Lucas Brook and Dr. Darren Hawkins, Political Science Department
Peru has long been afflicted by corruption and political malpractice. As a result, Peruvian citizens have become disenchanted with their government. In recent years, the Peruvian government has employed internet transparency campaigns in an attempt to improve these sentiments. But what if Peruvians feel they cannot trust the information? Does the source of the information have an influence on their attitudes? In July 2013, I conducted a field experiment in Lima, Peru aimed at answering this question.
Along with my faculty mentor and several other student researchers conducting their own experiments, I travelled to Peru’s capital city of Lima. As locations for our experiments, we selected public areas with a wide variety of people, such as parks, shopping malls, and busy commercial streets. We hired and trained Peruvian nationals to conduct the surveys, mostly 20- something students studying at a variety of technical institutes. Each of the enumerators was given a tablet and assigned a randomized treatment and a randomized selection number. The enumerator then approached the nth person that passed by according to the selection number. If the enumerator was rejected (the person they approached chose not to participate), s/he would then approach the next nth person. Enumerators repeated this process until they were successful in administering a survey. At the end of the treatment video and survey, the subjects were offered three Peruvian soles (the rough equivalent of one US dollar). The enumerators were then assigned a new random treatment and random selection number, repeating the process.
We administered the treatments in the form of three to four minute videos that displayed information from the transparency portals. Subjects (Peruvian citizens) viewed these videos on tablet computers and then answered 34 questions posed by hired surveyors. Twenty-five of these questions probed political attitudes while the remaining ten questions covered demographic information. We selected all questions from the most recent editions of the AmericasBarometer survey and Transparency International survey. These surveys are some of the most well-known and well-respected efforts to collect data on political attitudes, thus enabling us to utilize established, field-tested questions and to engage in comparisons with their findings.
My experiment is composed of three treatment groups and one control group. In each treatment, as well as the control, subjects were shown videos providing the same information on the municipality of Lima. The information shown in each video includes Lima’s budget per capita, illiteracy rate, infant mortality rate, percentage of households with access to water and sewage, a rating of the government’s capacity to execute plans for investments, information on whether the government published required details on the participative budget process, the rate of political participation of women, and information on the local government’s accountability to the national government. Where applicable, the statistics for Lima are compared with the national average. Lima generally compares very favorably to other areas of the country on these indicators. The videos for the treatment and control groups are identical except for the logo that is displayed as the source of the information and the narration for the video that calls attention to that logo. Treatment 1 attributes the information to the Comptroller General of Peru, an office within the government of Peru. Treatment 2 attributes the information to USAID, the foreign aid agency of the U.S. government. Treatment 3 attributes the information to Peru Prodecentralización, a Peruvian NGO. The control does not provide a source of the information. The information shown originated from a website jointly sponsored by these three entities. The demographic information we collected suggests that we surveyed a wide range of ages, education levels, incomes, and so forth.
Analysis of this survey data has shown that attributing the information to either the Peruvian government itself or the Peruvian national NGO had no affect on subjects’ political attitudes. Demonstrating USAID as the source of information, however, altered political attitudes in a number of ways. Perhaps most remarkably, it increased trust in the municipality and it increased the extent to which people say they value democracy. With respect to government performance, it increased President Ollanta Humala’s approval rating, increased satisfaction with government transparency, and increased pride in Peru’s political institutions.
These effects can be put in perspective by comparing the results we found in Peru with those from other countries in the AmericasBarometer survey. One of the most important changes occurred on a question asking respondents’ level of agreement with the following statement: “While democracy may have problems, it is better than any other form of government.” While the control group ranked 23rd out of the 26 countries in the AmericasBarometer survey, the treatment group with USAID as the source displayed levels of support for democracy at levels that would move Peru up to 6th place, at the level of Costa Rica and close to Canada and the United States. Hence, an endorsement of the transparency information from USAID raises support for democracy in Peru from near the bottom of all Western Hemisphere countries to levels seen frequently in North America. This same treatment group displayed levels of trust in the municipality that would rank 11th out of the 26 countries while the control was 25th. This heightened level of trust is even higher than that of the United States. These results show, rather conclusively, that U.S. endorsement of transparency information has real and substantial effects on political attitudes in Peru.
These results also suggest that the Peruvian government has a serious credibility problem. If the government could resolve that credibility problem, trust and support for the system would likely increase, resulting in increased civic engagement. This experiment has shown that a simple U.S. agency logo on transparency portals significantly changes Peruvian perceptions of their government and even democracy itself. The U.S. government could play a crucial role in this legitimization process by offering its endorsement of the transparency information that the Peruvian government is already providing to its citizenry