Publicizing malfeasance:

Transcription

1 Publicizing malfeasance: When media facilitates electoral accountability in Mexico Horacio Larreguy, John Marshall and James Snyder Harvard University May 1, 2015

2 Introduction Elections are key for political accountability but, for elections to function well, voters must be adequately informed. The mass media may play an essential role in informing voters. Particular relevance in developing contexts: greater malfeasance, less information. This paper: When does the broadcast media facilitate electoral sanctioning of malfeasant behavior such as corruption and diverting funds away from the projects for which they are earmarked in Mexico?

3 Challenges, and closest work and its limitations In developing contexts, there is little causal evidence establishing whether information provided by the media helps voters to sanction politicians at the polling booth, and what kinds of media market structures induce such sanctioning. Identification is challenging since identification requires exogenous variation in both malfeasance revelations, and voter access to media coverage. Ferraz and Finan (2008) find that incumbent mayors in Brazil who are randomly revealed to be corrupt just before an election suffer more at the polls in municipalities with more AM radio stations but lack randomness in AM radio coverage.

4 Related work and its limitations Work randomizing access to information about a voter s incumbent representative provides mixed evidence. Humphreys and Weinstein (2012) conduct survey experiments that indicate that Ugandan voters are strongly receptive to new information about the performance of their incumbent. However, Banerjee et al. (2011) and de Figueiredo, Hidalgo and Kasahara (2013) find no sanctioning of incumbents in India and Brazil. Chong et al. (2014) find the opposite effect in Mexico because the drop in turnout hurts more the opposition than the incumbent Moreover, they cannot separate the effects of such information from its mode of transmission since they lack exogenous variation in malfeasance revelations.

5 Literature on media market structure and its limitations There is also limited evidence identifying how the structure of the media market affects the electoral sanctioning of malfeasant politicians in developing contexts. No empirical work on the role of market concentration in developing contexts despite there are many theoretical reasons for it to matter. The literature has largely focused on identifying the electoral implications of access to a new media station or media market offering distinct content as opposed to more media stations with similar content.

6 Question and empirical strategy We identify electoral effects of local media stations, which emit within an electoral precinct s municipality, publicizing revelations of mayoral malfeasance in Mexico. We exploit two sources of plausibly enxogenous variation. As in Ferraz and Finan (2008), we leverage variation in the timing of the release of municipal audit reports around elections. We also leverage within-neighbor variation in commercial quality radio and television signals that differ across urban electoral precincts from within the same municipality because of plausibly exogenous reasons.

7 Audit reports Audit reports on the use of the Municipal Fund for Social Infrastructure (FISM). FISM represents about 25 % of mayors annual budgets and provides mayors with funds for infrastructural projects required to benefit citizens in extreme poverty. Reports reveal FISM money spent in an unauthorized manner and the share spent on projects not benefiting the poor. Both figures represent malfeasance.

8 Preview of baseline results Mayors cannot seek re-election (due to term limits), but in Mexico it appears that voters primarily choose between parties. Each additional local media station substantially increases voter punishment of the party of mayors revealed to be either corrupt or neglectful of the poor. Voters reward good performance in office. An additional local television station has substantially larger effects on electoral sanctioning than an additional radio station. (According to surveys, TV is the most important source of political information in Mexico.)

9 Preview of results on market structure Non-local media stations do not affect the vote share of the party of malfeasance mayors. Non-local media stations crowd out local media stations. Consistent with hypothesis that non-local media stations not only do not cover local news but also provide alternative, entertaining content that diverts voter s attention. The effect of a local media station is larger when the station s audience is principally from inside the municipality. Consistent with hypothesis that local news coverage increases with audience demand for such information. Findings are robust to a series of sensitivity checks.

10 Roadmap 1. Why the intensity of media coverage matters. 2. Background. 3. Data. 4. Empirical design. 5. Empirical Results. 6. Conclusion.

11 Roadmap 1. Why the intensity of media coverage matters. 2. Background. 3. Data. 4. Empirical design. 5. Empirical Results. 6. Conclusion.

12 Why the intensity of media coverage matters Empirical work on the political importance of access to a new media station or media market offering distinct content, both in terms of the focus of their news and their ideological stance. Despite significant theoretical interesting the role of market concentration, little is known about the marginal effect of an additional media station providing similar content for electoral accountability. We suggest that access to more media stations providing similar content may also play an essential role in informing voters about the behavior of their politicians in office.

