Transparency International’s Global Corruption Barometer (GCB) for Latin America and the Caribbean 20191 offers an early glimpse into Mexico’s anticorruption efforts under President Andrés Manuel López Obrador. Compared to the results from the GCB two years ago, and to other Latin American countries today, Mexico performed rather well.
The GCB, produced roughly every two years, measures different aspects of corruption based on public opinion polls (in contrast to TI’s better known Corruption Perception Index—CPI—which reflects expert opinions and produces a sole indicator of corruption). Normally, the GCB gauges perceptions of corruption generally and across institutions, perceived changes in levels of corruption, evaluations of government anticorruption efforts, and participation in corruption or victimization rates, among other areas. The 2019 version encompasses 18 Latin American countries and more than 18,000 interviews. For Mexico, the nationally representative sample included 1,000 interviews conducted between February 26 and March 9, 2019.
Results and Comparisons
Among the more positive results, 21% of respondents in Mexico expressed the view that corruption had decreased over the prior 12 months, while 44% said it had increased. While only a minority of responses point to any sort of decline in corruption, this is better than the regional average of 16% and 53%, respectively. More importantly, the percentage who said corruption had decreased in Mexico climbed 15 percentage points since 2017, while those pointing to an increase fell by 17 points.
This positive change in assessment of corruption is also reflected in the public’s evaluation of the “current government's” job in fighting corruption. Here, 61% rated the job as “good” (52% as “fairly well” and 9% as “very well”), compared to just 24% in 2017. In fact, this time a minority, 36%, deemed the effort “bad,” much lower than the 57% for the region and the 61% in Mexico two years earlier.
The GCB’s gauge of the public’s level of participation in corrupt acts (i.e., victimization rates) centers on asking respondents whether they accessed any of six services in the prior 12 months (police, courts, health care, schools, government identity document agencies, and utilities), and whether they paid a bribe while doing so. Overall, 34% of those accessing these services in Mexico reported paying a bribe. While this figure still remains much higher than the 21% of respondents in the region who paid bribes, it nonetheless represents a drop of 17 percentage points since 2017, a notable improvement. The changes, however, were neither uniform across the services, nor all in a positive direction. Results point to declines in bribe payments within public health clinics (from 39% in 2017 to 16% in 2019), public schools (from 33% to 19%), and ID-issuing agencies (from 37% to 25%), yet significant increases in bribes to police officials (from 30% in 2017 to 52%) and judges (7% to 35%).
While such trends are encouraging overall, others are less so. About 90% of respondents deemed corruption a “big problem,” for instance. Perceptions of corruption among institutions (those in which “most” or “all people” within the institution are considered corrupt) show few positive changes and some rather disturbing increases, particularly within the office of the president (from 51% in 2017 to 63% in 2019), Congress (56% to 65%), police (63% to 69%), and courts (50% to 58%), with negligible changes in the other areas listed in the survey (government officials, local government officials, religious leaders, and business executives).
In addition, 56% of respondents believed that people were “frequently” or “very frequently” offered bribes in exchange for votes, while 50% claimed to have experienced some form of vote buying over the past five years. Roughly 48% said voter intimidation occurred “frequently” or “very frequently,” while 25% reported actually experiencing voter intimidation over the past five years. Further, 55% believed that sextortion happens very frequently or often, while 20% reported having experience or knowledge of “a public official implying either openly or suggestively…that they will grant a government benefit in exchange for a sexual favor.” While 79% agreed that citizens can make a difference in the fight against corruption, 77% believed that people who report corruption risk retaliation, and 62% thought that if they were to report a corrupt act, it is unlikely that appropriate action would be taken against the public official.
Though many of the indicators point in the right direction, it is not entirely clear what they tell us. That the surveys were conducted roughly eight months after the historic July 2018 election and three to four months after the new government took power raises questions about whether to credit the new government for the positive changes or blame them for the negative ones. Much depends on: a) whether and which assessments, perceptions, and experiences relate to the current government and which to the prior government, especially since many of the questions reference a particular time frame (i.e., past 12 months or past year); b) the extent to which the views reflect popular expectations rather than assessments of current circumstances; and c) whether expectations or even approval of the current government and their policies may alter the memory of past experiences.
