Dear Editor,
I am Swiss-Guyanese and a non-partisan observer of Guyana’s politics and its economic development. I have followed the 2020 General Elections and its turbulent aftermath with great interest. The elections were generally considered to be free and fair, but the process of tabulating the votes was widely seen as fraudulent. Here, I would like to draw your attention to the reported voter turnout of 70.24% for the 2020 General Elections by the Guyana Elections Commission (GECOM). I believe that the number is misleading and that more realistic voter turnout estimates by myself (see below for details) and independently by the International Institute of Democracy and Electoral Assistance (IDEA) indicate numbers well above 90%. Such voter turnout numbers appear abnormally high in comparison to other liberal democracies.
All 65 seats of Guyana’s National Assembly are elected using closed list proportional representation. Seats are allocated using the Hare quota, where the total number of votes is divided by the number of seats to be filled. Given that the recount resulted in 460,352 valid votes, it takes 7,082 votes to gain one seat in the National Assembly. Ever since 1992, when the first free and fair elections were held since 1964, Guyanese elections have been closely contested between the two largest parties, the People’s National Congress (PNC) representing Guyanese of African or mixed-race heritage and the People’s Progressive Party (PPP), which is largely supported by Guyanese of Indian ancestry. For example, the 2015 general elections ended with a narrow victory for the APNU-Alliance for Change (an alliance of PNC and AFC) by one seat over the incumbent PPP government. Gaining one seat more, i.e. 7,082 votes, can make the difference, whether or not a party wins the general elections. It is therefore essential that all measures be taken to ensure that only eligible voters cast their vote once on Election Day.
Below, I have laid out a case documenting concerns that more votes could have been cast than reasonably possible. This observation requires close attention and further probing by the Commission of Inquiry (CoI) into the March 2020 General and Regional Elections. Although, I have to concede that it will be difficult to resolve the issue conclusively given that two and half years have passed since Election Day on March 2, 2020. Finally, I have made three key recommendations, which can be implemented fast, to ensure that future general elections are less subject to manipulations and that election results are reported faster to the Guyanese and international public. I had submitted on November 2, 2022, via email a similar letter raising concerns about the voter turnout calculations to the Commission of Inquiry. I had requested a confirmation from the Secretary of the CoI twice, but I have not received a response to date. The recount of the 2020 General Elections was completed by June 8, 2020. The total votes cast and the voter turnout reported by Guyana Elections Commission (GECOM) are as follows:
According to Guyana’s election laws any adult (18 years and older) holding Guyanese citizenship, who has been registered in the electoral list and is living in Guyana on election day, is eligible to vote in the general elections. The number of 661,378 registered voters reported by GECOM for the 2020 General Elections also includes many Guyanese living abroad, which for the most were not able to travel to Guyana and participate in the elections. The reported voter turnout of 70.24% is therefore too low. Hence, the number of registered voters cannot be used to calculate the voter turnout on Election Day, March 2, 2020. One needs to turn to the demographics of Guyana in 2020, particularly the number of adult Guyanese citizens, in order to estimate more realistically the voter turnout. Given that the last national census was carried out in 2010, all numbers given below are based on estimates.
Using the estimated number of adult Guyanese (18+) and the votes cast as reported by GECOM (see table above), the calculated voter turnout would amount to 91.3%. This number is essentially identical with the 91.4% reported for the 2020 general elections in an analysis of the voting and population trends from 1964 to 2020 in Guyana by the Sweden-headquartered International Institute for Democracy and Electoral Assistance (IDEA, https://www.idea.int/data-tools/country-view/121/40).
A voter turnout of 91.3% for the 2020 General Elections is likely to be an underestimation for the following reasons: (a) Not all voting-age Guyanese will have made the effort to register as voters, (b) Voting is not compulsory in Guyana, (c) Voting by mail is not permitted and (d) Eligible voters might be ill, living too distant from a polling station, or otherwise unable to cast their vote. To my knowledge, there is no accurate registry of all who are eligible to vote in Guyana and the voter registration rate is unknown. In the United States, for example, only about 70-75% of the population chooses to register themselves. In Guyana, the voter registration rate might be higher than in the US but it is unlikely to reach 100%. Guyanese living abroad could travel to Guyana to cast their vote on Election Day. Given the limited number of international flights to Guyana per week, the number of Guyanese returning up to two weeks prior to election day, March 2, 2020, would amount at best to a few thousand extra voters. In conclusion, the estimated voter turnout of 91.3% for the 2020 General Elections, which is also supported by IDEA’s 91.4%, is likely to be an underestimation.
The question arises whether the estimated high level of voter turnout of 91.3% or more is realistic for a liberal democracy, where voting is not compulsory? In Singapore, where voting is compulsory, turnout at the 2020 general election was 95.8%. By contrast, the voter turnout for the second round of the recent Brazilian presidential elections of October 31, 2022, was 79.4% despite the fact that voting was mandatory. In the Barbados General Elections 2022, where voting was voluntary 114,035 votes were cast. Barbados has currently an estimated population of 281,740 people. There were 266,330 registered voters reported for 2022, which is higher than the estimated resident adult (18+) population of 223,335. Using the latter number to calculate the voter turnout for Barbados General Elections 2022, this would amount to a mere 51.1%. Differing methods of measuring voter turnout can contribute to reported differences between the voter turnouts when comparing different nations. There are difficulties measuring the numerator, the number of voters that cast votes, and the denominator, the number of voters eligible to vote. In practice, voter turnouts in liberal democracies without mandatory voting are rarely higher than 80%. This rule of thumb is likely to apply also to Guyana.
On the premise that the estimated voter turnout of 91.3% is likely to be unrealistically high and that voter turnouts in liberal democracies without mandatory voting are rarely above 80%, the expected number of votes to cast in the 2020 General Elections would have to be in the range of 406,994 votes (80% of 508,742). The 2020 General Elections was expected to be the most significant since Guyanese independence in 1966 because of new large oil discoveries off the coast of Guyana. Therefore, voter turnout might have been higher, possibly up to 90%. This would amount to 457,868 votes cast. The range of “excess” votes cast would therefore range between 6,695 and 57,569. These numbers are sufficiently large to tip the balance in favour of one of the two dominant parties. It takes as little as 7,082 votes to gain one seat in the National Assembly.
It is reasonable to assume that the Commission of Inquiry (CoI) into the March 2020 General and Regional Elections will not be able to resolve the above discrepancies regarding the actual voter turnout number. I would therefore strongly recommend that the CoI considers including the following three measures with their final report to the Government of Guyana:
Introduction of an electronic voting system
An electronic voting system will automate the tabulation process and minimizes the possibilities of human manipulation aimed at altering the vote count in favour of one or more parties. In addition, the elections’ results can be made public within a few hours after polls have closed. Brazil has been successfully using an electronic voting system for more than 25 years. The details of this system is outlined below.
Adoption of a biometric voter identification system
The electronic voting system should be complemented by a biometric voter identification system. This will ensure that every registered voter can only cast one vote on voting day. The biometric system will match the voter identity with the registry of eligible voters. It will curb repeat voting at multiple polling stations and effectively prevent voter impersonation.
Reform of the method to calculate the voter turnout
Voter turnout should be calculated on the basis of the total votes cast and the number of registered voters present in Guyana on Election Day. This measure will serve as an early warning system to detect any voting irregularities.
Sincerely,
Andre Brandli, PhD
Professor
Ludwig-Maximilians-University Munich