More institutional reform in Mexico?

On Monday, President Felipe Calderón announced plans for some major structural reforms to Mexico’s political system (more here, here, and here). Among the numerous reforms announced, there are two I find both interesting and troubling. First, Calderón is going to try and push for allowing the immediate reelection of legislators and mayors. The second is the reduction in the number of legislators, presumably by reducing the number of legislators elected through proportional representation. I have written and the prohibition on reelection here before, please read that story for more background. While the issue of reelection comes up periodically, this time is notable both for the president’s public support and some signs that the other two major parties will also support it.

From the point of democratic theory, reelection is almost a necessary requisite for inducing both representation and accountability. Most of the arguments in favor of reelection are largely based on the theoretical benefits of reelection, and in fact, I know of no pundits or academics that are opposed to allowing for reelection in Mexico. Whenever reelection becomes a topic worthy of newspaper coverage, it is almost invariably in support of the reform, with these two most recent editorials in El Universal rehashing many of the same arguments in favor of reelection. The two major ones are: 1) reelection will force legislators to be accountable to voters, since voters will get to pass judgment after a single term, and 2) reelection will lead to a more professional legislature with institutional memory and career politicians dedicated to a legislative career.

One of the problems is, most Mexicans are against reelection, and this has been demonstrated again and again in national surveys. Some of the public resistance to reelection may be misguided based on a misinterpretation of Francisco Madero’s slogan of “effective suffrage, no reelection” during the 1910 revolution. Its not clear how this public resistance will manifest itself if the rules are changed, but it should not be ignored.

The other problem with the arguments in favor of reelection is that they ignore the complex interplay of numerous rules and practices that may work against the supposed benefits of reelection. In terms of the first benefit, greater accountability to voters, there are two issues that may impede accountability. First are the methods of candidate selection in Mexico. The use of primaries at the local level is growing, but there are still numerous cases where party leaders choose the candidate to be placed on the ballot without any voter input. The discretionary control that is written into party statutes and implemented in practice has numerous consequences for the relationship between representatives and voters. As the laws stand now, Mexico’s parties have complete control over who is placed on the ballot, meaning there is the possibility incumbent legislators could be forced off the ballot for angering party leaders (although likely a remote possibility). However, the greater issue is that legislators are going to be most responsive to those who help them get on the ballot and win the election. Understanding the relative role of party elites and voters in getting on the ballot and winning is crucial for understanding to whom the legislator will be most accountable. One recent editorial spent a single sentence on this issue, but really the benefits of reelection cannot be understood without a serious discussion of democratic candidate selection procedures. Second, Mexico’s politicians have a long history of progressive ambition, meaning they move around to various positions in political parties, as elected representatives, in the federal or state bureaucracy, or moving back and forth between politics and academia. While much of this is likely due to institutional rules that force this type of behavior, it is also a norm that may be hard to change. This type of ambition is fairly common in Latin America, even in places that allow reelection. The comparison is key, because in Latin American countries, party leaders exert fairly strong control over who gets on the ballot, and reelection to the same post is typically not a common goal for many legislators. Just because reelection may be allowed in Mexico, does not alter the goals of party leaders which may run against allowing individuals to develop strong bases of power in a single district, nor will it change in the short term the long term practice of changing positions every few years. In addition, since Mexico does allow non-consecutive reelection, when individuals do serve more than one term, they are more likely to be reelected through the proportional representation (PR) lists. The presence of the PR lists complicates the link between voters and representatives, if the informal practice of seeking reelection through the PR lists continues.

The above discussion leads to the potential problems with the second argument for reelection, a more professionalized legislature. Mexico already has a fairly powerful legislature, not only compared to its past prior to the late 1990s, but also compared to other Latin American legislatures. Allowing reelection may lead to more individuals with longer legislative careers who have greater experience navigating the legislative process. I do agree that professionalization is a problem. Right now, every three years the lower house, the Chamber of Deputies, is completely turned over. Since there are typically around 15-20% of legislators in each legislative term that have experience, these individuals help the newcomers learn the ropes. But, looking at the introduction of bills and the number of bills voted upon, the first year of each three year term is pretty thin in terms of activity. What we do see, is almost a year of wasted time every three years, and in the final year, a number of legislators who leave office early in search of other positions. Allowing reelection may lead to more legislative activity (although for better or worse really is something I can’t comment on). The other problem reelection might solve, is that there are a number of legislators who do virtually nothing. In terms of initiating a single bill, there are roughly 30% of legislators who can’t even do that, and another roughly 15-20% who either initiate their own bill or sign their name to someone else’s bill. These individuals show up to vote (although absenteeism on floor votes is also extremely high), but other than being warm bodies, there are a significant number of legislators who do not perform their most basic duty, to write legislation.

Yet, reelection might not be a panacea for solving these problems. Most legislators do not have the resources to engage in the research necessary to write good legislation. The party leaders have enormous control over the allocation of resources and staff, and without changing these internal rules and practices, Mexico’s legislators may not perform much differently.

