. −1 = decreased, 0 = unchanged, 1 = increased.^a	1,554	−0.324	0.622	−1	1
person_econ_evol	Personal economic situation evolution in 12 months. −1 = worse, 0 = unchanged, 1 = better.^a	1,554	−0.381	0.644	−1	1
pubsec_job	Government or state-owned enterprise employee. 0 = no, 1 = yes.^a	1,561	0.030	0.169	0	1
Participation and engagement
mun_meet_12m	Participation in municipal meetings in the last 12 months. 0 = no, 1 = yes.^a	1,557	0.100	0.300	0	1
engagcommun_12m	Participation in the solution of community problems in the last 12 months. 0 = no,1 = yes.^a	1,555	0.358	0.480	0	1
meet_polparty	Participation in political party meetings. From 0 = never, till 4 = once a week.^a	1,556	1.317	0.638	1	4
Democracy quality
perform_demo	Satisfaction with how democracy works. From 1 = very dissatisfied, till 4 = very satisfied.^a	1,530	2.396	0.691	1	4
Trust others
trust_commun	Trust in community’s people. From 1 = untrustworthy, till 4 = very trustworthy.^a	1,544	2.850	0.972	1	4

Note. In race dummy variables (race_w, race_b, race_n, race_o), the omitted category to avoid collinearity is mixed-race people. LAPOP = Latin American Public Opinion Project.

LAPOP project dataset.

Honduras Supreme Electoral Court.

Honduras Ministry of Finance.

Supreme Audit Institution.

From the Honduras Statistics National Institute.

Table 3.

Correlations.

	trust_muni	tran_muni	munserv_qual	mayo_ideology	majority	cit_ideology	sai_audit	age	race_b	often_news	income_evol	mun_meet_12m	engagcommun_12m	meet_polparty	perform_demo	trust_commun
trust_muni	1.0000
tran_muni	.4,312	1.0000
munserv_qual	.3,309	.2644	1.0000
mayo_ideology	.0104	.0251	.0301	1.0000
majority	.1012	.1278	.1218	.2042	1.0000
cit_ideology	.1837	.1343	.1248	.0535	.0318	1.0000
sai_audit	−.0423	−.0599	−.0426	.0895	−.4,275	.0445	1.0000
age	.0748	.0228	−.0497	.0090	−.0098	.0422	.0055	1.0000
race_b	−.0433	−.0512	−.0166	−.0679	.0365	.0265	.0629	.0010	1.0000
often_news	−.0067	−.0369	−.0289	−.0236	−.0360	.0750	.0628	.1273	−.0165	1.0000
income_evol	.0549	.0376	.0482	.0223	.0064	.0326	.0151	−.0846	.0184	−.0123	1.0000
mun_meet_12m	.0886	.0601	.0345	−.0177	.0172	−.1022	−.0826	.0693	.0205	.0544	−.0437	1.0000
engagcommun_12m	.0394	.0142	−.0232	.0021	−.0194	−.0360	−.0413	.0577	−.0142	.0184	.0476	.1813	1.0000
meet_polparty	.0728	.0187	−.0149	−.0546	−.0271	−.0264	−.0652	.0702	.0004	.0739	.0072	.2078	.1447	1.0000
perform_demo	.2061	.2530	.1351	.0198	.0244	.1388	−.0335	.0294	−.0288	−.0643	.1201	−.0021	−.0404	.0202	1.0000
trust_commun	.1191	.0488	.1067	.0076	−.0491	.0400	−.0194	.0762	−.0416	−.0085	.0222	.0118	.0704	−.0191	.1137	1.0000

Note. Bold numbers indicate 5% significance.

To provide a snapshot of the main variables, Figure 1 shows the evolution from 2004 to 2016 of the main LAPOP variables at stake: interviewees’ trust in local government, perception of quality of public services, and how widespread respondents think corruption is. The trend shows a positive impact of TCC on citizens’ trust in municipalities immediately after TCC were established in 2009. Average trust in the municipality significantly increases from 2008 (3.73) to 2010 (4.41): t = ***−12.08; K-W χ² = ***139.81. However, after the first initial impact of TCC, 2012 trust returns to 2008 levels. Statistically, 2008 trust (3.73) and 2012 trust (3.79) are equal: t = −1.06; K-W χ² = 1.10. Thus, the downward trend in trust reported by the literature has not happened in the Honduran municipalities.

Figure 1.

Evolution of trust in municipality, public services’ perceived quality, and perceived corruption.

Furthermore, Figure 2 presents the level of engagement in municipal issues, which affects trust. As we explain above, at least five open council meetings are held per year. Figure 2 shows that citizens’ participation in these council meetings is lower than their direct involvement in their communities’ problems. The difference between both variables, considering all LAPOP time series (2004-2016), is statistically significant (t student = ***45.2956, probability = .000; Wilcoxon z = ***41.492, probability = .000). These analyses indicate that people prefer to solve problems on their own, within their community, rather than airing their demands at the municipal councils.

Figure 2.

Citizens’ municipal engagement (municipal meetings vs. community engagement).

Finally, it is worth mentioning that average of citizens’ ideology variable (cit_ideology) is quite high (6.175), considering that respondents must range themselves in a Likert-type scale between 1 (progressive, left) and 10 (conservative, right). We expected a figure around 5.5. To check that respondents were unbiased in their answer, we check the coherence between presi_vote and cit_ideology over the whole LAPOP time span (2004-2016). The results indicate that those respondents who voted left governments in the last elections have a significantly lower value in variable cit_ideology: t student = ***−6.1178, significance = .000, number of observations = 5,450. Therefore, we prove the consistency of this political variable.

