How Nicaraguans Voted
Although the official election results present inconsistencies and have been questioned, they provide a starting point for analyzing the vote and beginning to imagine the political future of the country.
Nicaragua's elections finally came to an end on November 22 with a reading of the names of all new elected au thorities. This put a full stop to one of the most complex crises that the country has gone through, and left a new correlation of forces in the national and local governments.
On the afternoon of October 23, with 87% of the results counted, the Supreme Electoral Council (CSE) had halted the announcement of preliminary results begun in the early hours of October 21 and questioned by the FSLN and other parties within hours. Sixteen days later, on November 8, the CSE had offered a new set of preliminary results, this time total, after a wearying review of tallies and recounting of ballots also challenged by the FSLN and other parties. These were still provisional results, because the Electoral Law gave the political parties the right to challenge them, as in fact they did. Finally, after sixteen more days, the CSE announced the definitive winners on November 22, at close to midnight. According to the electoral law, these final results cannot be appealed.
At the time, the CSE provided no numbers and offered no details on how many--if any--of the hundreds of polling places (JRVs) that the various parties wanted annulled due to their serious anomalies had been eliminated. The only thing made clear was that the FSLN's petition to annul the entire elections in Managua and Matagalpa had been rejected.
Despite Everything, These are the NumbersBecause the CSE has still not published the final data, the following analysis is based on the CSE document, "Provisional Electoral Results," published on November 8. Unlike the 21 successive partial reports the CSE issued between October 21 and 23, this document breaks the figures down by the 8,995 polling place (JRVs) that operated on election day. envío was unable to learn whether there are any differences between these figures and the final ones of November 22.
It does not cease to be both embarrassing and complicated to analyze the results of elections clouded by multiple anomalies and irregularities before and during the counting and recounting of votes. Despite the doubts about these numbers, however, the CSE's published results of November 8--though provisional and with inconsistencies--are the only official data we have to know how and for whom Nicaraguans voted on October 20.
The Specter of FraudThe Liberal Alliance began to speak of fraud two years before the elections took place. Many other parties only used the term after they began to compare their own parallel counts with the CSE's first published results the day after the elections. The initial mathematical review of the results demanded by various parties confirmed that their accusations were not wholly unfounded.
The October 20 voting process was complicated by anomalies, possibly fraudulent actions, localized in various municipalities--even some departments--of the country. What was the true magnitude of the irregularities? What information is available to us to quantify its dimensions? Might a close analysis of the official results offer us some clues?
It must be said from the start that it is technically impossible to determine whether or not there was elec toral fraud, much less its magnitude, with the data provided by the CSE. That data, however, does sustain the thesis that there were important anomalies during the electoral process. It also allows us to state that there was a tendency on the CSE's part to try to manage the results. The data was organized and presented with the idea that certain anomalies and irregularities would not be so visible to the public view, and thus their importance could be considerably minimized.
Covering up Chaos: One CaseIn addition to the November 8 document mentioned above, the envío team got access to the CSE's computerized data base, designed and prepared by the Spanish organization INDRA SSI, also with the results of the 8,995 JRVs. According to that data base, Managua had a total of 233 "abandoned" JRVs, that is, JRVs that showed no voters--or, to put it another way, no votes. The data base shows a total of 67,185 registered voters in these 233 JRVs. Assuming that these citizens are not fictitious, it is impossible to know why their votes were reduced to zero, since there is no explanation in the data base. In other words, we can't know who those people voted for. Chart I demonstrates the distribution of these 233 "abandoned" JRVs and their 67,185 "nonexistent" voters in the seven municipalities of the department of Managua.
This is only one concrete example of the grave anomalies that occurred in the elections. It is important because it demonstrates some of the quantitative irregularities reflected in the CSE data. Analyzing just this one case can give us clues to a broader reflection on others that, although we know they existed, became almost impossible to discover in the data base.
Are these "ghost voting centers" a sign of fraud? One could argue that they aren't, but not categorically. Since all the votes from these 233 JRVs disappeared, this anomaly evidently affects not just one party but all political groupings that obtained any votes in any of these JRVs. Speaking technically about fraud, it would be logical for only the votes favoring the party against whom fraud was committed to disappear.
This does not mean, however, that there was no electoral fraud in the department of Managua. To prove it, however, one must look at other data not from the CSE, for example data possessed by the political parties that contested the elections.
It is possible to make endless conjectures about the case of the lost JRVs in Managua. One could suggest, for example, that someone took the full ballot boxes from these 233 JRVs, planning to return them after destroying or altering the ballots that favored the party they wanted to damage. One could then reason that this person or persons, for whatever reason, later saw the impossibility of returning the other ballots without becoming implicated and therefore finally chose to make all the ballots disappear, including those for their own party. If that were the case, we would be facing a sui generis fraud that would paradoxically harm all parties without distinction--unless, of course, those JRVs were located in pockets of support for the party targeted.
Where are the "Ghost" JRVs?Without discarding the fraud hypothesis, it's most likely that a good part of the 233 "ghost" JRVs in the Managua department disappeared due to the disorder that gummed up the functioning of all the institutional gears leading up to the elections. If it is said that from hand to hand even a cathedral can be lost, what luck can be expected for the ballots of these 233 JRVs, which were being passed from hand to hand along a chain that showed considerably deficient control mechanisms, both in the reception phase and in the storage of electoral materials?
