Preamble
The Nepali Congress’s 15th convention was a hopeful sign of a party trying to turn the page. But a closer look at the numbers reveals a problem every Nepali party should care about.

Delegates appear to have voted past the old factional panels and unseated familiar leaders.

Crowded races split the vote: where four or more stood, the typical winner took about 30%.

Capable candidates lost not for lack of support, but because too many ran for the same seat.

Simple fixes exist: modest entry thresholds, ranked ballots, and a smaller central committee.

Democracy began in Athens with imperfect rules and survived over the millennia through trial and error. The time has come for Nepal’s parties to do the same. This essay looks at what went wrong and offers a few practical fixes.

A curious convention
I have written about internal party democracy for a long time, often taking the Election Commission to task for not fulfilling its constitutional duty to ensure democracy inside our parties. A non-partisan democracy watch group, and an Internal Party Democracy Index to track how parties choose their candidates, officers and leaders, have long been my wishful thinking.

So I watched the Nepali Congress’s 15th general convention with some interest. Granted, the convention was not without controversy: a rival faction led by former president Sher Bahadur Deuba disputes its legitimacy and plans a convention of its own.

Still, something unusual seems to have happened there, all the more striking for an entrenched legacy party, one with a long record of democratic struggle and its share of turmoil.

More than 7,000 delegates voted for 229 positions. Over 800 candidates stood. Counting took a week. And the results did not seem to follow the old script: the 60-40 panel coalition, voting in blocs, that so often fractured the party along the lines of my votes, my bloc, my share, and, of course, the legacy claims.

Consider the results. A sitting general secretary, Gururaj Ghimire, finished fourth. A sitting vice-president, Dhanraj Gurung, finished fifth. Former joint general secretaries lost to newer faces. The old habit of voting by panel, where a faction hands delegates a list and they tick it, seemed weaker than at any convention I can remember. And so far I have not seen many reports of delegates being bused in, put up and paid for. The convention is still fresh, though; such stories may surface later. Let’s hope not.

Still: is this a sign of hope? It may be. But hope is not the same as electoral design. So I want to dig into the numbers.

When I looked closely, I noticed a few cracks.

Several winners took barely a quarter of the vote. Capable people lost because too many of them stood for the same seat. And at least one delegate reported receiving a secret note telling him whom to vote for. Gagan Thapa himself said as much after the convention: his supporters lost seats because enthusiastic candidates divided the same votes.

Those cracks are worth studying, not as complaints but as lessons in two things: candidate proliferation and vote fragmentation. Hence my title: more candidates, less choice!

To understand them, it doesn’t hurt to go back to where the idea of voting began: to Athens, and its version of the chowk, the agora, some 2,500 years ago.

And we need not look only to Greece. In the same centuries, the Shakya assembly at Kapilavastu settled questions of war and peace by a show of hands and wooden voting chips. It too was a republic of sorts, where its clan leaders deliberated and voted.

But let me take you to ancient Athens: to its mechanics of direct democracy and the philosophy behind them. Refined over centuries in the West, those ideas became part of the bedrock of liberal democracy around the world.

What Athens knew
Why go back as far as 500 BC, and why Athens? Its world was brutal. City-states like Athens and Sparta, and the great empires around them such as Persia, were locked in cycles of war, plunder and subjugation. Yet out of a deep desire for freedom, and a hatred of tyrants, the Athenians found ways to practise liberty and choice.

Those seeds took root again and again, in different forms. Rome built its own republic. In England, the barons forced King John to sign the Magna Carta in 1215, putting even a king under the law. The thirteen American colonies broke from the English crown with the Declaration of Independence in 1776 and began their own experiment. And France rose against its monarchy in the Revolution of 1789.

The first democrats of Athens did not start with a complex system of government. Citizens met in an assembly to decide matters directly, and to fill public offices they simply drew lots (lottery). Imperfect as it was, a remarkable product of human ingenuity was born: democracy.