Main Model and Robustness Checks

The main model follows this ordinary least squares (OLS) regression with heteroskedasticity-consistent standard errors for 2014:

y_{i} = γ + \sum β_{j} x_{ji} + c_{i} + ε_{i},

(1)

where y_i is municipal trust (trust_muni), γ the intercept, β_j the vector of parameters to be estimated, x_ji vector of explanatory variables, c_i the control for fixed regional effects (18-1 dummy variables accounting for the 18 Honduran regions), and ε_i the error term.

The main model is reported in Table 4, together with nine robustness checks. The main model presents the basic variables in our empirical assessment, in agreement with the literature discussed in “Literature Review on Trust in Government” section. First and second robustness checks seek to confirm the results of the impact of corruption on trust: We replace the main model corruption variable, num_polit_corrup, with two different corruption specifications, count_prob_corrup and corrupt_bribe. Robustness Check 3 substitutes r_revenue1pc by r_revenue2pc, to test the model sensitivity to different gauges of municipal revenues. Checks 4 and 5 test different specifications of household and per capita economic level of citizens. Different options of citizens’ engagement in municipal issues are tested in Checks 6 and 7.

Table 4.

Main Regression and Robustness Checks.

Variable	Main model	1	2	3	4	5	6	7	8^a	9^b
Transparency
tran_muni	***0.61408.92	***0.60758.86	***0.61428.98	***0.61388.92	***0.60518.51	***0.60688.75	***0.61008.87	***0.61649.02	***0.62648.57	***0.60249.12
Corruption
num_polit_corrup	0.02340.32			0.02180.30	0.02780.37	0.02100.29	0.03080.43	0.03710.51	0.01740.21	0.01670.24
count_prob_corrup		−0.1778−0.66
corrup_bribe			0.18950.79
Quality
munserv_qual	***0.44025.69	***0.44065.74	***0.44625.79	***0.44105.70	***0.46085.63	***0.44025.67	***0.45465.84	***0.44975.80	***0.44705.68	***0.45586.39
Political
mayo_ideology	*−0.2622−1.81	*−0.2521−1.77	*−0.2512−1.76	*−0.2606−1.80	−0.1818−1.20	*−0.2547−1.75	*−0.2487−1.74	*−0.2412−1.68	−0.2432−1.53	*−0.2288−1.77
majority	*0.35191.92	*0.30241.68	*0.30401.68	*0.35351.94	*0.33781.68	**0.38042.08	**0.36862.02	**0.37532.06	*0.38341.72	**0.34192.39
mayor_gender	0.06310.19	−0.0219−0.07	0.00830.03	0.05720.18	−0.3629−1.01	0.06480.19	0.09810.30	0.04860.15	−0.3600−0.64	−0.3524−1.32
cit_ideology	**0.05382.32	**0.05442.39	**0.05662.47	**0.05362.31	**0.05992.48	**0.05112.18	**0.04902.10	**0.05102.19	**0.05082.19	**0.05052.42
presi_vote	−0.0041−0.03	0.02250.18	0.01920.15	−0.0050−0.04	−0.0569−0.43	−0.0141−0.11	−0.0041−0.03	−0.0324−0.25	−0.0088−0.06	0.07730.63
Financial
r_revenue1pc	−0.0011−1.16	−0.0012−1.34	−0.0012−1.35		−0.0020−1.26	−0.0011−1.23	−0.0013−1.39	−0.0013−1.44	−0.0011−1.07	−0.0009−1.48
r_revenue2pc				−0.0005−1.20
r_debtpc	0.0000−0.06	0.0000−0.07	−0.0001−0.12	0.00000.02	−0.0001−0.26	−0.0001−0.15	−0.0001−0.20	−0.0001−0.14	−0.0001−0.23	−0.0002−0.81
r_deficitpc1	0.00121.41	0.00111.31	0.00121.39	0.00131.51	0.00181.52	0.00121.43	0.00121.39	0.00121.42	0.00181.52	**0.00142.11
sai_audit	**0.47812.06	*0.41401.84	*0.39991.79	**0.49072.11	*0.47011.93	**0.50962.18	**0.50302.15	*0.45491.93	*0.44531.79	*0.28631.87
Personal features
education	−0.0099−0.55	−0.0119−0.67	− 0.0120−0.68	−0.0098−0.54	−0.0040−0.22	−0.0077−0.43	−0.0105−0.58	−0.0079−0.44	−0.0034−0.18	−0.0028−0.17
age	**0.00962.00	**0.00942.00	**0.00982.08	**0.00962.01	***0.01402.80	**0.00972.04	**0.00972.02	**0.01022.15	**0.01242.54	**0.01022.35
gender	−0.1065−0.88	−0.0800−0.67	−0.0923−0.77	−0.1052−0.87	−0.1142−0.89	−0.1043−0.86	−0.1166−0.96	−0.1018−0.83	−0.1350−1.03	−0.1043−0.88
race_w	0.05580.38	0.01850.13	0.01770.12	0.05790.40	0.08150.52	0.08640.59	0.02620.18	0.04390.30	0.06840.43	0.08240.59
race_b	**−1.0148−2.41	**−1.0254−2.43	**−1.0228−2.42	**−1.0139−2.41	*−0.7548−1.64	**−0.9979−2.38	**−0.9872−2.29	**−0.9862−2.37	***−1.2042−2.57	**−0.9027−2.17
race_n	−0.0028−0.01	−0.0181−0.08	−0.0325−0.14	−0.0038−0.02	0.01330.05	0.02060.09	−0.0303−0.13	0.01520.07	0.02190.08	0.11650.51
race_o	0.41590.86	0.44440.97	0.43820.96	0.41120.85	0.69731.42	0.43570.89	0.34500.71	0.40420.81	0.56071.24	0.31520.77
Communication
often_news	*0.14371.92	0.11741.60	0.11751.59	*0.14451.93	0.13071.58	*0.13911.85	**0.14681.96	*0.13441.79	*0.14711.68	*0.13011.72
Income and labor
lnr_month_income	0.01480.15	0.02670.28	0.02650.28	0.01430.15		0.03920.40	0.02610.27	0.02960.31	0.06690.65	0.02240.25
lnr_income					0.26560.68
income_evol	**0.25122.56	**0.23832.45	**0.23592.43	***0.25172.57	**0.24812.42		**0.22512.31	**0.21852.24	**0.25942.43	**0.21082.23
person_econ_evol						0.13221.29
pubsec_job	−0.0714−0.22	−0.0756−0.23	−0.0965−0.30	−0.0708−0.22	0.06460.20	−0.0342−0.10	−0.0942−0.29	−0.0662−0.20	−0.0292−0.08	−0.2252−0.71
Participation and engagement
mun_meet_12m	***0.46452.68	***0.45772.63	***0.44952.60	***0.46962.68	***0.51602.63	**0.42742.45			**0.53162.48	***0.48382.66
engagcommun_12m							***0.31582.59
meet_polparty								*0.15171.71
Democracy quality
perform_demo	***0.31573.44	***0.30413.34	***0.30443.35	***0.31703.46	***0.32203.30	***0.31743.42	***0.32523.53	***0.31373.43	***0.27612.85	***0.30733.55
Trust others
trust_commun	0.10351.56	0.10841.63	*0.10901.64	0.10301.55	**0.14252.02	0.10791.63	0.09451.40	*0.11591.72	*0.13591.90	0.09451.49
R ²	0.3424	0.3430	0.3431	0.3425	0.3278	0.3381	0.3428	0.3403	0.3309
Probability > F	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000	0.0000
Maximum VIF	3.98	3.97	3.96	3.88	^c10.07	3.99	3.97	3.99
Heteroskedasticity test (B-C)	4.300.0380	4.390.0361	4.320.0377	4.380.0364	2.170.1404	3.650.0560	3.940.0471	3.680.0550
Homoskedasticity test (White)	610.030.3470	618.680.2521	606.490.3739	612.240.3240	547.950.3624	627.780.1854	609.300.3655	610.350.3650
Wald χ²(26)										***367.95
LR test vs. linear model: χ²(2) =										0.00