Unlike in its data base, the CSE's November 8 report never explicitly mentions in the existence of Managua's 233 "ghost" JRVs, or of their 403,048 ballots--six for every voter. Perhaps this is an attempt to maintain its own institutional prestige, since the CSE could have damaged its own image by systematically introducing a column in all its reports showing the total of--how many more?--"ghost" JRVs plus the voters that disappeared with them. Instead of doing this--which was what a spirit of transparency would advise--the CSE omitted from its data base any reference, however minimal, to anomalies or irregularities.
In a statement after the publication of the November 8 results, the Carter Center wrote: "Regrettably, the CSE did not release information about the corrections made during the review [of the results], or the number of votes counted to determine the provisional results announced on November 8.... It is imperative that the CSE release this information as soon as possible so that the Nicaraguan people, as well as the international observers, can evaluate the appeals of the parties and the announcement of the Council. Only through such an open and comprehensive evaluation can the Nicaraguan people have confidence in the official results."
In all of its reports, the CSE limited itself to offering the following six types of information for the electoral results of parties or associations: Total JRVs, JRVs Reported, Total Registered Voters, Total Voters, Valid Votes, Null Votes.
Where do the "ghost" JRVs appear? There wouldn't even have to be a column titled Missing JRVs; after all, we learn how many registered voters did not vote on election day by subtracting Total Voters from Total Registered Voters. We should thus be able to find the "ghost" JRVs by subtracting JRVs Reported from Total JRVs. Nonetheless, envío discovered that they are included in both the category of "Total JRVs" and of "JRVs Reported." Their inclusion in the first category is a perfectly normal procedure, but their inclusion in the second one is biased accounting. The "ghost" JRVs are not, in reality, "Reported." On the contrary, they are "unreported." The CSE knows they existed because they are on the electoral list, but to then say that they were "reported" leaps a huge chasm. If the CSE had acted more transparently, it would have carefully separated the "reported" JRVs from those that were never reported, since they disappeared leaving no trace but their number on the electoral list. The artifice employed by the CSE allows it to hide the existence of the "ghost" JRVs, forcing the total JRVs and the reported JRVs to coincide. This way, the mathematical results match perfectly, although at the cost of manipulating data and concepts used to order and categorize them.
It remains to be elucidated where the CSE finally assigned the "ghost" votes from these JRVs. Having included the JRVs themselves, it could have chosen to include the votes in the column of "null votes" (deposited ballots that were unmarked, doubly marked, etc.), but it did not do so. It chose rather to include them all in a category also not explicitly mentioned in the CSE reports: the "no voter" category. As mentioned above, this information is obtained in the report only by subtracting "total voters" from "total registered voters." In theory, the result obtained from this mathematical operation should give us the number of citizens who abstained from voting. But this is only "in theory" because in reality, the CSE diluted that abstention category by including in it people who in fact voted but whose votes mysteriously disappeared. One does not have to be an expert in electoral issues to realize that disappeared votes are not the same as abstentions. An example: in the election for departmental representatives to the National Assembly in the municipality of Managua, envío discovered that 55,295 of the 142,154 citizens in the "no voter" category--(39%)--are "disappeared" voters. The rest are, in theory, citizens who abstained from voting. This case is only a small demonstration of how whitewashed the CSE's "official" provisional results are.
How Many Didn't Vote?If we include "disappeared" voters with abstaining voters, as the CSE did, the percentage of electoral abstention will obviously rise. Unfortunately we have no other data, nor did we have the time needed to "clean up" all the data we do have, so our analysis of abstentions in the 1996 elections is not fully freed of the bias introduced by the CSE. For this reason, it should be interpreted with caution.
In the case of the presidential elections, if we calculate the abstentions at the national level using the data offered by the CSE, the rate is 22.9%. This is a high percentage that doesn't appear to correspond to the participation level observed on election day. According to our estimates, if we subtract from that percentage the voters whose JRVs we think vanished, the possibly real abstention level falls to approximately 13.9%--very close to the 13.7% abstention rate in the 1990 elections.
We can only make assumptions about the abstention rate by department, since the CSE provides this figure only in its breakdown for departmental National Assembly representatives. The reason it's risky to automatically assume that people who didn't vote for departmental representatives didn't vote at all is that we don't know how the CSE disposed of cases of JRVs in which the ballots for only some elections were sent. Qualifying the following with that caveat, however, it appears that the four departments with the highest abstention rates were, in descending order, the RAAS, the RAAN, Río San Juan, Jinotega and Matagalpa. It is likely that genuine abstention has two separate motivations in these and in other areas of the country: citizens who were undecided, disenchanted with politics, etc., so consciously and freely chose not to participate in the elections; and others who decided it would be impossible to vote either because of difficult access to voting centers--particularly in the RAAS, the RAAN and Río San Juan--or because many JRVs opened their doors only after long delays. This latter factor many have led many citizens to return home even after making the effort to get to the JRV, thus giving up their right to vote.
