The debate on Fiji’s electoral system has been raging on for some months now. The discourse, however, has been disproportionately dominated by legal experts — the most notable being Prof Anthony Regan, Attorney-General Graham Leung, Richard Naidu and lately Daniel Fatiaki — whose analyses, while well-intentioned, reveal a fundamental misunderstanding of electoral system design. We have not been hearing much from the actual experts in this area – the political scientists!
I have responded to Prof Regan’s incorrect assertions on Fiji’s electoral system in an earlier article. Recently, Richard Naidu penned an op-ed in The Fiji Times where he described Fiji’s current OLPR system as “strange and unrepresentative” while suggesting it undermines democracy. In this article, I unpack those claims.
Claim 1: “Strange”
Calling OLPR “strange” betrays a lack of familiarity with its global pedigree. Open-list PR — where voters select candidates within party lists, and seats are allocated proportionally —is a mainstay in diverse nations like Brazil, Finland, and Indonesia.
According to International IDEA, List PR is the most popular electoral system in the world with 38 per cent of 217 countries using a form of List PR. Moreover, the PR system is used in the majority (53 per cent) of democratic (free) countries. Political science literature, overwhelmingly supports PR for divided societies, citing its ability to mitigate conflict through inclusivity — unlike winner-takes-all systems like FPTP or AV. Critics of the system in Fiji need to realise that Aiyaz Sayed-Khaiyum has not invented the PR system, rather that it is a tried and tested system that has proven itself in top democracies whose societies have benefited from its centripetal effects.
The AV system used in the 1997 constitution had led to some of the worst possible political outcomes for an ethnically divided society like Fiji, effectively leading to a two-party system based on race, directly contrary to the intentions of the architects of the 1997 constitution who had believed that embracing an AV system in favour of the previous FPTP will foster moderation in a polarised society. Fraenkel and Grofman have thoroughly dismantled the arguments of AV proponents — most notably Donald Horowitz, who strongly argued for its adoption in Fiji — by demonstrating how AV had spectacularly failed as a tool of ethnic moderation in Fiji.
Mr Naidu’s label that Fiji’s OLPR is strange, thus, seems more a rhetorical jab than a reasoned critique, a perspective that overlooks the nuanced design considerations political scientists weigh.
Claim 2: “Unrepresentative”
The charge of unrepresentativeness collapses when measured against the Gallagher Index. The Gallagher Index, also known as the Least Squares Index (LSq), is a widely used measure in political science to assess the disproportionality between the percentage of votes a party receives in an election and the percentage of seats it wins in a legislature.
The index is calculated using the following formula:
Where:
• Vi = the percentage of votes received by party i
• Si = the percentage of seats won by party i
• n = the number of parties
The score is interpreted as follows:
• 0-5: Highly proportional (e.g., pure PR systems like South Africa).
• 5-10: Moderately proportional (e.g., mixed systems like Germany’s MMP).
• 10+: Highly disproportional (e.g., FPTP systems like the UK or Fiji’s 1997 AV).
Below, I have used the results of Fiji’s 1999 and 2014 general elections to calculate the Gallagher index. These were the first elections held under two contrasting electoral systems, and this analysis show how design choices impact outcomes in a divided society.
Fiji’s 1999 Election: High Disproportionality Under AV
The 1999 Fijian general election, held under the 1997 Constitution, introduced the Alternative Vote (AV) system across 71 single-member constituencies. Of these, 46 seats were ethnically reserved (23 for iTaukei, 19 for Indo-Fijians, 3 for General Electors, 1 for Rotuma), with 25 open seats. Voters ranked candidates, and preferences were redistributed until a majority winner emerged.
Election Results:
Gallagher Index Calculation:
1. Differences squared: FLP: (32.2−52.1)2=396.01, SVT: (19.6−11.3)2=68.89, FAP: (10.1−15.5)2=29.16 , NFP: (14.5−0.0)2=210.25, PANU: (4.4−5.6)2=1.44, VLV: (9.7−4.2)2=30.25, NVTLP: (4.4−1.4)2=9.00, Others/Independents: (9.5−9.9)2=0.16
2. Sum: 745.16
3. Divide by 2: 372.58
4. Square root: 19.3
The Gallagher index of 19 shows very high disproportionality. The FLP, with just around a third of the vote, secured over half the seats, while the NFP, despite 14.5% support, won nothing. SVT, the incumbent iTaukei-led party, was the other casualty of this system.