Note. Dependent variable: trust_muni. Dummy variable on electoral participation (1 = voted in last elections, 0 = otherwise) omitted because of collinearity (Table 1). Below each coefficient, t value is reported. All models include the following not reported variables:Intercept and dummy variables dept* (Honduran regions), which account for each municipality’s region, as a way to control for specific differences in the socioeconomic development among regions. Heteroskedasticity tests report Breusch−Pagan/Cook−Weisberg test for heteroskedasticity, and White’s test for homoskedasticity. VIF = variance inflation factor; LR = likelihood ratio.

Tests of endogeneity of num_polit_corrup r_revenue1pc: Wu−Hausman F test: 0.04308, F(2, 644) p value = .95784; Durbin−Wu−Hausman χ² test: 0.09176, χ²(2) p value = .95515. Tests of overidentifying restrictions: Sargan 2.101, χ²(4) p value = .7172; Basmann 1.972, χ²(4) p value = .7409. Instruments: num_polit_corrup, count_prob_corrup, corrup_bribe, lnr_income, person_econ_evol, race_w, race_b, race_n, race_o, and r_revenue2pc.

Multilevel (hierarchical) model is grouped (or nested) in two categories: 18 regions and 51 municipalities.

This VIF is on the critical value of 10 (Pop-Eleches, 2007). However, we could not drop neither lnr_income nor r_revenue1pc, which are the correlated variables that make VIF become 10.07. The former variable is the purpose of the robustness check. The latter variable captures municipal taxes, which is a control variable that the literature includes as independent variable. This econometric issue makes us take this regression cautiously. In any case, coefficients and significances differ very little compared with the remaining equations (only mayo_ideology loses significance).

10% significant. **5% significant. ***1% significant.

Robustness Check 8 controls for the possibility of endogeneity and reverse causality of two variables: corruption (num_polit_corrup) and municipal revenues (r_revenue1pc). This Model 8 is summarized in this equation:

y_{i} = γ + \sum β_{j} x_{j i} + β n u m_p o l i t_c o r r u p_{i} + β r_r e v e n u e 1 p c_{i} + c_{i} + ε_{i} .

(2)

Instrumented variables are num_polit_corrup_i and r_revenue1pc_i. Regarding corruption, there is a wealth of field evidence on the endogeneity and reverse causality of corruption (Mauro, 1998). If politicians appear to be free of corruption, they are more likely to receive people’s trust, and this can give politicians more flexibility (Alt et al., 2006). This flexibility could enhance possibilities of dishonest conducts. Similarly, there may simultaneously be corruption and high institutional trust and, in particular, confidence in politicians to fight corruption. Thus, there may be a lack of citizens’ supervision and control of corruption, because people trust the government to curb corruption (Bauhr, 2012).

Reverse causality of municipal revenues is clear, as without trust, citizens are less willing to pay taxes. Governments typically need more resources to accomplish its objectives. To claim those resources, governments have to improve their performance to earn citizens’ trust (Yang & Holzer, 2006).