Null VotesIn the CSE report, the "null votes" category is one of few which appears--at first glance--to offer no confusion or ambiguity. In theory, this category exclusively designates deposited ballots that were rejected because they were defaced, unmarked, marked in more than one box, etc. However, this is not exactly the case. A high level CSE official explained to envío that for JRVs in which only part of the ballots disappeared, these missing ballots were included in the "null vote" category.
Chart II shows the national percentage of null votes in four of the six elections held on October 20.
As can be observed, the presidential elections had the lowest rate of null votes (4.95%). This could be related to the fact that the ballot for this race offered the least possibilities for mistakes, particularly for illiterate people, because it included the photograph of each of the 24 presidential candidates. This was not the case for all but one of the other five ballots (National Assembly representatives, mayors, Municipal Council representatives). Only the ballot for Managua mayor also included photographs. This feature of the presidential ballot may also explain why the percentage of null votes in these elections was lower than the 6% national average in 1990.
Another possible factor is that the presidential elections were the most important in peoples' minds and also the most highly publicized. According to all polls, more people had decided who would get their vote for president than they had in any of the other elections. Combining this clearer determination with the fact that the presidential ballot was the first of the six given each voter, on which they could identify their choice's photo, meant a greater likelihood of voting correctly.
The further voters went through the intimidating stack of ballots, the greater their chances of making errors. For one thing, each ballot had a different number of columns. In the case of National Assembly representatives, for example, some parties had separated from the presidential alliance they had joined to run their own legislative candidates. And in the case of the municipal elections, the ballots were further elongated by the popular subscription candidates formed in a number of municipalities, who were only permitted to run candidates for mayor and municipal councilor.
To make matters even more complicated, both these popular subscription associations and the smaller parties with no real shot at the presidency had a vested interest in encouraging voters to split their ticket. According to all Managua polls, for example, independent mayoral candidate Pedro Solórzano was the front runner, even though the Liberal Alliance led the polls for the presidential race. During the ballot scrutiny after the JRVs closed on election day, one could discern that a number of people had not quite grasped the ticket splitting idea: more than a few ballots for mayor had been marked for one of the two leading parties as well as for the local independent favorite for mayor.
A final complicating factor that surely led to the annul ling of some ballots was that the ink from the felt tip pens issued for marking them was not always dry when voters folded their ballots for insertion into the ballot boxes. The CSE had instructed that ballots should be accepted if voter "intent" was evident, but in some cases the ink bled through too much for intent to be clear, and surely in others there were squabbles over the issue.
On the subject of intent, it sometimes became a judgment call. There was no debate when, as in one case seen, the voter had scrawled "NO" over the picture of a presidential candidate, but what was the proper call in the following example, also seen on a presidential ballot? Instead of an x, the voter had written the words "God help us" in the circle for one of the parties.
Examining the percentage of null votes for departmental National Assembly representatives--again the only category for which this information is provided by department--we find that the highest percentages were in the RAAN (9.84%) and Río San Juan (9.11%). In addition, the RAAS (7.92%), Rivas (7.57%), Nueva Segovia (7.57%), Jinotega (7.5%), Madriz (7.43%) and Matagalpa (7.26%) all showed null vote percentages over the 6.15% national average.
Presidential ElectionsThe Liberal Alliance presidential ticket, made up of Arnoldo Alemán and Enrique Bolaños, won the presidential elections with 51.03% of the valid votes. In second place, separated by 13.28 points, was the Daniel Ortega Juan Manuel Caldera ticket of the FSLN, with 37.75% of valid votes. (In 1990 the National Opposition Union's Violeta Chamorro Virgilio Godoy ticket won with 54.74% of valid votes while the FSLN's Daniel Ortega Sergio Ramírez ticket pulled 40.82%.)
The other 22 parties or alliances that ran in the presidential race this year accumulated a total 11.22% of the valid votes, with 19 of them getting a maximum of 0.60% each. The most extreme case was the Nicaraguan Democratic Alliance Party (PADENIC), which only pulled 904 valid votes from all over the country--0.05% of the total. Only the novice Christian Way (Camino Cristiano, or CCN), led by Assemblies of God pastor Guillermo Osorno, and the historic Conservative Party of Nicaragua (PCN), led by Noel Vidaurre, managed to break through the 1% barrier of national votes. Osorno and Vidaurre respectively obtained 4.1% and 2.26% of the valid votes.
The fact that Alemán won the presidential elections with over 50% of the votes caused surprise. For the majority of analysts, Alemán's turning point can be attributed to the call made by the Supreme Council of Private Enterprise, the Catholic bishops and priests--with Cardinal Obando at their head--and the President of the Republic herself to vote against the FSLN. What would the results of the elections have been without the public invitation by such business, religious and political leaders to close ranks against the Sandinistas?
The results of the last CID Gallup poll before the anti Sandinista offensive of the final pre electoral days suggested three distinct possibilities, depending on the mobility of statistical margins of error and on the election day behavior of the shrinking undecided vote. Only in one of those scenarios did the FSLN appear as a winner in the first round, with 45% of the votes, 3 points more than the Liberal Alliance. In the other two, the Liberal Alliance appeared as the winner in the first round: in one case with 45% of the votes, leaving the FSLN 3 points behind, and in the other with 48%, leaving the FSLN with 39%. Given this technical neck and neck race, illustrated as well in the final polls of four other pollsters, only Borge and Associates, in the pay of the Liberal Alliance--dared predict an unequivocal Alemán win in the first round, categorically ruling out the possibility of a second round.