Every year, names went into a stone allotment machine, the kleroterion, and the lot (a random drawing mechanism) chose the council and most magistrates. The logic was radical: any citizen could govern, and no faction could buy a lottery.

But Athens did not use the lottery for everything. Its ten generals were elected, and re-elected, because leading an army needed proven skill. The grip of faction was addressed too: Cleisthenes redrew the city’s ten tribes so that each mixed coastal, urban and inland districts, breaking the hold of regional clans.

So the people we credit with inventing democracy did not rely on one method. Nor did those who came after them, the Romans, the English and the French. Each adapted the idea to its own history and traditions, as modern democracies still do.

Athens’ seed of democracy was far from complete: women, slaves and foreigners had no voice. But its lessons are priceless. Democracy began in an incomplete form, at times even an extreme one, and it survived by refining its rules.

Loosely speaking, a well-functioning system of government, whether presidential or parliamentary, rests on a few features: a strong and efficient executive; institutional guardrails and the rule of law; representation of diverse voices, across gender, ethnicity and geography; and, along the way, a habit of learning, applying and repeating. A work in progress, indeed.

That is the spirit in which I want to understand our own democratic struggle, and of course this convention. My focus, to repeat, is on two issues: candidate proliferation and vote fragmentation.

What the numbers say
I ran a simple experiment with the publicly reported data, compiling the final tallies for every contested single-seat post with published results: twelve in all. A simple graph and a table revealed four patterns.

First, where two or three candidates stood, the median winner took about 72% of the vote. Where four or more stood, the median winner took about 30%. The seat for Europe’s overseas members drew nine candidates; its winner took just 22%. At the 14th convention in 2021, by contrast, every reserved joint-general-secretary race had only two or three candidates, and seven of the eight winners took more than half the vote. Openness invited more candidates, and more candidates meant thinner wins.

[Source and method —see the footnote at the end.]

The correlation between the number of candidates and the winner’s share is about -0.8, a strong relationship, and an intuitive one: as the field widens, the winner’s share of the vote shrinks. This simple observation has a profound implication, which I will turn to shortly.

Second, the two open general-secretary seats went to Pradip Paudel and Udaya Shumsher Rana. Dr Sunil Sharma finished third, just 73 votes behind Rana. One delegate told Ratopati that, fearing Sharma might win, they had been instructed to vote for Udaya and Pradip, and did. When delegates may tick two names, a disciplined group that agrees on the same pair can win both seats. Weaken the formal lists, and informal ones fill the gap.

Third, Binod Chaudhary won the inclusive vice-presidency with 1,745 votes, 87 ahead of Jeevan Pariyar: 28% of the votes cast for that seat. Much of the public debate has been about his wealth. I think that misses the point. A party needs successful entrepreneurs, and he won fairly under the rules as written. The real questions lie in the design. A seat reserved for groups historically left out is chosen by the whole convention, not by those groups. And with five candidates in the race, it can be won with about a quarter of the vote: the correlation graph in action.

The overseas seats show both problems at once: in four of the five regional seats, winners took a third of the vote or less, chosen by delegates in Nepal rather than by the diaspora they represent.

Fourth, there is the cognitive burden on voters. Office-bearer posts grew from 14 to 25, partly to make room for more leaders. Delegates faced 229 positions in all, and no delegate can weigh hundreds of candidates on their merits. The unintended consequence is predictable: a faction’s list becomes a rational shortcut. Expanding posts to accommodate leaders ends up strengthening the panel-voting culture.

How others manage crowded fields
Nepal is not the first place to face this. Parties elsewhere have learned to separate two questions: who may enter the race, and how the votes are counted. The first is about filtering: keeping the field to serious candidates without letting party bosses decide who is worthy. The second is about counting: making sure a small but disciplined bloc cannot win simply because the majority’s votes are scattered among many candidates.

The Congress leadership and the media are currently debating this issue.

I summarize below a few best practices from some democratic countries that may shed some light and provide some implementable solutions.