The literature has not succeeded in setting relevant and valid instruments for corruption (Liu & Mikesell, 2014). Table 4 shows the instruments we have chosen, which are valid according to the Sargan and Basmann tests of overidentifying restrictions, that is, our structural model is correctly specified. In any case, the endogeneity test reported in Table 4 indicates that, in our sample, both variables, num_polit_corrup and r_revenue1pc, are not endogenous. Thus, we follow a very conservative approach, to ensure that our estimations are robust and meaningful.

The Robustness Check 9 presents a multilevel regression model (Montalvo, 2010). The model simultaneously takes into account both regional and municipal-level factors. Our sample has individual respondents, nested within municipalities, which are nested within regions. This multilevel equation produces accurate regression estimates, as a way to prove the reliability of the OLS regression.

Results and Discussion

Table 4 shows the main regression and the nine robustness checks. Different specifications of independent variables, controlling for endogeneity, and the multilevel model roughly yield the same coefficients and significances, which confirms the robustness of our analysis.

Transparency

Transparency significantly increases trust (tran_muni, Table 4). This result confirms the findings of Cooper and Yoder (2002), Pérez et al. (2010), H. Park and Blenkinsopp (2011), Bauhr and Grimes (2014), Wasike (2016), and Zhao and Hu (2017).

As an additional check, we present an alternative gauge of transparency. This transparency variable, based on documents posted by Honduran municipalities on an official web, is split into three dimensions (see Table 5). These web-based transparency variables have a lower impact on trust than the transparency perceived by LAPOP respondents. Therefore, the web-based transparency, though providing more detailed, technical raw data, has a lower impact on citizens’ trust. Besides, although web-based transparency variables correlate significantly with LAPOP tran_muni, the correlation coefficients are lower than .13. Both facts together lead us to conclude that it is not how many technical documents are made available to citizens, but, instead, the way these data are conveyed so that people get the relevant, essential information. In other words, a considerable number of technical documents posted on a website are less relevant for citizens than other processed, more understandable data. Here, either government departments or even journalists with specific training could analyze the data and summarize it in a way that is intelligible to the average citizen. According to Heald (2006), there is as a contrast between nominal transparency and effective transparency, meaning that there may be a gap between what is reported as happening and the underlying reality. Considering this inefficient scenario, we posit the Theory of Effective Transparency, which states that transparency is not achieved just by making excessive, over technical, raw data available to citizens. Quite the opposite, just a few key indicators of government performance would be better understood by citizens and would help them to hold the government accountable. Thus, the three basic requirements of government transparency to be met to comply with this newly formulated theory are as follows:

Transparency should convey relevant information to citizens, not data overload or over technical documents.

Transparency should be based on unbiased, fair information.

Data should be compared with other similar public institutions.

Table 5.

Correlations Between Alternative Variable of Transparency and Trust.

	globtran ^a	tran_acc ^a	tran_fin ^a	tran_bud ^a
tran_muni
Pearson correlation	.1102	.0735	.1108	.1253
Significance	.0020	.0397	.0016	.0004
n	784	784	808	808
trust_muni
Pearson correlation	.1851	.1624	.1842	.1224
Significance	.0000	.0000	.0000	.0004
n	801	801	824	824
globtran	tran_acc ^a + tran_fin ^a + tran_bud ^a
tran_acc ^a	Sum of annual accountability reports posted on web^a (operations, purchasing, revenues, expenditures, capital)
tran_fin ^a	Sum of annual financial reports posted on web^a (balance, net income, transfers, non-financial investments, financial investments, debt, and delinquency).
tran_bud ^a	Sum of annual budget reports published on web^a (actual and budgeted expenditures and revenues).

Note. These correlations take year 2016 because it is the first year of full enforcement of the website; 2016 is the last available year for Latin American Public Opinion Project.

https://portalunico.iaip.gob.hn/

Data overload is to some extent connected to the fact that citizens’ information-processing capacity cannot cope with information flows on a timescale relevant to achieving effective transparency (Heald, 2012). Similarly, government abuse of information release feeds media circulation and constructs barriers to effective transparency. Regarding the requirement of data to be unbiased, Heald (2012) advocates the role of public audit to ensure effective transparency through the validation of financial reporting. As far as the comparison claim is concerned, data on municipal deficit per capita, for example, should be compared with similar municipalities, so that tax payers can evaluate how the local government is managing tax funds.

Corruption

Our corruption variable, respecified in two ways, shows no impact on citizens’ trust (Table 4: main model and Checks 1 and 2). This result disagrees with the wealth of field evidence for the relation between corruption and trust (Canache & Allison, 2005; Espinal et al., 2006; Montalvo, 2010; Seligson, 2002). Specifically, we cannot confirm Córdova and Layton (2016), who show that people who have been asked for a bribe trust municipal governments less (corrupt_bribe). On one hand, our results would be in line with Bauhr and Grimes (2014), because if corruption is endemic, this may lead to public resignation, that is, strong belief that corruption is impossible to eradicate. Corruption is endemic in Honduras—the variable num_polit_corrup ranks above 3 in all years of the LAPOP time series (see Figure 1). This variable takes values 3 and 4 when respondents answer that half or more than half of politicians are corrupt, respectively. This situation, in financial terms, would imply that citizens have discounted corruption, and it does not impact on trust anymore. On the other hand, a less plausible explanation would be the coexistence of corruption with high institutional trust and, in particular, confidence in the ability of the political system to fight corruption (Bauhr, 2012). However, this would happen only in developed, low corrupt countries.

Municipal Service Quality

Municipal services quality has a clear, robust, and significant impact on trust (munserv_qual, Table 4). One point higher on the Likert-type scale of perceived service quality means around 0.45 greater trust. These results confirm previous literature (Bouckaert & van de Walle, 2003; Córdova & Layton, 2016; Montalvo, 2010; van de Walle & Bouckaert, 2003; Walle, 2003; Wu et al., 2017). In addition, according to F. L. Cook et al. (2010), governments’ weak performance does not impact on trust, but rather, it is a better-informed public that trusts more. Our data refute this assumption, as the extent to which people are informed (often_news) has a lower impact on trust than the perceived municipal service quality.