CID Gallup's third scenario is the one that most closely approximates the October 20 electoral results. Precisely because of this it offers us a parameter, a reference point, to be able to quantify a posteriori the possible effect on electoral behavior, especially among undecided voters, of the call by the Catholic hierarchy, COSEP, Violeta Chamorro and others to vote against the FSLN.
The data of this third scenario suggests that the impact of those anti Sandinista calls was not as massive as some analysts suggest, which would indicate that only a relatively small part of the electorate reacts blindly to the public opinions of its religious or political leaders. Chart III, which compares the actual results of the presidential election with each of the three CID Gallup projections, suggests that the FSLN lost 1.25% of the votes from that third scenario as a result of the public interventions of various power sectors, while the Liberal Alliance gained 3.03% of the votes (73,358). Taking one of the other two projections as the benchmark, however, shows far greater effects. The chart also suggests that the other 22 parties lost 1.78% of their total votes, slightly more than the percentage lost by the FSLN. This would means that the Liberal Alliance took more presidential votes away from those political parties than it could from the FSLN.
The primary effect of the anti Sandinista call by religious, political and business leaders was to mobilize more non Sandinista and non Alemán voters to opt for the Liberal Alliance's presidential candidate, but not the parliamentary and municipal ones, or at least not in the same proportion. The votes for National Assembly representatives and municipal mayors and council members corroborates this hypothesis. It is also probable that another effect of the anti Sandinista crusade was to dissuade some of the undecideds who had been leaning toward the FSLN--above all those from low income sectors who still recall the benefits of the Sandinista government's social policies but also haven't forgotten the war. These sectors, with their simple religiosity and inclination to mystical interpretations, would likely be particularly susceptible to Cardinal Obando's viper parable.
Elections for National AssemblyThere were two elections with separate ballots for the National Assembly: one ballot for the 20 national, at large representatives and 17 separate ballots for the 70 representatives from the 17 departments. In the national representative election, the Liberal Alliance won 46.03% of the valid votes, the FSLN won 36.55% and the rest of the parties got the other 17.42%. In the departmental representative elections, the overall results were: 45.30% for the Liberal Alliance, 36.42% for the FSLN and 18.28% for the remaining parties.
Chart IV merits various comments. In the first place, the percentages of votes obtained by the Liberal Alliance, the FSLN and the rest of the parties differ in the two separate elections. Comparing them, the Liberal Alliance and the FSLN both lost some votes in the departmental balloting, though the FSLN lost fewer (0.13%) than the Liberal Alliance (0.73%). This shows that a small sector of supporters of both parties preferred to split their ticket, voting for departmental representatives from other parties. In absolute terms, these percentages represent the loss of 2,305 votes for the FSLN and 12,945 votes for the Liberal Alliance in the elections for departmental representatives over those for national representatives.
This behavior of the Sandinista and Liberal Alliance electorate explains how the remaining parties won a higher percentage of votes in the elections for departmental representatives (18.28%) than in those for national representatives (17.42%). Conversely, the electorate of these other parties did not all vote for their own parties' presidential candidates; some voted for the Liberal Alliance candidate and others, to a lesser degree, for the FSLN candidate. Using as a base line the 17.42% who voted for the national representatives of the other parties, it is 5.20% greater than the total of those who voted for these parties' presidential candidates.
The New National AssemblyOf all the parties participating in the two parliamentary elections, only 11 won seats in the National Assembly. The other 13, failing to win even one seat, will lose their legal status as stipulated in the Electoral Law. Those that no longer exist are: Conservative Popular Alliance Party (APC), National Renovation Movement (MORENA), Central American Integrationist Party (PIAC), Pan y Fuerza Alliance, Nicaraguan Justice Party (PJN), National Democratic Party (PADENIC), Liberal Unity Party (PUL), Renovating Action Movement (MAR), Nicaraguan Workers, Peasants and Professionals Unity Party (PUNOCP), Marxist Leninist Popular Action Movement (MAP ML), Communist Party of Nicaragua (PC de N), Nicaraguan Socialist Party (PSN) and Democratic Action Party (PAD).
Chart V shows the distribution of National Assembly seats among the 11 parties or alliances that won at least one representative. The distribution is divided by those who won with valid vote quotients in the first round of counting (Elected), those who got an Assembly seat due to the quotient system for losing presidential candidates (Pres. C.), and those who won due to various mechanisms of counting the residual votes after full quotients have been assigned (Resid.).
As the chart shows, the Liberal Alliance has 42 of the 93 representatives in the new National Assembly and the FSLN 36, while the other 15 are distributed among a total of 11 political parties or alliances. The latter have only one representative each, with the exception of the CCN with 4, the PCN with 3 and PRONAL with 2.