In Britain’s Labour Party, anyone who wants to run for leader must first be nominated by a fifth of the party’s members of parliament, and then win backing from local branches or trade unions. In Japan’s ruling Liberal Democratic Party, a candidate for party president needs the endorsement of 20 members of parliament. These rules keep the field small and serious. Given Nepal’s political culture, though, I would not be surprised if such a rule simply moved factional bargaining up to the parliamentary party. But, it does have its merit.

Canadian parties ask leadership candidates to collect member signatures from several provinces. A candidate with support in only one region cannot qualify. It is a modern echo of Cleisthenes mixing districts into each tribe.

South Korea’s parties often hold a preliminary round that trims a crowded field to about three candidates. The main vote then happens among those who have already shown real support.

Ireland, Malta and Australia’s Senate use ranked ballots. Voters number the candidates in order of preference. If a voter’s first choice cannot win, the vote moves to their second choice. Supporters of similar candidates no longer split their camp.

Two lessons stand out.

First, good filters test support and spread, not loyalty to a leader. Set at the right threshold, they also screen out marginal candidates and keep the field manageable.

Second, ranked ballots discourage candidates from winning by appealing only to a small, hardcore base. A candidate who is the second choice of most voters may represent the party better than one who is the first choice of a determined few. And it takes only one trip to the ballot box: voters rank their choices once, and the counting does the rest in rounds. (The appendix explains how it works.)

Some Lessons
Over the last three decades, Nepal has stayed on a democratic track, but largely an electoral one, built from the top down. Inside political parties, a handful of leaders consolidated power and gradually captured public institutions to sustain their party machines. I have called this the extraction–control nexus: entrenched hierarchies that control candidate selection, divide parties into factions, and sustain themselves through patronage, corruption and nepotism.

The Nepali Congress’s 15th convention offers a glimmer of hope of breaking that pattern, though one convention does not make a trend. But it also brought two challenges into sharper focus: candidate proliferation and vote fragmentation.

The lesson is not to limit competition, but to improve how it works. Modest nomination thresholds, drawing on the practices from other countries discussed above, could keep ballots manageable without handing control back to party bosses. Ranked-choice voting in crowded contests could give delegates greater freedom to express their preferences (see the appendix for a worked example).

Likewise, fewer elected positions could reduce the burden on voters and their dependence on factional lists. A Central Committee of nearly 300 members is unusually large. India’s Congress Working Committee, the executive of a party far bigger than ours, has 39 members, plus about 30 invitees. Japan’s ruling Liberal Democratic Party manages party affairs through a General Council of 25. South Korea’s Democratic Party is led by a Supreme Council of just seven. Large committees can also become instruments for rewarding loyalists and accommodating factions. For Nepali Congress, reducing to around 100 members may be a sensible step toward a leaner, more effective and accountable party organization.

More transparent rules for reserved seats and presidential nominations could further strengthen representation and accountability. But using the power of appointment to reward loyalists, especially those defeated in open competition, would send the wrong signal. Appointments should complement the delegates’ choices, not reverse them.

None of these changes is a cure. Factions will adapt, and new rules will bring their own complications. Why not begin with a few small experiments, learn from the results, and refine the system?

This is not just a Congress problem. Nor is the Congress the first to experiment. The RSP, a much younger party, has tried primaries, letting members have a say in choosing candidates. Each party is testing pieces of internal democracy; none has the whole design.

Basically, it concerns every Nepali political party. The convention data compiled for this essay could be a small beginning toward the Internal Party Democracy Index I have long wished to see.

Athens did not stop at the lottery. It kept refining its institutions. Changing who leads a party and changing how its leaders and candidates are chosen are two different reforms. The first happened this October. The second is the harder, more lasting work.

Dr. Alok K. Bohara, Emeritus Professor of Economics at the University of New Mexico, writes as an independent observer of Nepal’s democratic evolution through the lens of complexity and emergence science. His systems-policy essays on Nepal’s socio-economic and political landscape appear on Nepal Unplugged.

