Nepal's Electoral System Gives Every Voter Two Ballots. Here's Why That Matters More Than You Think.

When a Nepali voter walks into a polling station, they get two ballot papers. One lets them pick a person. The other lets them pick a party. Those two choices feed into two completely different counting systems that produce two different sets of winners — and the combination of the two is what determines who runs the country.

This isn't a quirk. It's a deliberate design choice baked into the 2015 Constitution, and it has enormous consequences for how power gets distributed. In 2022, the system helped produce the most fragmented parliament in Nepal's democratic history, with an effective number of parties hitting 4.75 — a level of splintering that made coalition government not just likely but mathematically inevitable. In 2017, the same system helped UML claim 121 of 275 seats (44%) on far less than 44% of the vote.

The rules of the game shape who wins. And in Nepal, the rules are more interesting than most people realize.

Two elections in one

Here's the basic setup. Nepal's House of Representatives has 275 seats. Of those, 165 come from First-Past-The-Post (FPTP) districts — single-member constituencies where the candidate with the most votes wins, full stop. The remaining 110 seats come from Proportional Representation (PR), allocated nationally based on party vote shares.

Every voter participates in both. Ballot one: pick your local candidate. Ballot two: pick your preferred party. The two choices don't have to match. You can vote for a Nepali Congress candidate in your district and then cast your PR ballot for UML. Plenty of people do exactly that, which is part of what makes the results so hard to predict.

The FPTP side is straightforward. If you've followed elections in the UK, Canada, or India, you know how this works. One hundred sixty-five districts, one winner each, most votes wins. No runoff. No minimum threshold. If there are eight candidates and you win with 22% of the vote, congratulations — you're a member of parliament.

The PR side is where things get mathematically interesting.

The FPTP problem (and why Nepal added PR in the first place)

Nepal ran three elections under a pure FPTP system: 1991, 1994, and 1999. All 205 seats (the chamber was larger then) were decided by winner-take-all district races. And the results showed exactly the distortions that political scientists warn about.

In 1991, the Nepali Congress won a comfortable majority. In 1994, UML won the most seats but not a majority, producing a short-lived minority government. By 1999, the system was producing parliaments where a party could win 36% of the national vote and end up with well over half the seats. Smaller parties — including those representing Madheshi communities and ethnic minorities — were systematically shut out.

The decade-long Maoist insurgency (1996-2006) scrambled everything, and when Nepal held its first post-conflict election in 2008, the system had changed. The Constituent Assembly that year had 601 members: 240 from FPTP districts and 335 from PR lists, plus 26 nominated members. The PR component was deliberately oversized. The message was clear: the old winner-take-all system had failed to represent Nepal's diversity, and proportionality was the fix.

The 2013 Constituent Assembly used the same structure. But when the 2015 Constitution established the permanent House of Representatives, the ratio flipped. FPTP got 165 seats. PR got 110. The winner-take-all component was back in the driver's seat — 60% of the chamber versus 40% for PR.

That ratio matters a lot. It means FPTP still dominates outcomes. Big parties with geographically concentrated support get rewarded. But the PR component acts as a partial corrective, giving smaller parties a foothold they'd never get in a pure FPTP system.

Source: Election Commission of Nepal. 2008 and 2013 figures exclude nominated members.

Figure: Effective number of parties over time (higher = more fragmented)

How the Sainte-Laguë method actually works

OK, here's where I need to get a little nerdy. Bear with me — this is the part that actually determines who gets the PR seats, and it's not as complicated as it sounds.

Nepal uses something called the Sainte-Laguë method to convert party vote shares into seats. It's named after André Sainte-Laguë, a French mathematician, and it works like this:

Take each party's total PR vote count. Divide it by 1. That's the first quotient. Then divide by 3. That's the second. Then by 5, then 7, then 9 — always odd numbers. Keep going until you've generated enough quotients for every party.

Now line up all the quotients from all the parties, biggest to smallest. The top 110 quotients each win a seat for their respective party.

Let me walk through a simplified example. Say there are three parties and 10 seats to allocate:

• Party A: 100,000 votes
• Party B: 80,000 votes
• Party C: 30,000 votes

The quotients would look like this:

Now rank the top 10 quotients: 100,000 (A), 80,000 (B), 33,333 (A), 30,000 (C), 26,667 (B), 20,000 (A), 16,000 (B), 14,286 (A), 11,429 (B), 11,111 (A).

Result? Party A gets 5 seats, Party B gets 4, Party C gets 1.

With perfect proportionality, Party A would get 4.76 seats, Party B 3.81, and Party C 1.43. So Sainte-Laguë gets pretty close — Party C is slightly underrepresented, but not badly.

Why Sainte-Laguë and not D'Hondt?

This is a real choice with real consequences. The other common method, D'Hondt, divides by 1, 2, 3, 4, 5 — sequential integers instead of odd numbers. The difference? D'Hondt systematically favors larger parties. Sainte-Laguë is more generous to mid-sized and smaller parties.

Nepal's choice of Sainte-Laguë was intentional. The whole point of adding PR was to give smaller parties and marginalized communities a voice. Using D'Hondt would have partially defeated that purpose. In a country with as many ethnic groups, languages, and regional identities as Nepal, that choice matters.