Of these new National Assembly representatives, 63 were elected simply and directly. Another 3 (Ortega, Osorno and Vidaurre) are losing presidential candidates who join the Assembly because, as stipulated in the Constitution, they obtained a sufficient electoral quotient. The rest, a total of 27, won their seat due to the complicated set of vote counting mechanisms set up in the 1984 Electoral Law to open National Assembly doors to the greatest degree possible of proportional representation by small political parties. According to these "electoral residual" mechanisms, "averages of averages" are obtained through a successive mathematical regrouping of "left over" votes; parties whose votes are equal to or higher than those ever lower new mathematical quotients automatically win a seat in the Assembly. That was how the CCN won three more representatives, the PCN two more and PRONAL its only two. It was also through this mathematical juggling that the PRN, MRS, Unity Alliance, PLI, ANC and UNO 96 each won a seat. The residual mechanism, whose original goal was to foment pluralism, ended up in 1996 favoring party opportunism and politicians with little party representation and scant popular backing.
Although both the Liberal Alliance and the FSLN have solid blocs in the Assembly, neither alone has the 48 vote simple majority needed to approve ordinary laws. And both fall far short of the 56 votes needed to approve constitutional ranked laws (Electoral Law and others) or reform the Constitution itself. To reach either of these two key numbers, the Liberal Alliance and the FSLN will have to make alliances with some or all of the representatives from the other 11 parties in the new Assembly. Since the FSLN could only get the number of votes required for constitutional ranked laws by cutting into the Liberal Alliance, it will probably have to content itself with trying to pull enough votes away from the Liberals to block their own ability to reach that number.
Ever since the preliminary results were made known, control of the National Assembly has been a priority point on both the Liberal Alliance and FSLN agendas, since the new legislature will have considerably more power than it has ever had before. Some of the 1995 constitutional reforms--whose implementation was, after months of negotiation, put off until this new government takes office--significantly changed the rules of the game, giving the Assembly many attributes previously belonging to the executive branch. If Alemán cannot control the Assembly, it will be difficult for him to carry out his plan of government using only the executive power.
Getting control of the Assembly will not be an easy task for either the Liberals or the Sandinistas. The main obstacle for the Liberals will be, paradoxically, themselves. The Liberal Alliance is made up of six different member parties (PLC, PLN, PLIUN, PALI, Liberal Convergence and PUCA), as well as politicians and technocrats with no party affiliation at all or with traditions far from or even opposite to the Liberal ones. This heterogeneity is already, and will continue to be, a source of fissures and tensions that could weaken the Liberal Alliance's ability to pressure and negotiate within the Assembly over the long haul.
National vs. Departmental RepresentativesAnother new aspect introduced by the reformed Constitution was the creation of 20 national representatives. With this reform, the departmental representatives are not eliminated, but are reduced to 70. A more worthy reform closed the loophole allowing candidates to run for departmental representatives of regions or departments in which they didn't actually live.
Of the 20 new national representatives in the Assembly, 9 are from the Liberal Alliance, 8 from the FSLN and the 3 others are distributed among the CCN, PRONAL and PCN. In terms of the percentage of votes obtained in the election of national representatives, the Liberal Alliance pulled 46.03% of the total, the FSLN 36.55%, the CCN 3.73%, PRONAL 2.36% and the PCN 2.12%.
The distribution of the 70 departmental representatives is very complex. A fixed number of seats are established by the Electoral Law for each of the country's departments or regions, which are distributed to each party based on the electoral quotients and residual vote counts described above. The various dispositions determining how to assign the seats obtained through residual votes to departments are apparently what allowed a totally paradoxical situation to occur: parties were assigned their seats in departments in which they won their most insignificant percentage of votes.
The most absurd cases have been those of the ANC, which won a representative in the North Atlantic Autonomous Region (RAAN), even though it got only 0.64% of that region's votes--the ridiculous quantity of 311. And the Unity Alliance, which also got a representative in the RAAN--and only 557 votes. And the PRN, which got its seat in the department of Estelí, where it only pulled 0.53% of the votes, a whopping 408. And the PLI, which got its representative in Granada, where it won 626 votes. Situations such as these, in which the figure of departmental representative becomes adulterated, can presumably be avoided in the future by reforming the Electoral Law to establish correspondence between the seats obtained by a political party in a department through the residual votes and the percentage of votes that party obtains in the same department. The word presumably is used advisedly here, since we have not seen all the relevant dispositions. It may even be that the other parties that could have been assigned the seats above had even smaller votes in the respective departments.
As can be observed in Chart VI, the Liberal Alliance obtained at least one seat in each of the country's departments and regions. It's highest percentages were in Chontales (63.82%), Boaco (57.27%), RAAS (57.16%), Jinotega (53.43%) and RAAN (50.18%). In all of them it managed to capitalize on the population's anti Sandinista sentiment better than other non Sandinista political parties did.
During the war in the 1980s, the "contras" had their greatest support in Boaco, Chontales and Jinotega, with a convinced social base among people negatively affected by the Sandinista government's rural economic policies. In the RAAS, on the other hand, the Liberal Alliance has controlled the autonomous government for two years, after winning the 1994 Atlantic Coast elections, and in the RAAN it has controlled the government for about a year, after the governor chosen in 1994, a member of the Miskito organization YATAMA, was removed. In both regions the Alliance used this institutional platform to maintain or even broaden its own base of sympathizers.