Remaining Determinants of Trust

Political variables that impact on trust are mayors’ ideology (mayo_ideology), absolute majority (majority), and citizens’ ideology (cit_ideology). We find a contradictory result. On one hand, conservative citizens significantly trust their municipal governments to a greater extent (cit_ideology). This finding confirms previous research, such as Stoyan, Niedzwiecki, Morgan, Hartlyn, and Espinal (2016) and Rudolph and Evans (2005). However, our result differs from that shown for another developing country, El Salvador, by Córdova and Layton (2016), who report a weak effect of political partisanship on trust. On the other, progressive mayors are trusted significantly more (mayo_ideology). This result appears to indicate that conservative people trust their municipal governments more, even when the mayor is progressive. To test this unexpected finding, we proceed with a double check. First, we recode cit_ideology, which ranges from 1 (left) to 10 (right), into a new dummy variable, dum_cit_ideology, which takes value 0 when the respondent is left wing (cit_ideology ≤ 5) and value 1 when right wing (cit_ideology ≥ 6). Later, we compute a new dummy variable called pol_agree, which takes value 0 when the mayor’s ideology and the interviewee’s ideology do not coincide (dum_cit_ideology ≠ mayo_ideology), and 1 when they do (dum_cit_ideology = mayo_ideology). Table 6 reports t test difference for the variable trust_muni conditional to pol_agree. The test clearly and significantly shows that political alignment leads to greater trust in the municipal government. This result was expected, as people trust more politicians who match their own political views (Wang, 2016). We ran an additional test to prove the contradictory finding explained above. We compute a new variable, L_whenmayor_R, with value 1 when a citizen is left wing and the mayor is right wing, and value 0 when the opposite. After running a t test difference for the variable trust_muni conditional to L_whenmayor_R, we conclude that our assumption is right, right wing citizens trust their left wing municipal governments more than when the opposite occurs, that is, left wing citizens in right wing municipalities (Table 6).

Table 6.

Impact of Political Alignment on Trust.

	Group		t student
Variable	pol_agree = 0M	pol_agree = 1M	Difference	Pr(T < t)
trust_muni	4.1106	4.1937	**−2.1482	0.0159
	Observation = 4,366	Observation = 4,548
	L_whenmayor_R = 0M	L_whenmayor_R = 1M	Difference	Pr(T < t)
trust_muni	4.3636	3.8753	***8.8220	0.0000
	Observation = 2,104	Observation = 2,262

Note. dum_cit_ideology takes value 0 when respondent is leftist (cit_ideology ≤ 5) and value 1 when rightist (cit_ideology ≥ 6). pol_agree takes value 0 when mayor’s ideology and interviewee’s ideology disagree (dum_cit_ideology ≠ mayo_ideology), and 1 when both ideologies coincide (dum_cit_ideology = mayo_ideology). L_whenmayor_R takes value 0 when citizen is rightist and mayor is leftist, and value 1 vice versa. In this table, we take the whole Latin American Public Opinion Project time span, that is, 2004-2016.

10% significant. **5% significant. ***1% significant.

With respect to the impact of the majority variable, the results indicate that the political strength of the governing party has a positive and statistically significant effect on trust. Furthermore, municipal financial indicators do not influence trust, which disagrees with Stoyan et al. (2016) and Espinal et al. (2006). However, we find sai_audit to be significant, that is, the fact that the municipal financial statements have been audited by SAI generates a positive and significant impact on trust. As previously mentioned, the law obliges a minimum of five open town councils per year. We believe that local officials inform citizens in these councils that financial reports were audited by the Supreme Audit Institution, which would generate greater trust among citizens.

Among the personal characteristics, the education coefficient shows that less educated people trust the municipal government more, although the coefficients are not statistically significant, which is in line with Sønderskov and Dinesen (2016), who find that education does not have a significant impact on trust. To provide an additional check on this relationship, we run a correlation between the variables education and trust_muni for the year of the regression (2014) and for the whole-time span (2004-2016): correlation = ***−.09 with n = 1,539 and correlation = ***−.05 with n = 10,777, respectively. Correlations are low, although significant, and negative, which, with the inherent limitations of univariate analyses, reinforces the idea that more educated citizens evaluate and question their municipalities’ performance to a higher extent and, accordingly, have less trust. Most of the literature from developing countries also find a negative impact of education on trust (Corbacho, Philipp, & Ruiz-Vega, 2015; Montalvo, 2010; Morris & Klesner, 2010; Seligson, 2002).

Age is positively related to trust, and the results are solid. Older people trust the government more. The average age of our sample is 39 years, which shows that most of the respondents have had more experiences or greater interaction with the government than younger ones. Age was found to increase trust in several studies on Latin American countries (Corbacho et al., 2015; Espinal et al., 2006; Montalvo, 2010). Regarding race, we find that Black people place less trust in government (race_b), which agrees with Wilkes (2015).

More informed people (often_news) appear to trust more, due to the impact of a smaller information gap on people’s trust (Welch et al., 2005). However, in some regressions, the coefficient is not significant, which indicates that the relationship is not strong.

Regarding the different specifications of the economic level (lnr_month_income, lnr_income, income_evol, person_econ_evol), only the evolution of the household’s economic income (income_evol) shows a positive and significant impact, that is, improvement in the interviewee’s economic situation leads to greater trust (Córdova & Layton, 2016; Montalvo, 2010). However, when economic evolution is respecified (Check 5), significance drops, so we should be cautious about this result.