How much does it matter? In practice, the difference between Sainte-Laguë and D'Hondt might shift 3-5 seats in a 110-seat allocation. That's not nothing — in 2022, the difference between the fourth and seventh largest parties was just 8 seats. A few seats can determine whether a party is a viable coalition partner or an afterthought.

The 3% threshold: a gatekeeper with teeth

Before any of the Sainte-Laguë math kicks in, there's a filter. Parties must win at least 3% of the total national PR vote to qualify for any PR seats at all. Fall below that line and your PR votes evaporate — redistributed, effectively, to the parties that cleared the bar.

Three percent sounds low. It's not.

In 2022, roughly 10.56 million valid PR votes were cast. Three percent of that is about 317,000 votes. For a small regional party, that's a steep hill. A party could win 250,000 votes — a quarter-million people choosing them — and get zero PR seats for it.

The threshold creates a strategic dilemma for voters. If you support a small party that's polling at 2.5%, do you vote for them on the PR ballot and risk wasting your vote? Or do you strategically vote for a larger party that you like less but that will actually clear the threshold? Political scientists call this the "wasted vote" problem, and it's real in Nepal.

The threshold also creates weird incentives for parties. In 2022, the Rastriya Swatantra Party (RSP) — a genuinely new force in Nepali politics — cleared the threshold comfortably and won 20 seats total. But other small parties that might have won 1-2 PR seats under a threshold-free system got nothing. The threshold rewards parties that can build national support and punishes those with geographically concentrated but nationally thin bases.

Is 3% the right number? That's debatable. Germany uses 5%. Israel uses 3.25%. The Netherlands has effectively no threshold. Lower thresholds produce more fragmented parliaments. Higher thresholds produce more concentrated ones. Nepal's 3% is moderate by international standards — but in a country with deep regional and ethnic diversity, even a moderate threshold excludes voices that might otherwise be heard.

The gap between votes and seats

Here's where the two systems collide — and where things get really interesting.

Because FPTP and PR operate on different logics, a party's total seat count can look very different from its total vote share. FPTP rewards geographic concentration. If your voters are spread evenly across the country, you might win 20% of the vote in every district and get zero FPTP seats. But if they're clustered in specific areas, you could win 30% of the national vote and take 45% of FPTP seats.

Look at 2017. UML won 121 of 275 seats — 44% of the chamber. That's a massive haul. But their national vote share was considerably lower. The FPTP system amplified their support into a disproportionate seat count, which is exactly what FPTP does in every country that uses it. The PR component partially corrected for this, but since PR only controls 110 of 275 seats, the correction was incomplete.

Now look at 2022. NC won 89 seats (32.4%), UML won 78 (28.4%), and then things got fragmented fast: Maoist Centre took 32 (11.6%), RSP got 20 (7.3%), RPP won 14 (5.1%), and JSPN picked up 12 (4.4%). Six parties with meaningful seat counts. No one close to a majority. The effective number of parties — a standard measure of fragmentation — hit 4.75, the highest in Nepal's history.

Key stat: The 2022 election produced a parliament where the largest party held just 32% of seats — the lowest for any leading party in Nepal's democratic era.

What happened? Partly it was voter behavior — people genuinely wanted more options. But the electoral system shaped those preferences into outcomes. FPTP split unpredictably across multiple competitive parties, and PR rewarded several mid-sized parties that cleared the 3% threshold. The math made coalition government mandatory.

How the two ballots interact

Something that doesn't get enough attention: voters can and do split their tickets. You might vote for a well-known local NC candidate on your FPTP ballot because you trust them personally, then vote for RSP on your PR ballot because you like their national platform. This happens all the time, and it means the FPTP and PR results can tell very different stories about what voters actually want.

In theory, the PR ballot is the purer signal of national party preference — it's not contaminated by the personal popularity of individual candidates or the tactical considerations of district-level races. In practice, the FPTP ballot gets more attention because those races are more dramatic. Everyone knows who won their district. Fewer people understand how the PR math works.

This creates an information asymmetry that benefits big parties. NC and UML have the name recognition and organizational muscle to win FPTP districts. Smaller parties often do better on the PR ballot, where they can appeal to voters on ideology and platform without needing a strong local candidate in every district.

What the system gets right

Let's give credit where it's due. Nepal's mixed system does several things well.

First, it balances local and national representation. Your FPTP member of parliament is someone from your area who (in theory) represents your community's specific interests. Your PR members represent broader national and demographic constituencies. You get both.

Second, the PR component has meaningfully increased the diversity of parliament. The PR lists are required to include women, Dalits, indigenous groups, Madheshi communities, and other historically marginalized populations. This isn't just aspirational — it's mandatory. And it works. Nepal's parliament is more demographically representative than it would be under pure FPTP, where candidates from dominant groups tend to win district races.

Third, the system allows new parties to break through. RSP's 2022 performance is the clearest example. A brand-new party, running for the first time, won 20 seats. Under pure FPTP, they might have won a handful of districts at best. The PR component gave them a platform to convert national support into real parliamentary power.