The FSLN received its highest percentages in León (46.06%), Estelí (45.74%), Chinandega (45.72%) and Managua (42.80%), though it did not reach 50% of the votes in any department. In León, Estelí and Chinandega its percentage of votes was greater than that of the Liberal Alliance.
On the other hand, the FSLN didn't win even one representative in the departments of Río San Juan and Granada or in the RAAN, although it came in second behind the Liberal Alliance in each of those departments and regions. Exempting Río San Juan, which gets only one parliamentary seat by law, the FSLN was the victim of the "electoral residuals" mechanism in the other two departments. For example, it won no seat in Granada, even though its vote percentage was 36 times more than the PLI's, which did. The same thing happened in the RAAN, where the Unity Alliance and the ANC each received a seat, even though the FSLN received 22 and 66 times more votes than those two parties respectively did.
Ten parties obtained only one departmental representative throughout the country--always by the residual mechanism. In none of the respective departments or regions where these parties were each assigned their seat did they receive over 10% of the total votes.
The CCN and the PCN each got two departmental seats. The CCN obtained its greatest number of votes in the departments of Managua, León and Matagalpa, followed by Chinandega, Masaya and Jinotega, though with significantly lower results in the latter three. The PCN won its greatest percentages in Chontales and Matagalpa, followed by Granada, Masaya and Chinandega although again with significantly fewer points. In distant tenth place for the PCN was Boaco, a department that, together with Chontales, is considered a "hotbed of Conservatism" in Nicaragua.
PARLACEN ElectionsRepresentatives to the Central American Parliament (PARLACEN) were elected for the first time in Nicaragua's history. Never before had Nicaraguans used their vote to choose who would represent them in a regional body.
As is apparent in Chart VII, the results of the PARLACEN election are very similar to those for national representatives to the National Assembly. The same parties obtained exactly the same number of seats in both elections and the percentages of votes obtained by each does not differ by even one percentage point in any case. The two election results suggest that the electorate basically followed the same voting pattern for each, except for giving a few more votes to PRONAL and a few less to the PCN.
According to the reformed Constitution, municipal governments will have greater power and autonomy in relation to the central government in managing and obtaining their own material and financial resources. They will thus be an important source of power in the future government. Due partly to the resources they will manage (if they succeed in their task of attracting them) and partly to the legal and institutional gaps that the Assembly must still fill with respect to municipal responsibilities, mayors' offices will also become important spaces in the struggles for power at both the local and national levels. In addition, municipal elections come again in four years. Parties that control the mayors' offices and win the local populations' approval through their effective management will be accumulating votes for the presidential elections the following year. It should not be forgotten that Arnoldo Alemán reached the presidency and managed to reconstruct and rebuild his own political party from his position as mayor of Managua between 1990 and 1995.
Nicaragua has 145 municipalities, of which the Liberal Alliance won 92 (63.44%) and the FSLN 51 (35%). Between them, they control all but two mayors' offices. Those two, El Rosario in Carazo and Potosí in Rivas, were won, respectively, by the MRS and by the Civic Association of Potosí, the only popular subscription association to win any municipality.
The Liberal Alliance won 11 of the country's 17 departmental capitals, including Managua, the most important of them, while the FSLN won the remaining 6. The departmental capitals won by the Liberal Alliance were Chinandega, Managua, Masaya, Granada, Rivas, Juigalpa, Boaco, Matagalpa, Jinotega, Bluefields (RAAS) and San Carlos (Río San Juan). The FSLN won Ocotal (Nueva Segovia), Somoto (Madriz), Estelí, León, Jinotepe (Carazo) and Puerto Cabezas (RAAN).
Chart VIII shows the number of municipalities per department won by each political group. The Liberal Alliance beat out the FSLN in 12 (76.5%) of the 17 departments or regions in the country. In 2 of those--Chontales and Río San Juan--it won all the municipal governments, and in the other 10--Boaco, Granada, Jinotega, Managua, Masaya, Matagalpa, Nueva Segovia, RAAN, RAAS and Rivas--it won between 54% and 85.7% of them. Of the remaining 5 departments, the two parties split Estelí down the middle, the FSLN won 50% in Carazo (the MRS, with 1 mayor's office, took another 12.5%) and the majority in 3--León (80%), Chinandega (69.3%) and Madriz (55.6%).
The FSLN's greatest historic social base is in the northwestern part of the country (León and Chinandega), a zone now seriously affected by the crisis in traditional cotton production. The FSLN also held off the Liberal Alliance in the greater part of the Segovias (Madriz, Estelí and Nueva Segovia), traditionally also a Sandinista region and scene in the 1930s of the feats of General Sandino's army.
Any evaluation of municipal victories or defeats, however, must deal not only with the total number of municipalities won or lost by Liberals or Sandinistas. As or more important is the number and economic situation of the inhabitants of each of the municipalities. This can be the focus of future analyses by the envío team.
Municipal Elections: 1990 to 1996It is interesting to compare the results that the FSLN obtained in the February 1990 municipal elections with those it obtained almost seven years later in these elections. The comparison should be interpreted with some perspective, however, because we must compare elections for municipal representatives in 1990 with those for mayor in 1996. In 1990, voters elected only Municipal Council members, who then chose the mayor from among themselves. In 1996, citizens directly elected both mayors and Municipal Council members.