The other potential reform I want to discuss is the reduction in the number of proportional representation legislators, typically referred to as los plurinominales, or for short, los pluris. Mexico’s Chamber of Deputies currently has 500 seats, 300 elected through single-member district plurality elections much like in the U.S., and 200 elected through a closed party list. The Senate has 128 seats, 32 of which are also selected through closed party lists. Each state legislature also has a varying number of pluris. Mexico first added these pluris way back in 1963 as a concession to opposition parties, and the PR seats were traditionally the only way the opposition parties could gain seats, as the ruling PRI (Institutional Revolutionary Party) completely dominated all of the district elections. Now, the pluris are typically chosen by the national party leadership of each major party, feature most of the key party leaders in both the Chamber of Deputies and the Senate, and are also used as patronage to hand out seats to important interest groups. There is some public anger against the pluris, largely because voters have absolutely no say over who can be placed on these lists or who ends up in the most powerful positions in the legislature. During the 2009 mid-term election, there was a small but vocal movement advocating for annulling one’s vote, in a protest against the perceived lack of responsiveness of the political system. One of the major demands of this movement was the elimination of the pluris. Some websites and facebook pages are now popping up pushing for their elimination, and a number of commentators are advocating for their elimination or reduction (here, here, here, and here).

I generally think this is a terrible idea, and there exist a couple of alternatives that may deal with the perceived unresponsiveness of the pluris. One recent editorial suggests that they get in the way and have no reason to exist. After looking at the evidence of what legislators do, not only do I disagree with this claim, but I found the opposite is true. It is the district legislators who are more likely to do nothing. Why is this the case? I think it is because the parties put their more experienced and talented people on the lists. Evidence is key in the discussion of eliminating the pluris. There definitely seems to be more public support for eliminating the pluris than reelection, so if reelection did not pass, but the reduction or elimination of pluris did, it is likely the legislature will perform even worse than it already does.

However, the major reason I think eliminating the pluris is a terrible idea, is that this might lead to a resurgence of PRI dominance through district elections. The PRI still has the best organized party in Mexico, and the reduction of the pluris could nearly cripple the center-left Party of the Democratic Revolution (although they are doing a good job destroying themselves), and return to one-party dominance in a number of states where the center-right National Action Party is not competitive. The party list seats were implemented to increase competition into the Mexican political system, and Mexico is not that far removed from the days of one-party dominance. Using proportional representation helps balance out disproportionality that comes from only using single-member districts.

How to deal with the lack of responsiveness of the party list legislators? The most basic is to open up the candidate selection of who occupies the lists. This could be done by using open-list PR, allowing voters to rank members of the list on a separate ballot, or through more direct primaries to allow voters to have a voice of who gets placed on the ballot. Another possible reform is to promote the increased use of dual candidacy. In most countries that employ a mixed electoral system like Mexico, candidates are allowed to run in a district race and also simultaneously be placed on the party list. Mexico’s major parties rarely use dual candidacy (although the PAN uses it more than the PRD or the PRI). If the party list candidates were forced to also run in a district race, then they would be more visible to voters, and potentially, more responsive. One interesting proposition is what is done in Japan, known as the “best loser” provision. In Japan, district candidates who lose are ranked on the party list after the election according how they performed in their district, encouraging them to do their best to attract votes in the district races regardless of their ability to win. I could envision in Mexico a system where the pluris are nothing more than those candidates who lost in the district race, rather a completely separate group of candidates who never have to campaign or communicate with voters. Electing losers has its own sets of problems and created controversy in a number of places such as Ukraine, New Zealand, and Japan (see numerous posts here this post about dual candidacy in mixed electoral systems). Yet, these types of changes would not only increase the responsiveness of the pluris, but also not lead to the potentially undesired consequence of reducing party competition in Mexico.

I do not want to come across as completely opposing either reelection or reducing the number of party list legislators. However, the lack of serious debate about both of these much discussed reforms in Mexico is troubling. (I want to thank Patrick Corcoran at Gancho for providing a number of links used here.)

A link between political institutions and crime in Mexico?

In a previous post about rising violence in Mexico, I mentioned in the comments that Mexico’s constitutional prohibition on consecutive reelection had an impact on citizens’ abilities to hold their politicians accountable for their actions. The comment was a little off the cuff and not well thought out, but I was careful not to suggest that removing this impediment to reelection was a partial solution to inducing Mexican politicians to be more responsive to citizen concerns about security issues. However, this post over at Mexidata has suggested just that, that allowing for reelection would allow Mexican voters to hold their leaders accountable for the way they address issues of crime and violence. A recent editorial in the Mexican daily El Universal has also addressed the issue of reelection and its link to accountability. In the growing debate over how to respond to the rising violence, many people in Mexico have been voicing their support for a recall provision, in a sort of knee-jerk, throw the bums out, type of response to the seeming incompetent government attempt to combat drug traffickers and kidnappers. In the editorial, the author suggests recalling public officials is less effective in holding leaders accountable than allowing for reelection.