13 Limiting media capture and bias, and multiple signals Besley and Prat (2006) show that political capture of media stations decreases with the number of stations in the market. Without political capture, media may bias its news coverage toward the priors of their consumers in order to increase their reputation as a high-quality news outlet (Gentzkow and Shapiro 2006). However, greater competition in the media market might restrict this behavior ensuring that voters receive truthful signals about political performance. Moreover, receiving the same information from multiple sources should increase voter certainty regarding incumbent performance (Gentzkow, Shapiro and Sinkinson 2014)

14 Informing voters and new voters More media can also expose new listeners or viewers. Prat and Stromberg (2005) show that the introduction of commercial television disproportionately attracts relatively uninformed voters, causing them to become more politically knowledgeable and increase their political participation. Similarly, the introduction of television stations with different partisan biases can attract new audiences. We offer suggestive evidence that the availability of additional local media station increases news consumption in our context of Mexico. More media reduces the likelihood that listeners or viewers do not get local news because media outlets also focus on other political markets (Snyder and Stromberg 2010).

15 Roadmap 1. Why the intensity of media coverage matters 2. Background. 3. Data. 4. Empirical design. 5. Empirical Results. 6. Conclusion.

16 Mayors in Mexico and the FISM Mexico s 2,435 municipalities account for 20% total government spending. Mayors manage local infrastructure and public services. The Municipal Fund for Social Infrastructure (FISM) represents 25% of a municipality s total resources. The FISM is exclusively for the funding of public works, basic social actions, and investments that directly benefit the socially disadvantaged population living in extreme poverty.

17 Audits in Mexico Audit of federal funds by Mexico s Federal Auditor s Office (ASF) since 1999 in response to widespread concerns regarding the mismanagement of public resources. The list of municipalities to be audited in a given year is announced the year after the spending occurred. Between 2007 and 2012, 14% of Mexican municipalities were audited at least once, with about 120 municipalities being audited each year. Although the municipalities that have been audited are not randomly chosen, the timing of an audit is essentially random. Correspondence with the ASF confirms that the selection of municipalities for audit does not depend upon the electoral cycle or the government s political identity. Our balance tests confirm we capture plausibly exogenous variation in the timing of audits.

18 Release of audit reports Audit reports are publicly released two years after the spending actually occurred, when they are presented in Congress in late February and posted on ASF s website. Jan. Dec. Jan. Dec. Feb. July FISM funds spent FISM funds audited Audits released Elections conducted Jan. Dec. Jan. July Dec. Feb. FISM funds spent FISM funds audited Elections conducted Audits released

19 Audit reports (Cont.) Audits highlight: Percentage of funds spent in an unauthorized way (e.g. diverting resources, personal largess). Percentage of funds not spent on the poor (e.g. public works in richer areas, diverting funds for electoral campaigns). Figure: This mayor wrote a check for 69,130 USD to his wife.

20 Ajalpan report from 2008

21 Electoral punishment for malfeasance The ASF can recommend various sanctions on culpable officials but these rarely occur. The largest punishment may be electoral. While mayors were elected (until recently) for non-renewable terms, there are good reasons to believe that a mayor s political party may be punished by voters at the next election (Chong et al. 2014). Mexico s strong party system and little information about local candidates ensure that party labels play a key role in voting decisions. Also, the top-down internal structure of Mexican parties at the state level ensures that within-party candidate choice is highly correlated.

22 Media in Mexico Voter are uninformed about mayoral responsibilities and the use of public funds. Voters learn about public spending primarily through media coverage. According to Castañeda Sabido (2011), 83% of individuals report that they receive information about malfeasance in the management of public resources through media, and 61% regard such information as reliable.

23 Media in Mexico (Cont.) Radio and television stations provide about 2 hours of news coverage a day. While the majority of entertainment content is common across media stations within Mexico s large media networks, news programming is typically local. Results in the Online Appendix suggest that the likelihood that a voter follows the news increases with the availability of local media. We expect that television may be the most the important media source for electoral accountability. 83 % of Mexican households respondents gather political information from television. 41 % of Mexican household gather political information from radio.

24 Coverage of audits reports The release of municipal audit report results each February is widely covered. News reports generally cover mayors within the local vicinity, and typically focus on cases of corruption and mayors not spending FISM funds on projects targeting the poor. Most reports accurately cite the proportions of unauthorized spending and spending on projects not targeted at the poor, and many dig deeper to describe the nature of the malfeasance.

25 Roadmap 1. Why the intensity of media coverage matters. 2. Background. 3. Data. 4. Empirical design. 5. Empirical Results. 6. Conclusion.

26 Election and audit data Main outcome is change in incumbent party vote share. Municipal election data for 2004 to 2012 on electoral precincts (within municipalities) from various state electoral institutes. Proportion of funds spent in an unauthorized manner and the proportion of funds not spent on projects benefiting the poor from audit report data for 2005 to There are 742 municipal audits covering 351 unique municipalities but only 470 reports from 321 different municipalities were released in an election year or the year after.

