1994 Elections in Mexico: Still Neither Fully Free Nor Fully Fair
by Andrew RedingThe 1994 Mexican elections were widely hailed by foreign observers as a great advance in democracy. Millions of dollars were spent on improvements in the electoral process, and there was far less election day fraud than in the 1988 presidential race.
Despite improvements, however, the elections were neither fully free nor fair. The elections preserved a virtual monopoly on power for the ruling Institutional Revolutionary Party (PRI), which has governed Mexico for two-thirds of a century.
The two principal problems remain the persistence of constitutional structures and electoral principles specifically designed to sustain one-party rule and the PRI's overwhelming advantage in campaign financing, media exposure and patronage.
Constitutional Design and Electoral System
Congress consists of two chambers: a 128-seat Senate, and a 500-seat Chamber of Deputies. Each federal entity (the 31 states and the Federal District) elects four senators for six-year terms, coinciding with the presidential term of office. Three senate seats go to whichever party wins a plurality; the remaining seats to the runner-up.
For the Chamber of Deputies, the country is split into 300 federal electoral districts, each of which elects one deputy by plurality. Another 200 deputies are elected at-large, with 40 seats in each of five regions apportioned in proportion to the vote received by parties that attain a threshold of 1.5%.
Deputies are elected to three-year terms, which means the entire Chamber of Deputies is renewed both in presidential election years and in mid-term elections. Neither senators nor deputies may seek reelection to consecutive terms.
On paper, the Mexican Congress has virtually the same powers as the U.S. Congress. In fact, however, the Congress has never seriously challenged any legislative proposal, budget or appointment submitted to it by a PRI president. That is in part because the PRI has always held a majority of seats in both houses of Congress -- a majority deliberately exaggerated by the method of election.
This bias is most obvious in the Senate, where the party that wins a state gets three of its four seats. Thus, if the PRI gets 40% of the vote, the PAN 35%, and the PRD 25%, the PRI ends up with 75% of the representation, the PAN 25% and the PRD with nothing. Not surprisingly, the PRI emerged from the August 1994 election with 74% control of the Senate, enough to make confirmation of presidential appointments a mere formality.
A similar, though less pronounced, bias exists in the method of election of the Chamber of Deputies. With the opposition divided and with electoral fraud still a problem in many rural districts, winner-take-all districts offer a bonanza for the PRI. In 1994 the PRI won 274 (92%) of 300 such districts.
Although the remaining 200 seats are allocated in proportion to the vote received by each party -- and no party can win more than 60% of total seats -- they do nothing to correct the over-representation of the PRI in the 300 district seats. As a result, the party that won 50% of the vote in 1994 ended up with its full allotment of 60% of the seats.
Of course, control of Congress by the President's party does not of itself ensure a docile legislature. Three other factors ensure the compliance of individual legislators:
- The only way to become a PRI candidate is to be selected by party leaders. There are no internal party caucuses and no primary elections in which to challenge machine candidates.
- There is no way to get reelected. The ban on reelection prevents development of the independent power bases that are a key feature of the relationship between the U.S. Congress and President.
- The Mexican Congress lacks the committee structure and staffing that enables the U.S. Congress to act as a watchdog on the executive branch. The result is a legislature that acts as little more than a rubber stamp for executive authority.
The PRI had an overwhelming advantage in campaign financing, media exposure and patronage. The country was festooned with PRI posters, banners, and billboards. In a country where two-thirds of the population relies on television as their primary source of news, both of the nationwide private television networks (Televisa and Television Azteca) did little to conceal their preference for the PRI.
The PRI also took advantage of the government-funded National Solidarity Program (Programa Nacional de Solidaridad, Pronasol) of grants to the poor. The Pronasol symbol was a variation of the PRI symbol (itself based on the Mexican flag). Since these grants are made to communities, and it is easy to determine how communities vote, recipients and would-be recipients were informally warned that future allocations would depend on their response on election day.
Finally, despite improvements in the electoral process, there still were troubling election-day irregularities. Civic Alliance (Alianza Civica), a nationwide coalition of human rights and pro-democracy organizations, fielded thousands of trained poll-watchers. In more than two-thirds of the polling stations covered, they saw voters with valid photo-ID cards being turned away because their names did not appear on voter lists. This is of particular concern because it suggests continued use of an electoral fraud known as la razurada, or "shaving" of suspected opposition voters from the lists.
Impact of Election Day Corruption
These irregularities were not, however, of such a scale as to affect the outcome of the presidential race. Officially, PRI candidate Ernesto Zedillo won with 50.17% of the vote. The right-wing PAN candidate Diego Fernandez de Cevallos came in second with 26% and the left-wing PRD candidate Cuauhtemoc Cardenas -- the likely true winner of the 1988 race -- came in third with 17%.
Unlike his predecessor Carlos Salinas, Zedillo clearly won election to the presidency. Yet, once allowance is made for irregularities, he almost certainly did not win more than 50% of the vote cast for the three leading candidates, as reported by the Federal Electoral Institute. As in 1988, the PRI vote was inflated to just a fraction of a percentage point over 50%, in order to claim a clear mandate, and secure continued one-party rule.
The difference was that in 1994, the extent of PRI vote inflation was much smaller and did not cast doubt on the outcome of the presidential election. The same cannot be said for some congressional elections nor Chiapas' gubernatorial election
The PRI won very narrow victories in many parts of the country, including Jalisco. The PAN won handily in Guadalajara, Jalisco's state capital and Mexico's second largest city. Yet in rural areas, where irregularities were common and poll-watchers scarce, the PRI won just enough votes for a razor-thin victory, denying PAN another senatorship.
The gubernatorial race in Chiapas was similarly tarnished by widespread irregularities. Officially, PRI candidate Eduardo Robledo won with 50.4% of the vote, to 34.9% for PRD candidate Amado Avendano. Yet observers from Civic Alliance documented a pattern of widespread vote fraud, the worst of any of the 31 states. Ballot secrecy was violated in two-thirds of the polls observed; there were attempts to influence voters in 45% of locations; and voters were seen in 9% of sites casting multiple ballots rolled inside each other like tacos.
In rebel-held areas, on the other hand, where irregularities were slight, about 70% voted for the PRD. Further undermining the credibility of the official results is that, as in the presidential election, the PRI claimed to have won a fraction over 50% of the vote, just enough to claim a clear mandate.
Need for Real Democratic Reform
As Mexico struggles in the wake of the peso devaluation that sent a shockwave through worldwide financial markets and has caused the loss of hundreds of thousands of jobs, the need for real democratic reforms becomes all the more pressing. One such reform that would help in the transition from authoritarian to more democratic governance would be to convert the semi-proportional representation system to full proportional representation.
Mexico now has a curious variant of the German mixed-member model, one that reserves enough seats to opposition parties to provide an appearance of pluralism, but also guarantees a safe working majority for the ruling PRI. By increasing the fraction of proportional seats from 40% to 50% and by allocating those seats to compensate for the ruling party's over-representation in winner-take-all seats, the federal and state legislatures can be made to fully reflect the diversity of Mexican society.
More faithful representation would not only introduce legislative checks and balances on executive authority, but would allow conflicts currently being waged through guns (as in Chiapas) and mass protests to be converted into parliamentary conflicts. In parliament, the very logic of the process and legitimacy of the participants would make compromise and peaceful resolution of disputes more likely.
Andrew Reding directs the North America Project of the World Policy Institute, where he is Senior Fellow for Hemispheric Affairs. He has published widely on Mexican politics. This article was a contribution to chapter seven of the 1995 Voting and Democracy Report.