Engagement and participation (mun_meet_12m, engagcommun_12m, and meet_polparty) show an unambiguous, robust, and significant impact on trust. It is clear that more participative citizens trust more, which agrees with the mainstream literature (Bauhr, 2017; Brehm & Rahn, 1997; Montalvo, 2010; Stoyan et al., 2016). As a practical implication, municipal governments should encourage citizens to get involved in municipal issues, as a way to increase trust. However, the above result contradicts the findings of Espinal et al. (2006) in a developing country (Dominican Republic), similar to the one considered in this research (Honduras). These authors find little evidence of a relationship between civic engagement and institutional trust.

Those who consider that democracy works well (perform_demo) report a greater level of trust in the government, that is, the perception of the health of the democratic system influences trust in local government, as Marlowe (2004) indicates. Finally, we find a weak relation between trusting others and trust in government (trust_commun). We cannot confirm Montalvo (2010) and Marlowe (2004), as our results do not hold under the robustness checks. Therefore, we take these findings with caution.

Conclusions and Further Research

This article empirically assesses the determinants of trust in Honduran local governments. In contrast to the mainstream literature, corruption in this country does not impact on citizens’ trust. Our guess is that as corruption is endemic in Honduras, there is firm belief that corruption is impossible to eradicate. In financial terms, citizens have discounted corruption, and it does not impact on trust anymore. However, we do find a clear influence of transparency on trust. Complementary web-based transparency variables, though providing more detailed, technical raw data, are found to have a lower impact on trust. This leads us to conclude that it is not the number of technical documents that are made available but rather how these data are conveyed that is important, and that people should get the key, relevant information (i.e., it is about “quality” rather than “quantity” of transparency). There is as a contrast between nominal transparency and effective transparency, so there may be a gap between what is reported as happening and the underlying reality (Heald, 2006). Accordingly, we posit the Theory of Effective Transparency, which sets three basic requirements for government transparency to be efficient:

Transparency should report relevant information, not data overload or over technical documents.

Transparency should be based on unbiased, fair information.

Data should be objective and comparable with counterparts.

As a point for practitioners, these requirements should be considered by governments when implementing transparency, so it becomes a useful tool for citizens to hold governments accountable. In this way, the gap between what is reported and what is actually happening would be narrowed.

Quality of municipal services strongly impacts on trust. We refute previous evidence that governments’ weak performance does not impact on trust; rather, it is better-informed citizens who trust more. Our data show that the extent to which people are informed has a lower impact on trust than the perceived quality of municipal services.

Regarding partisanship, we get a contradictory result. On one hand, conservative citizens trust their municipalities significantly more. On the other, progressive mayors are trusted significantly more. An additional test shows that right wing citizens trust their left wing municipal governments more than in the reverse case. With respect to the governing party’s strength (majority) and audited municipal financial statements, both generate a positive and significant impact on trust.

Older people trust governments more, while Black people trust them less. Engagement and participation show a robust and significant impact on trust. As a practical implication, municipalities should involve citizens in municipal issues to increase trust. Several policies could be adopted, such as participative budgets and more open council meetings with greater decision-making power. Finally, those who consider that democracy works well show a greater level of trust.

The main limitation of this research stems from LAPOP variable about transparency. Out of all LAPOP countries, this variable is available only in Honduras in 2014 and 2016. We hope transparency will be included in future LAPOP surveys for all countries.

Regarding future research, panel data could be used on future LAPOP datasets combined with our own web-based variable on transparency. Besides, as municipal services quality is a recurrent questionnaire item in LAPOP, again panel data analysis may disentangle which factors impact on this quality, for example, if greater expenditure per capita means, per se, higher quality municipal services.

Footnotes

Acknowledgements

We thank the Latin American Public Opinion Project (LAPOP) and its major supporters (the U.S. Agency for International Development, the Inter-American Development Bank, and Vanderbilt University) for making the data available.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Lorenzo Estrada

Francisco Bastida

Author Biographies

Lorenzo Estrada is a full-time professor of Business Education Department at the Francisco Morazán National Pedagogical University -UPNFM- (Honduras). He is a PhD student at the University of Murcia. He teaches Financial Accounting, Cost Accounting, Finance Management at the UPNFM. His current research focuses on the impact of transparency on government financial reporting and on citizens’ trust.

Francisco Bastida is a full time professor of Financial Economics and Accounting at the University of Murcia (Spain). He teaches management accounting at the University of Murcia and he has taught it at the University of Illinois at Chicago. He also teaches accounting information systems courses and courses in several master degrees at the University of Murcia. His current research focuses on the impact of transparency in government financial reporting from an international perspective, and on public finances. He has published, among others, in the following journals: Applied Economics, Public Administration, Public Administration Review, International Public Management Journal, Kyklos, Public Choice, American Review of Public Administration, Fiscal Studies, and Journal of Urban Planning and Development-ASCE. He is the Editorial Board Member of the journal International Journal of Economics and Accounting. He also works on consulting projects about management accounting implementation.

References

Alt

J. E.

Lassen

D. D.

Rose

(2006). The causes of fiscal transparency: Evidence from the U.S. states. IMF Staff Paper, 53, 30-57.

Araujo

J. F. F. E.

Tejedo-Romero

(2016). Women’s political representation and transparency in local governance. Local Government Studies, 42, 885-906. doi:10.1080/03003930.2016.1194266

Arbatli

Escolano

(2015). Fiscal transparency, fiscal performance and credit ratings. Fiscal Studies, 36, 237-270. doi:10.1111/1475-5890.12051

Bauhr

(2012). Need or greed corruption. In S. Holmberg & B. Rothstein (Eds.), Good government: The relevance of political science (pp. 68-86). Northampton, MA: Edward Elgar Publishing Limited.