Why So disproportional?
The AV system, intended to moderate ethnic divisions, backfired. Ethnic voting continued but the distortions due to the AV system was a key reason leading to a landslide for FLP despite its narrow vote share. This winner-takes-all effect, combined with reserved seats entrenching identity politics, produced a Gallagher Index far higher than proportional systems like New Zealand’s MMP (typically 2-5).
Fiji’s 2014 Election: A shift to proportionality with PR
The 2014 election, under the 2013 Constitution, marked a radical shift to an open-list Proportional Representation (PR) system in a single nationwide constituency with 50 seats. Ethnic reservations were abolished, voters chose one candidate, and seats were allocated proportionally to parties surpassing a 5% threshold using the D’Hondt method.
Election Results:
Gallagher Index Calculation:
1. Differences squared: FijiFirst: (59.2−64.00)2=23.04, SODELPA: (28.2−30.00)2=3.24, NFP: (5.5−6.00)2=0.25 , PDP: (3.2−0.00)2=10.24, FLP: (2.4−0.00)2=5.76 , OFP: (1.2−0.00)2=1.44,
FUFP: (0.2−0.00)2=0.04, Independents: (0.3−0.00)2=0.09
2. Sum: 44.10
3. Divide by 2: 22.05
4. Square root: ≈4.7
A score of 4.7 reflects moderate to high proportionality, typical of PR systems with thresholds. It’s higher than pure PR systems like Sweden’s 2022 elections (0.64) but far lower than majoritarian systems like Fiji’s 1999 AV (19) or the UK’s FPTP (typically 10-15, increased to 23.8 in 2024!). The 5% threshold contributes significantly: 7.21% of votes (35,727) went to parties with no seats, inflating the index. Without this, the score would approach 2-3, closer to ideal proportionality.
Why More Proportional?
OLPR eliminated single-member distortions and ethnic quotas, forcing parties to appeal nationally. The 5% threshold did distort proportionality, but the Gallagher Index rivaled proportional benchmarks like Sweden’s 2022 election (0.64) or Ireland’s STV (1.96), a vast improvement over 1999.
Comparative Insights
• 1999 (AV): A Gallagher Index of 19 mirrors majoritarian systems like the UK’s FPTP (often 10-15), where concentrated support trumps broad representation. It failed Fiji’s divided society by amplifying ethnic silos.
• 2014 (PR): A score of 4-5 aligns with mixed or PR systems (e.g., Germany’s MMP at 5-10), balancing stability and fairness. It reduced zero-sum outcomes, and moderated impacts of ethnic politics.
The 2014 PR system’s lower score reflects a successful shift to inclusivity, mirroring voter intent more faithfully. I have demonstrated earlier in 2019 that parties seen as moderates (NFP, in Fiji’s case) have been the main beneficiary of the PR system in Fiji, reflecting the system’s capacity to reward platforms that bridge, rather than exploit, ethnic divisions.
There is growing dissatisfaction with majoritarian systems (like FTP and even AV) globally given the unrepresentativeness of the electoral outcomes it has been producing. An insightful read for those commenting on Fiji’s electoral system would be the Electoral Reform Society’s report titled “A System Out of Step: The 2024 General Election” where they have examined how the FPTP system produced one of the most unrepresentative outcomes in UK’s 2024 elections.
Conclusion
Fiji’s OLPR is neither strange (it’s globally tested), nor unrepresentative (Gallagher Index proves it), nor less democratic. Fiji’s electoral system isn’t perfect—thresholds could be debated—but it’s a marked improvement over the AV system in the 1997 constitution, delivering a fairer, more democratic outcome for an ethnically divided nation. Those who critique Fiji’s OLPR should engage with the wealth of political science research confirming the efficacy of proportional systems in divided societies. Dismissing it as “strange” or “undemocratic” is not only inaccurate but risks undermining one of Fiji’s most promising institutional reforms in decades.