We have limited ourselves to listing the municipalities, without examining why the FSLN may have won or lost in each one of them. Even without this analysis, however, envío considers it worthwhile to present the list for our readers.
* In 22 municipalities the FSLN won the elections in 1990 and again in 1996, the majority of them in the Estelí and Nueva Segovia departments: Jalapa, Dipilto, Ocotal, San Fernando, Pueblo Nuevo, Condega, Estelí, Cinco Pinos, Somotillo, Villanueva, Larreynaga, León, San Francisco Libre, Mateare, Masatepe, San Marcos, Jinotepe, Santa Teresa, San Juan del Sur, Cárdenas, Tuma La Dalia and San Ramón.
* The FSLN won 25 municipalities in 1996 that it had lost in 1990, the majority of them in the departments of Madriz, Chinandega, León and Masaya: El Jícaro, Somoto, Telpaneca, San Juan del Río Coco, San Lucas, San José de Cusmapa, San Francisco del Norte, El Viejo, Puerto Morazán, El Realejo, Chichigalpa, Posoltega, Achuapa, El Sauce, Quezalguaque, Telica, El Jicaral, La Paz Centro, Tisma, Nandasmo, Niquinohomo, Nandaime, Buenos Aires, Esquipulas and La Concordia.
* The FSLN lost 8 municipalities in 1996 that it had won in 1990: Macuelizo (Nueva Segovia); San Pedro del Norte (Chinandega); Bocana de Paiwas (Matagalpa); Wiwilí (Jinotega); San Miguelito, San Carlos and El Castillo (Río San Juan); and El Rosario (Carazo)--where the winner, mayor since 1979, is an FSLN member who went with the MRS in 1994.
* In 1996 the FSLN won 4 of the 13 municipalities on the Atlantic Coast whose municipal limits were not yet defined in 1990: Puerto Cabezas, Bonanza, Corn Island and the Río Grande delta.
* The FSLN lost 76 municipalities in both 1990 and 1996: Santa María, Mozonte, Ciudad Antigua, Murra, Quilalí, Yalaguina, Totogalpa, Palacaguina, Las Sábanas, San Juan de Limay, San Nicolás, La Trinidad, Santo Tomás del Norte, Corinto, Chinandega, Santa Rosa del Peñón, Nagarote, Tipitapa, Villa Carlos Fonseca, Managua, Ticuantepe, San Rafael del Sur, Nindirí, Masaya, La Concepción, Catarina, San Juan de Oriente, Diriamba, Dolores, La Paz de Oriente, La Conquista, Granada, Diriá, Diriomo, Belén, Moyogalpa, Altagracia, Tola, Rivas, San Jorge, Rancho Grande, Río Blanco, San Isidro, Sébaco, Matagalpa, Matiguás, Ciudad Darío, Terrabona, San Dionisio, Muy Muy, Waslala, Wiwilí, El Cuá Bocay, Santa María de Pantasma, San Sebastián de Yalí, San Rafael del Norte, Jinotega, Morrito; plus all 6 municipalities in the department of Boaco and all 12 in Chontales.
Popular Subscription AssociationsNot only political parties and alliances participated in the October 20 elections. What the Electoral Law terms "popular subscription associations" also competed, although they only had the right to present candidates for municipal offices. According to the Electoral Law, to be formed, an association needed, among other things, to present to the Supreme Electoral Council a "written request signed by a minimum of 5% of the citizens on the electoral rolls corresponding to the respective electoral area."
A total of 53 popular subscription associations participated in the elections, distributed by department as indicated in Chart IX.
Only one of these associations won the mayor's post in the whole country: the Civic Association of Potosí (ACP), which received 37.65% of the municipal votes in Potosí, department of Rivas. The great expectation awakened with the right of local associations to participate was not realized. Was it that the big party machines and national structures bowled over the initiatives and structures of these local political associations? Or did the enormous complexity of voting with six different ballots block the way to the ticket splitting expected to benefit the new associations?
Despite winning only one municipality, an important number of association candidates finished in second or third place. The most prominent case was Viva Managua, which with 33% of the decided votes appeared in a comfortable first place in the last CID Gallup pre electoral poll in Managua. All polls, in fact, predicted that Viva Managua's candidate, Pedro Solórzano, would be Managua's new mayor. Though Liberal Alliance candidate Roberto Cedeño, who placed third in most polls, rode Alemán's coattails to victory, Solórzano got 26% of the votes, only 2 points behind Cedeño. His 100,089 votes are what third place CCN presidential candidate Guillermo Osorno won in the whole country.
Of all the 53 popular subscription associations participating, 20 (37.7%) won third place in their respective municipalities, which is very significant given that this is the first time in the country's political history that they were competing.
Even though the SOL (sun) association, led by Herty Lewites, ended up in fourth place in the elections for Managua mayor, it is worth mentioning. Many electors considered Lewites a moderate and modernizing Sandinista alternative to the confrontational and rigid official FSLN candidate Carlos Guadamúz, director of the popular Radio Ya. With a total of 47,442 votes, Lewites finished behind Guadamúz (98,809 votes).