Bauhr

(2017). Need or greed? Conditions for collective action against corruption. Governance, 30, 561-581. doi:10.1111/gove.12232

Bauhr

Grimes

(2014). Indignation or resignation: The implications of transparency for societal accountability. Governance, 27, 291-320. doi:10.1111/gove.12033

Bouckaert

van de Walle

(2003). Comparing measures of citizen trust and user satisfaction as indicators of “good governance”: Difficulties in linking trust and satisfaction indicators. International Review of Administrative Sciences, 69, 329-343. doi:10.1177/00208523030693003

Brehm

Rahn

(1997). Individual-level evidence for the causes and consequences of social capital. American Journal of Political Science, 41, 999-1023. doi:10.2307/2111684

Buehn

Dell’Anno

Schneider

(2017). Exploring the dark side of tax policy: An analysis of the interactions between fiscal illusion and the shadow economy. Empirical Economics, 54, 1609-1630. doi:10.1007/s00181-017-1277-7

10.

Canache

Allison

M. E.

(2005). Perceptions of political corruption in Latin American democracies. Latin American Politics and Society, 47, 91-111. doi:10.1111/j.1548-2456.2005.tb00320.x

11.

Cook

F. L.

Jacobs

L. R.

Kim

(2010). Trusting what you know: Information, knowledge, and confidence in social security. The Journal of Politics, 72, 397-412. doi:10.1017/S0022381610000034

12.

Cook

T. E.

Gronke

(2005). The skeptical American: Revisiting the meanings of trust in government and confidence in institutions. The Journal of Politics, 67, 784-803. doi:10.1111/j.1468-2508.2005.00339.x

13.

Cooper

T. L.

Yoder

D. E.

(2002). Public management ethics standards in a transnational world. Public Integrity, 4, 333-352. doi:10.1080/15580989.2002.11770926

14.

Corbacho

Philipp

Ruiz-Vega

(2015). Crime and erosion of trust: Evidence for Latin America. World Development, 70, 400-415.

15.

Córdova

Layton

M. L.

(2016). When is “delivering the goods” not good enough? World Politics, 68, 74-110. doi:10.1017/S0043887115000441

16.

Cucciniello

Porumbescu

G. A.

Grimmelikhuijsen

(2017). 25 years of transparency research: Evidence and future directions. Public Administration Review, 77, 32-44. doi:10.1111/puar.12685

17.

de Fine Licht

(2014). Policy area as a potential moderator of transparency effects: An experiment. Public Administration Review, 74, 361-371. doi:10.1111/puar.12194

18.

Emig

A. G.

Hesse

M. B.

Fisher

S. H.

III . (1996). Black-white differences in political efficacy, trust, and sociopolitical participation: A critique of the empowerment hypothesis. Urban Affairs Review, 32, 264-276. doi:10.1177/107808749603200206

19.

Espinal

Hartlyn

Kelly

J. M.

(2006). Performance still matters: Explaining trust in government in the Dominican Republic. Comparative Political Studies, 39, 200-223. doi:10.1177/0010414005281933

20.

Godefroidt

Langer

Meuleman

(2017). Developing political trust in a developing country: The impact of institutional and cultural factors on political trust in Ghana. Democratization, 24, 906-928. doi:10.1080/13510347.2016.1248416

21.

Gordon

M. T.

(2000). Public trust in government: The US media as an agent of accountability? International Review of Administrative Sciences, 66, 297-310. doi:10.1177/0020852300662006

22.

Grimmelikhuijsen

S. G.

Meijer

A. J.

(2014). Effects of transparency on the perceived trustworthiness of a government organization: Evidence from an online experiment. Journal of Public Administration Research and Theory, 24, 137-157. doi:10.1093/jopart/mus048

23.

Grimmelikhuijsen

S. G.

Porumbescu

Hong

(2013). The effect of transparency on trust in government: A cross-national comparative experiment. Public Administration Review, 73, 575-586. doi:10.1111/puar.12047

24.

Heald

D. A.

(2006). Varieties of transparency. In Hood

Heald

D. A.

(Eds.), Transparency: The key to better governance? Proceedings of the British academy 135 (pp. 25-43). Oxford, UK: Oxford University Press.

25.

Heald

D. A.

(2012). Why is transparency about public expenditure so elusive? International Review of Administrative Sciences, 78, 30-49.

26.

Howell

S. E.

Fagan

(1988). Race and trust in government: Testing the political reality model. Public Opinion Quarterly, 52, 343-350.

27.

Intawan

Nicholson

S. P.

(2018). My trust in government is implicit: Automatic trust in government and system support. The Journal of Politics, 80, 601-614. doi:10.1086/694785

28.

Lee

(2017). Why have policies often remained symbolic? Understanding the reasons for decoupling between policy and practice. Review of Policy Research, 34, 617-635. doi:10.1111/ropr.12241

29.

Liu

Mikesell

J. L.

(2014). The impact of public officials’ corruption on the size and allocation of U.S. state spending. Public Administration Review, 74, 346-359. doi:10.1111/puar.12212

30.

López-Arceiz

F. J.

Bellostas Pérezgrueso

A. J.

Rivera Torres

M. P.

(2017). Accessibility and transparency: Impact on social economy. Online Information Review, 41, 35-52. doi:10.1108/OIR-09-2015-0296

31.

Marlowe

(2004). Part of the solution or cogs in the system? The origins and consequences of trust in public administrators. Public Integrity, 6, 93-113. doi:10.1080/10999922.2004.11051251

32.