What the system gets wrong

But the system has real problems too.

The 60-40 FPTP-PR split means winner-take-all dynamics still dominate. If the goal was proportionality, Nepal's system only partially delivers. A party that wins 25% of the national vote might end up with anywhere from 15% to 35% of total seats, depending on how their support is distributed geographically. That's a wide range, and it means election outcomes depend heavily on factors that have nothing to do with how many people support a given party.

The two-ballot system is also genuinely confusing for some voters. Nepal's literacy rate has improved dramatically, but explaining Sainte-Laguë divisors to a first-time voter in a remote hill district is... a challenge. Voter education efforts have helped, but there's still a gap between how the system works and how voters understand it.

And then there's the coalition problem. The 2022 parliament required a coalition to govern. So did the parliament after 2017, effectively. Coalition politics isn't inherently bad — it forces compromise and power-sharing. But Nepal's coalitions have been notoriously unstable. The country has had more than a dozen prime ministers since 1990. The electoral system doesn't cause that instability directly, but by making majority governments nearly impossible, it creates the conditions for it.

The fragmentation question

Is Nepal's parliament getting too fragmented? The data suggests a clear trend.

In the 1990s, under pure FPTP, Nepal essentially had a two-party system — NC and UML dominated, with smaller parties on the margins. The effective number of parties hovered around 2-3.

The Constituent Assembly elections of 2008 and 2013 introduced the Maoists as a major third force. By 2017, three parties controlled 86% of seats (NC with 63, UML with 121, Maoist Centre with 53). That's concentrated power.

Then 2022 blew things open. Six parties won 10 or more seats. The effective number of parties jumped to 4.75. RSP came from nowhere. RPP re-established itself. JSPN held its ground in Madhesh.

The question is whether this fragmentation is temporary or structural. If RSP consolidates and grows, Nepal might settle into a four-party system. If it splinters or fades, the system could revert to three-party dominance. The electoral system itself is agnostic — it can produce either outcome depending on how voters behave and how parties organize.

But here's what the system does guarantee: as long as the 3% PR threshold exists and FPTP controls 60% of seats, Nepal will almost certainly never return to single-party majority government. The math just doesn't support it unless one party can dominate FPTP districts while also winning the PR vote — and in a country this diverse, that's extremely hard to do.

What we'd see under alternative systems

It's worth thinking about what Nepal's parliament would look like under different rules.

Pure FPTP (like the 1990s): Bigger parties would dominate even more. RSP probably wins 5-8 seats instead of 20. Madheshi parties might be wiped out in districts where they're competitive but not dominant. Parliament would be less diverse, less representative, but potentially more stable.

Pure PR (like the Netherlands or Israel): Every vote counts equally regardless of geography. Small parties thrive. Nepal might have 8-10 parties in parliament. Coalition-building becomes even more complicated, but representation is maximized. The 3% threshold would still filter out the smallest parties.

Different FPTP-PR ratio: If Nepal flipped to 40% FPTP and 60% PR (like the Constituent Assembly elections), smaller parties would gain significantly. The corrective effect of PR would be stronger. But geographic representation would weaken.

None of these alternatives is obviously better. They involve trade-offs — between representation and governability, between local accountability and national proportionality, between inclusion and fragmentation. Nepal's current system is a compromise, and like most compromises, it satisfies no one completely.

The provincial layer adds complexity

One more thing. The 2015 Constitution didn't just create a new electoral system for the federal parliament — it created seven provincial assemblies, each with their own mixed FPTP-PR elections. Provincial elections happen simultaneously with federal elections (they did in 2017 and 2022), and the results don't always match.

A voter might support NC for their federal FPTP candidate, RSP for their federal PR ballot, and UML for their provincial race. Ticket-splitting across three different ballots creates outcomes that can seem contradictory. A province might have a UML-led government while the federal districts in that same province sent NC members to Kathmandu.

This isn't a bug in the system — it's federalism working as designed. But it does make Nepali politics harder to read from the outside. You can't just look at federal results and assume you understand what's happening at the provincial level.

What we know and what we don't

Here's what we can say with confidence: Nepal's mixed electoral system produces parliaments that are more representative than pure FPTP would, but less proportional than pure PR would. The Sainte-Laguë method and 3% threshold work together to give mid-sized parties a fair shot while filtering out very small ones. The system structurally favors coalition government, which means the post-election bargaining over who becomes prime minister is often more consequential than the election itself.

What we don't know is whether the current 60-40 FPTP-PR ratio is the right one. We don't know if the 3% threshold is set at the optimal level. We don't know if the increasing fragmentation we saw in 2022 will continue or reverse. And we don't know whether Nepali voters, given more experience with the system, will become more strategic in how they split their tickets — or less.

The electoral system is the plumbing of democracy. Most people don't think about plumbing until something breaks. But in Nepal, the plumbing determines whether a party with 28% of the vote gets to lead the government or sits in opposition. It determines whether Madheshi communities and indigenous groups have a voice in parliament or get shut out. It determines whether new parties like RSP can break through or get crushed by the big two.

The math matters. And in Nepal, the math is more interesting — and more consequential — than almost anyone gives it credit for.