27 Audit report outcomes

28 Media data Following a major media reform in 2007 the IFE required that every AM and FM radio station and every television station in the country provide signal coverage data. We have the location and coverage of each of the 852 AM, 1097 FM and 1255 television stations. Our principal measure of local media coverage is the total number of AM, FM or television stations covering a given electoral precinct that broadcast from within the precinct s municipality. We also computed the number of media stations that cover a precinct but transmit from outside their municipality.

29 AM radio station coverage (4.4 per precinct)

30 FM radio station coverage (5.4 per precinct)

31 TV station coverage (2.4 per precinct)

32 Roadmap 1. Why the intensity of media coverage matters. 2. Background. 3. Data. 4. Empirical design. 5. Empirical Results. 6. Conclusion.

33 Identification strategy The difference-in-differences (DD) component of our design rests upon exogenous variation in the timing of audit report releases. The independent ASF s procedure for determining which municipalities will be audited in a given year is apolitical: the timing of elections does not feature in their selection rule. Compare 321 municipality (47,938 precinct-election) reports released just before and after elections. Table 1 in the paper shows only one statistically significant difference in 49 political, demographic, media, and economic characteristics.

34 Difference-in-differences incumbent vote p,m,t = β 1 audit m,t + β 2 audit outcome Q3 m,t +β 3 audit outcome Q4 m,t ( ) +β 4 audit m,t audit outcome Q3 m,t ( ) +β 5 audit m,t audit outcome Q4 m,t +X p,m,t γ + ε p,m,t, Q3 and Q4 are indicators for municipalities in the third and fourth quartiles of the distributions of corrupt or neglectful mayors. We weight each observation by the number of registered voters Cluster by municipality-year.

35 Effect of local media We cannot simply look at the interaction of our DD estimates and the number of local media that cover every precinct. In the Online Appendix, the number of local media stations received by an electoral precinct in our Mexican sample is significantly correlated with almost every census characteristic. We then compare neighboring electoral precincts from within a given municipality that differ in the total number of local AM radio, FM radio and television stations that they are covered by. Our identifying assumption is that neighboring precincts differ only in their local media coverage.

36 Effect of local media (Cont.) Restricting attention to within-neighbor variation removes a wide variety of potential confounds. To maximize the plausibility of our identification strategy, we focus on relatively urban precincts (with at most an area of 10 km2) to avoid rural - urban comparisons. Table 2 tests for balance and finds no statistically significant difference in 36 variables.

37 Effect of local media (Cont.) We estimate the following triple-difference (DDD) specification using our neighboring precincts sample: Y p,k,m,t = β 1 audit m,t + β 2 outcome m,t + β 3 media p,m ( ) +β 4 audit m,t outcome Q3 m,t ( ) +β 5 audit m,t outcome Q4 m,t ( ) +β 6 audit m,t media p,m ( ) +β 7 audit m,t outcome Q3 m,t media p,m ( ) +β 8 audit m,t outcome Q4 m,t media p,m +ξ k + ε p,k,m,t, where ξ k are neighbor fixed-effects.

38 Roadmap 1. Why the intensity of media coverage matters. 2. Background. 3. Data. 4. Empirical design. 5. Empirical Results. 6. Conclusion.

39 Difference-in-differences results Change in incumbent Incumbent party party vote share re-elected (1) (2) (3) (4) Audit (0.019) (0.021) (0.122) (0.113) Corrupt Q (0.027) (0.159) Audit Corrupt Q (0.037) (0.191) Corrupt Q (0.036) (0.153) Audit Corrupt Q (0.045) (0.208) Not poor Q (0.022) (0.156) Audit Not poor Q (0.035) (0.196) Not poor Q (0.041) (0.162) Audit Not poor Q (0.047) (0.203) Observations 47,938 47,938 47,938 47,938 Outcome mean

40 Local media estimates Change in incumbent party vote share (1) (2) (3) (4) Audit (0.054) (0.050) (0.069) (0.080) Audit Local media 0.007*** 0.006* 0.008*** (0.002) (0.003) (0.002) (0.003) Audit Corrupt Q3 Local media * ** (0.004) (0.003) Audit Corrupt Q ** (0.074) (0.095) Audit Corrupt Q4 Local media * ** (0.004) (0.004) Audit Not poor Q (0.098) (0.139) Audit Not poor Q3 Local media * (0.006) (0.006) Audit Not poor Q (0.076) (0.101) Audit Not poor Q4 Local media *** *** (0.004) (0.004) Non-local media interaction Observations 17,312 17,312 17,312 17,312 Local mean Local std. dev

41 Local media estimates (Cont.) A standard deviation increase in the number of media stations, which entails 10.6 more media stations, thus reduces the vote share of an incumbent revealed to be corrupt by about 7.5 percentage points. This represents half of the average municipal victory margin (in the sample) of 15 percent. A standard deviation increase in the number of local media stations thus entails an 13 percentage point decrease in the vote share of the most neglectful mayors if their behavior is revealed before an election. This represents a 25% reduction in their precinct vote share. Clean incumbent parties are rewarded for their good performance in the presence of local media stations.