Why is this important? One of the unique features of the Mexican political system is this prohibition on consecutive reelection, with only one other country in the world (Costa Rica) having a somewhat similar set of institutions. Therefore, the potential impact of prohibiting reelection or removing it once it has been in place after 70-plus years is not well understood either theoretically or empirically. I think in this case, especially with the serious problems Mexico is currently facing, allowing for reelection would not only have several unintended consequences, but would actually be dangerous.  Hopefully it becomes clear by the end of this post why I think so.

First, a little background. In Mexico, term limits have been in place since 1933, when the Mexican Congress approved a constitutional amendment that prohibited consecutive reelection for federal deputies and senators, as well as for state legislators and municipal presidents (i.e. mayors). Governors, as well as the President, are barred from reelection for life. Unlike term limits in many U.S. states and a few other countries around the world, Mexican politicians must sit out one term in most offices (except for executive positions), before being allowed to serve again. This means that an individual can potentially move up and down the ladder from being mayor to being senator, but cannot make a career in any single elected position.

            Part of the justification for term limits was to reduce the instability over presidential succession that had plagued Mexico since the end of the Revolution. The ban on presidential reelection has stabilized alternation in power since 1934. In addition, the reform was seen as implementing a key tenet of the Revolution, “Sufragio Efectivo, No Reelección,” or “Effective Suffrage, No Reelection,” which had been raised by Francisco Madero against the dictatorship of Porfirio Díaz in 1910. However, the ban on reelection for all public offices was not part of the revolutionary platform, and several scholars have convincingly argued it was put in place for other motives. A ban on consecutive reelection contributed to the centralization of power in the hands of the PRI’s national party leaders and the president, and also centralized power nationally by weakening local political parties and local power structures.

Since the 1990s, many scholars of Mexican politics have advocated for removing the ban on reelection for two reasons. First, allowing reelection is supposed to create politicians who are responsive to voters and allow voters to hold their elected officials accountable. Second, allowing reelection creates incentives for individuals to develop a career in a particular office. Drafting legislation is not easy and those who are allowed to become professional legislators have the skills and experience necessary to make better public policy. I don’t have much of a problem with the second argument, but I take issue with the idealism of the first.

The link between reelection and democratic representation is a fairly basic tenet of democratic theory and is a fairly pervasive part of most work on American politics. Democratic elections are theoretically supposed to allow for the representation of people’s interests because if voters are not adequately represented, they have the possibility to vote that person out of office the next election cycle. It is thought that reelection is a powerful incentive to force politicians to respond to people’s demands. The theory is great, the reality is something entirely different. The theory assumes all voters have perfect information about an incumbent’s record and the challenger’s policy stances. It also assumes equal competition between incumbents and challengers. Finally, it also assumes that people vote for particular individuals based on their own policy interests, rather than other non-policy reasons. Any casual observer of the reality of American politics should laugh at these assumptions, and political scientists have written literally millions of pages showing these assumptions don’t hold. What is the reality? Most voters are uninformed, many vote based on easy cues such as party, or the personal qualities of the candidate, and incumbents have a resource advantage in campaigns that makes it difficult for many challengers to gain recognition.

Moreover, just as there is a resource bias that favors incumbents over challengers, there is a bias in who is represented by these incumbents. Organized interests are generally better represented than the unorganized, primarily because organized interests are easier to mobilize to support one’s reelection and/or to donate money to one’s campaign. In the U.S. context, organized interests can mean many things, both positive and negative. However, translated to the Mexican context, the type of organized interests present at the local level are not comparable to what exists in the United States. In Mexico’s current state with the prevalence of organized crime and drug traffickers, the better comparison to what might happen with reelection is Guatemala and maybe Colombia, not the United States.

While this claim is really speculative, the involvement of drug traffickers in Mexican politics seems to be much lower than in other Latin American countries where drugs are a major part of the economy. Sure, stories surface every once and a while of individual politicians having links to different cartels, but the lack of reelection also seems to be a powerful incentive to keep organized crime out of electoral politics, unlike Guatemala. Politicians have to depend on their political parties, not local organized interests for their next job and are therefore not dependent on raising money to wage a reelection campaign. The cartels may have short-term interests in trying to buy off politicians, but they would have to do this every few years since the individuals would always be changing. For this reason, allowing reelection in Mexico could have some dangerous consequences.

            Right now, the possibility of removing the constitutional ban on reelection seems remote, although the issue has not gone away since the mid-1990s. Under the Fox Administration (2000-2006), numerous initiatives were introduced in the Congress to allow for reelection, and several months ago, a couple members of the National Action Party (PAN) introduced new proposals to remove the ban. The major parties are generally against removing the ban, likely because it would reduce their control over the party membership and elected leaders who use the party label. The people are overwhelmingly against removing the ban, at least according to a 2006 survey (See table on page 143 of the document). However, major institutional reforms are generally more likely in times of crisis, and if supporters of reelection are able to make a link between reelection and the security crisis, then it is possible we could see dramatic changes, maybe not for the better, and not in the directions advocates of reelection had hoped.

This post ended up being much longer than I had planned. In the unlikely event readers are interested in this topic, I definitely have more to say on this issue.