It should also be stated that not all popular subscription associations achieved positive results; 29 of them (54.7% of the total) failed to get 5% of their local votes. That is, their votes did not even equal the number of signatures they needed to win the right to participate in the elections.
Municipal Council ElectionsFor reasons of time, the envío team was unfortunately unable to process the election results for municipal council members in the country's 145 municipalities. Therefore we are presenting Chart X only with the results in the 17 departmental capitals. This, however, represents a significant sample, since the departmental capitals are their main municipalities, in terms of both population density and the economic and political life of each department.
To understand fully what the correlation of forces in the Municipal Councils of the country's departmental capitals will be, it must be remembered that the elected mayors are also council members by law and, as such, have a vote on the councils. Taking into account both this and the political groupings with similar goals within the municipal councils, the FSLN has a majority of council members in Jinotepe, Estelí, León, Somoto, Ocotal and Puerto Cabezas. In the case of Managua's Municipal Council, it could have a virtual tie with the Liberal Alliance, since the Alliance will have the votes of its 7 members plus Mayor Cedeño and probably that of Pedro Solórzano, from Viva Managua--9 in total. The FSLN could have this same number if, in addition to its own 7 members, it wins the votes of the CCN and Herty Lewites, leader of SOL.
In 11 of the departmental capitals, the correlation of forces within the councils is, in general, disadvantageous for the FSLN and overwhelmingly favorable for the Liberal Alliance. It is possible that in some capitals the FSLN could, through careful alliance building, win an additional vote in its favor, though it will not be easy.
Women ElectedIn all six elections, the great majority of winning candidates were men, but women comprise a notable minority. Charts XI and XII are eloquent: of the 403 officials that appear in the chart, only 12% (49) are women. Women achieved their highest proportion of seats in the PARLACEN (25%) and their lowest in the National Assembly departmental race (10%).
What political parties do elected women belong to? Chart XII indicates just that. Of the 49 women elected, 51% belong to the Liberal Alliance, 45% to the FSLN and the remaining 2% (two women) to the PLI and to UNO 96. It should be pointed out that while 92% of the Liberal Alliance women elected will occupy local posts (mayors and vice mayors), 54% of those elected for the FSLN will occupy high level posts: national and departmental representatives to the National Assembly and PARLACEN. This is largely because the FSLN reached an internal party resolution to alternate men and women candidates on all such slate elections. For this same reason, the number of Sandinista women would probably increase if Municipal Councilors were included in the above charts.
Straight or Split Ticket?Not all Nicaraguan citizens voted for the same party in all six electoral ballots. Some practiced ticket splitting: distributing their six votes among various parties, alliances or popular subscription associations. This type of vote allowed Arnoldo Alemán to increase the percentage by which he won the presidential election.
To offer an approximation of how extensive ticket splitting was, the envío team designed a coefficient that attributes the value of 100 to the presidential election results of each party and thus serves as a reference point to determine the proportion to which the results obtained that same party in other elections were equal to, less or greater than 100. Chart XIV on the next page presents the values obtained using this methodology. To simplify the presentation, we only included those parties that won seats in the National Assembly or PARLACEN.
As can be observed in the chart, the parties whose results show the least degree of variation between elections are the FSLN followed by the Conservative Party of Nicaragua (PCN). This would mean most Sandinistas and Conservatives did not split their tickets, but rather voted for the same party in all four elections shown in the chart.
The other two parties whose sympathizers tended to vote a straight party ticket were the CCN and the Liberal Alliance, although this type of voting was somewhat more frequent among CCN followers than among Alemán supporters. And compared to the Sandinistas and the Conservatives, both the Liberals and the evangelicals of the CCN split their tickets more frequently.
All other parties in the chart, in contrast, show a strong pattern among their supporters of switching to other candidates for president and voting for their own only in the other elections. The highest proportions of crossover votes are in PRONAL (particularly in the elections for national and departmental representatives), the PRN (in elections for departmental representatives) and the MRS (in elections for national representatives). Based on the information in the chart it can be stated that, in general terms, ticket splitting tended to be greater among the sympathizers of the smaller parties or alliances than the larger ones. (It should be taken into account that using the base of 100 on the extremely small number of presidential votes garnered by these six parties makes the crossover appear quite exaggerated.)
How Many "Ghosts"?We want to finish this analysis where we began, at the CSE coverup of the electoral chaos. We choose to conclude this way because it is important to focus on the real national dimensions of the inconsistencies discovered in the CSE data base.
Does the CSE's manipulation of data hide some proportion of the magnitude of irregularities that took place on October 20? A preliminary and exploratory analysis of the CSE data base--and therefore one still subject to later verification--allows us to estimate that the departments or regions appearing to have the greatest number of "ghost" JRVs could be significantly higher than the percentage observed in the case study of the department of Managua with which we began our reflection.
The evidence seems to indicate that the percentages of "ghost" JRVs in the departments of Chinandega, Chontales and Nueva Segovia are just under that of Managua. There are also "lost" JRVs in the rest of the departments, but to a lesser degree.
The CSE owes us many explanations. It would be healthy if we could shed some light on all these big and troublesome realities. For the good of Nicaragua.