42 Separate FM and TV estimates Change in incumbent party vote share (1) (2) (3) (4) Audit Local FM media (0.004) (0.004) Audit Corrupt Q3 Local FM media (0.006) Audit Corrupt Q4 Local FM media (0.007) Audit Not poor Q3 Local FM media 0.013** (0.007) Audit Not poor Q4 Local FM media (0.005) Audit Local TV media 0.037*** 0.037*** (0.005) (0.008) Audit Corrupt Q3 Local TV media *** (0.008) Audit Corrupt Q4 Local TV media *** (0.011) Audit Not poor Q3 Local TV media ** (0.015) Audit Not poor Q4 Local TV media ** (0.018) Observations 9,194 9,194 9,686 9,686 FM/TV mean FM/TV std. dev

43 Non-local media estimates Change in incumbent party vote share (1) (2) Audit (0.070) (0.051) Audit Non-local media 0.004* (0.002) (0.001) Audit Corrupt Q (0.086) Audit Corrupt Q3 Non-local media (0.003) Audit Corrupt Q (0.103) Audit Corrupt Q4 Non-local media (0.003) Audit Not poor Q (0.084) Audit Not poor Q3 Non-local media (0.002) Audit Not poor Q (0.111) Audit Not poor Q4 Non-local media (0.002) Observations 40,007 40,007 Non-local media mean Non-local media std. dev

44 Robustness checks Our main findings for local media are robust to a variety of robustness tests. Our results are robust to including an interaction with the number of non-local media. Our results are robust to simultaneously including the interaction of audit results with second-order polynomials for four important background indicators: non-local media, distance to the municipal head (from the precinct centroid), average years of education, and the proportion of households with a car. Our results are not driven by restricting attention to relatively small neighboring precincts. Our results are robust alternative operationalizations of malfeasance.

45 Media market structure Our results so far indicate the importance of local media for electoral accountability. We further investigate the media market structure focus on two important features: crowd-out by non-local media, and the geographic composition of a station s potential audience. A new media station attracts listeners or viewers away from local media stations, plausibly because voters also consume media for reasons other than acquiring politically-relevant information. We look at the quadruple interaction.

46 Non-local media crowd out Change in incumbent party vote share (1) (2) Audit Local media 0.014*** 0.020*** (0.005) (0.005) Audit Non-local media Local media *** (0.0002) (0.0001) Audit Corrupt Q3 Local media (0.008) Audit Corrupt Q3 Non-local media Local media (0.0002) Audit Corrupt Q4 Local media * (0.009) Audit Corrupt Q4 Non-local media Local media (0.0003) Audit Not poor Q3 Local media *** (0.008) Audit Not poor Q3 Non-local media Local media *** (0.0004) Audit Not poor Q4 Local media *** (0.008) Audit Not poor Q4 Non-local media Local media *** (0.0002) Observations 17,312 17,312 Non-local media mean Non-local media std. dev

47 Local media market share We test whether the incentives of the supply matter. Local media stations predominantly cover consumers within their municipality have the strongest incentives to cover audit reports relating to their municipality. For each local media station, we calculate the proportion of the population that receives a commercial quality signal that reside inside the municipal of the emitter. We then computed the average such share across all AM radio, FM radio and television stations covering a given electoral precinct.

48 Local media market share results Change in incumbent party vote share (1) (2) Audit Local media ** (0.005) (0.003) Audit Local media Local market 0.017** 0.027*** (0.008) (0.009) Audit Corrupt Q3 Local media (0.006) Audit Corrupt Q3 Local media Local market (0.015) Audit Corrupt Q4 Local media (0.006) Audit Corrupt Q4 Local media Local market (0.013) Audit Not poor Q3 Local media (0.007) Audit Not poor Q3 Local media Local market ** (0.012) Audit Not poor Q4 Local media (0.004) Audit Not poor Q4 Local media Local market *** (0.013) Observations 17,312 17,312 Local market mean Local market std. dev

49 Roadmap 1. Why the intensity of media coverage matters. 2. Background. 3. Data and empirical design. 4. Empirical Results. 5. Conclusion.

50 Conclusion Importance of local media for local electoral accountability in a developing context. Effective electoral accountability requires more than just a single local media outlet. Non-local media does not contribute for local electoral accountability. Local media market structure matters: Non-local media crowd out the effects of local media. Local media may not as effective to support electoral accountability when their audience contains many consumers from outside the municipality. To do: How much of the effect is due to news coverage and how much is due to campaign advertising?