Mauro

(1998). Corruption and the composition of government expenditure. Journal of Public Economics, 69, 263-279. doi:10.1016/S0047-2727(98)00025-5

33.

Maxwell

(2010). Trust in government among British Muslims: The importance of migration status. Political Behavior, 32, 89-109. doi:10.1007/s11109-009-9093-1

34.

Michener

(2015). Assessing freedom of information in Latin America a decade later: Illuminating a transparency causal mechanism. Latin American Politics and Society, 57(3), 77-99. doi:10.1111/j.1548-2456.2015.00275.x

35.

Montalvo

(2010). Understanding trust in municipal governments (Americas Barometer Insights Series No. 35). Retrieved from https://www.vanderbilt.edu/lapop/insights/I0835en.pdf

36.

Morris

S. D.

Klesner

J. L.

(2010). Corruption and trust: Theoretical considerations and evidence from Mexico. Comparative Political Studies, 43, 1258-1285. doi:10.1177/0010414010369072

37.

Park

Blenkinsopp

(2011). The roles of transparency and trust in the relationship between corruption and citizen satisfaction. International Review of Administrative Sciences, 77, 254-274. doi:10.1177/0020852311399230

38.

Park

M. J.

Choi

Kim

S. K.

Rho

J. J.

(2015). Trust in government’s social media service and citizen’s patronage behavior. Telematics and Informatics, 32, 629-641. doi:10.1016/j.tele.2015.02.006

39.

Pérez

M. D. C. C.

Bolívar

M. P. R.

Hernández

A. M. L.

(2010). Transparency and e-government in developing countries: The case of Latin-American municipalities. In Citizens and e-government: Evaluating policy and management (pp. 158-183). IGI Global. doi:10.4018/978-1-61520-931-6.ch009

40.

Piotrowski

S. J.

van Ryzin

G. G.

(2007). Citizen attitudes toward transparency in local government. The American Review of Public Administration, 37, 306-323. doi:10.1177/0275074006296777

41.

Pop-Eleches

(2007). Historical legacies and post-communist regime change. The Journal of Politics, 69, 908-926. doi:10.1111/j.1468-2508.2007.00598.x

42.

Porumbescu

G. A.

(2015). Using transparency to enhance responsiveness and trust in local government. State and Local Government Review, 47, 205-213. doi:10.1177/0160323X15599427

43.

Porumbescu

G. A.

(2016). Placing the effect? Gleaning Insights into the relationship between citizens’ use of e-government and trust in government. Public Management Review, 18, 1504-1535. doi:10.1080/14719037.2015.1122827

44.

Rudolph

T. J.

Evans

(2005). Political trust, ideology, and public support for government spending. American Journal of Political Science, 49, 660-671. doi:10.2307/3647738

45.

Seligson

M. A.

(2002). The impact of corruption on regime legitimacy: A comparative study of four Latin American countries. The Journal of Politics, 64, 408-433. doi:10.1111/1468-2508.00132

46.

Sønderskov

K. M.

Dinesen

P. T.

(2016). Trusting the state, trusting each other? The effect of institutional trust on social trust. Political Behavior, 38, 179-202. doi:10.1007/s11109-015-9322-8

47.

Steves

(2001). Regional integration and democratic consolidation in the Southern Cone of Latin America. Democratization, 8(3), 75-100. doi:10.1080/714000210

48.

Stoyan

A. T.

Niedzwiecki

Morgan

Hartlyn

Espinal

(2016). Trust in government institutions: The effects of performance and participation in the Dominican Republic and Haiti. International Political Science Review, 37, 18-35. doi:10.1177/0192512114534703

49.

van de Walle

Bouckaert

(2003). Public service performance and trust in government: The problem of causality. International Journal of Public Administration, 26, 891-913. doi:10.1081/PAD-120019352

50.

Wang

C.-H.

(2016). Government performance, corruption, and political trust in East Asia. Social Science Quarterly, 97, 211-231. doi:10.1111/ssqu.12223

51.

Wasike

(2016). FoIA in the age of “Open. Gov”: An analysis of the performance of the Freedom of Information Act under the Obama and Bush administrations. Government Information Quarterly, 33, 417-426. doi:10.1016/j.giq.2016.05.001

52.

Welch

E. W.

Hinnant

C. C.

Moon

J. M.

(2005). Linking citizen satisfaction with e-government and trust in government. Journal of Public Administration Research and Theory, 15, 371-391. doi:10.1093/jopart/mui021

53.

Whiteley

Clarke

H. D.

Sanders

Stewart

(2016). Why do voters lose trust in governments? Public perceptions of government honesty and trustworthiness in Britain 2000-2013. The British Journal of Politics and International Relations, 18, 234-254. doi:10.1111/1467-856X.12073

54.

Wilkes

(2015). We trust in government, just not in yours: Race, partisanship, and political trust, 1958-2012. Social Science Research, 49, 356-371. doi:10.1016/j.ssresearch.2014.08.008

55.

Worthy

(2010). More open but not more trusted? The effect of the Freedom of Information Act 2000 on the United Kingdom central government. Governance, 23, 561-582. doi:10.1111/j.1468-0491.2010.01498.x

56.

(2017). Government transparency and perceived social equity: Assessing the moderating effect of citizen trust in China. Administration & Society, 49, 882-906. doi:10.1177/0095399716685799

57.

Yang

Holzer

(2006). The performance-trust link: Implications for performance measurement. Public Administration Review, 66, 114-126. doi:10.1111/j.1540-6210.2006.00560.x

58.

Zhao

(2017). Determinants of public trust in government: Empirical evidence from urban China. International Review of Administrative Sciences, 83, 358-377. doi:10.1177/0020852315582136