The heightened view that corruption has decreased references a period—past 12 months—before the election and transition. Yet it is hard to imagine that such positive results could relate to the Peña-Nieto government given the public’s documented assessments of corruption prior to the election. At the same time, it is somewhat difficult to attribute this view to real changes put in place by the current government given the brief period of time the administration had been in power at the time of the surveys. This may suggest, in part at least, the influence of expectation and anticipation.
However, no confusion surrounds the evaluation of the government’s efforts to fight corruption, since the question refers specifically to the “current government.” Still, it is unclear whether the positive assessment and shift in opinion reflects actual assessment of the government’s efforts to fight corruption or are rooted more in expectations based on promises and the announcement of high-profile initiatives.
And yet, changes in the public’s level of participation in corruption—a question based on actual experience but referencing the prior 12-month period—suggest either real changes on the ground during the brief window since the administration has taken power or perhaps even altered memory of corruption under the Peña-Nieto administration. Even if there have been real changes in the actual delivery of services, the question then emerges as to whether this is due to the tone, for example, of policies or pronouncements by the government that have induced a new spirit of honest government among lower-level officials (less corruption), or whether the changes reflect actual reforms to the institutional mechanics of service delivery (lowering incentives for corruption and individual discretion, increasing risks, etc.). However, if any of these alternative hypotheses are correct, then what accounts for the dramatic increases vis-à-vis police and judges? Were these areas immune from the influences of the tone and/or policies of the new government or were they simply resistant to the changes?
Perhaps even more confounding are the results on perceptions of corruption among institutions. Not only were few positive changes registered, but the increase in perceived corruption within the presidency is wholly inconsistent with the idea that expectations of change and support and approval of anticorruption initiatives under López Obrador have led to changes in experiences. Here, it is important to note that in contrast to the services referenced in the participation question, the public has very little interaction with the presidency, members of Congress, or judges. In other words, these perceptions rely less on actual experience and more on perception. While respondents may be drawing on views associated with the prior government, these results nevertheless complicate our interpretation.
In the end, despite difficulties in explaining the results or their meaning, there is no denying that seemingly for the first time, TI’s polling provided some good news on Mexico’s fight against corruption. Many of the shifts point in the right direction, and the government understandably wishes to tout the results as signs of early success and a validation of their strategy. Other evidence bolsters these tendencies, such as an increase in public denunciations of corruption, increased investigations by the government, or new collective action initiatives by the Public Affairs Ministry.2 Still, the country has a long way to go, and the GCB points to those problem areas and challenges —immense public concern about corruption, high levels of perceived and actual vote buying and voter intimidation, high levels of sextortion, fears of reporting corruption, among others. To the extent the positive shifts stem from actual changes implemented by the Lopez Obrador administration, then clearly the government needs to continue to strengthen those initiatives. To the extent the shifts reflect public expectations and the hopes tied to a new government, the government must meet and even mobilize those expectations. Either way, hopefully, these positive trends will continue.
1. See Coralie Pring and Jon Vrushi, Global Corruption Barometer, Latin American and the Caribbean 2019: Citizens’ Views and Experiences of Corruption (Berlin: Transparency International, 2019), www.transparency.org/ files/content/pages/2019_GCB_LatinAmerica_ Caribbean_Full_Report.pdf.
2. See Secretaría de la Función Pública, 1 Informe de Labores (Mexico City: Secretaría de la Función Pública, 2019), www.gob.mx/ sfp/articulos/85305.
This material may be quoted or reproduced without prior permission, provided appropriate credit is given to the author and Rice University’s Baker Institute for Public Policy. The views expressed herein are those of the individual author(s), and do not necessarily represent the views of Rice University’s Baker Institute for Public Policy.