P.S. Something new: Khari-Pati.
Once a month, I’ll send a short, curious kurakani: one puzzle worth turning over, plus a lighter reflection. The first upcoming issue asks how Roman gold ended up in Indian ports two thousand years ago, and what that ancient trade can teach a landlocked country between two giants. It’s opt-in, so it won’t fill your inbox unless you ask. Sign up for Khari-Pati: http://nepalunplugged.Substack.com/s/khari-pati

Appendix A: The data.
Every figure in this essay comes from final tallies reported by the Kathmandu Post, Onlinekhabar, Ratopati, the Annapurna Express and the Himalayan Times. I compiled them using the Google search engine and AI into one table, with a note on each place the sources disagree. Errors and updates aside, the larger message of the essay should be the focus of attention. Download the data (PDF)

Appendix B: How a ranked ballot works
In a ranked ballot, often called instant-runoff voting, each delegate numbers the candidates in order of preference: 1 for their favourite, 2 for their next choice, and so on. Delegates vote only once. The counting then runs in rounds. If no candidate has more than half the first choices, the last-placed candidate is dropped, and each of those ballots moves to that voter’s next choice. This repeats until someone has a majority. Ireland uses this method to elect its president, and Australia to elect its House of Representatives (Wikipedia: Instant-runoff voting).

A worked example. Imagine 100 delegates choosing one person for a reserved seat from five candidates: A, B, C, D and E. The candidates and rankings are invented, but the first-choice shares are close to the real inclusive vice-president race. To win, a candidate needs a majority: 51 votes.

How the delegates rank them:

28 delegates: A first, then E

26 delegates: B first, then C

18 delegates: C first, then B

10 delegates: D first, then B

7 delegates: D first, then C, then B

6 delegates: E first, then A

5 delegates: E first, then B

A is the favourite of a committed bloc, but few others rank A at all. B is the first choice of fewer delegates but the second choice of many.

Round 1. First choices: A 28, B 26, C 18, D 17, E 11. No one has 51. Under today’s plurality rule, A would win here with 28%. Instead, E, the last-placed candidate, is dropped.

Round 2. E’s 11 ballots move to their next choices: 6 to A, 5 to B. Now A 34, B 31, C 18, D 17. Still no majority. D is dropped.

Round 3. D’s 17 ballots move: 10 to B, 7 to C. Now B 41, A 34, C 25. Still no majority. C is dropped.

Round 4. C’s 25 ballots all rank B next. Final count: B 66, A 34. B wins with a clear majority.

What the example shows. The candidate with the most first choices is not always the one most delegates can accept. A ranked ballot finds the candidate with the broadest support, in a single round of voting, without anyone handing out a list. It does not always change the result: when the leader is also widely acceptable, the leader still wins. But no one can win a seat with 28% while most delegates preferred someone else.

Sources
- [Kathmandu Post: vice-president and general-secretary results](https://kathmandupost.com/politics/2026/10/12/bhandari-acharya-and-chaudhary-elected-congress-vice-presidents)

- [Kathmandu Post: counting of all 229 positions](https://kathmandupost.com/national/2026/10/16/congress-completes-convention-vote-count)

- [Kathmandu Post: office-bearers raised from 14 to 25](https://kathmandupost.com/politics/2026/10/07/five-things-to-know-about-nepali-congress-general-convention)

- [Ratopati: results, reactions and the delegate’s account of instructions](https://english.ratopati.com/story/83411/congress-office-bearer-election-discussion-on-winner-chaudhary-and-loser-dr-sharma)

- [Ratopati: overseas seats](https://english.ratopati.com/story/83508/who-is-ahead-when-6877-votes-are-counted-for-the-public-relations-committee)

- [Onlinekhabar: presidential result](https://english.onlinekhabar.com/gagan-thapa-elected-nepali-congress-president.html)

- [Annapurna Express: joint general secretaries](https://theannapurnaexpress.com/story/67901/)

- [Himalayan Times: 14th convention joint general secretaries](https://thehimalayantimes.com/nepal/seven-of-13-nc-office-bearers-from-deuba-camp)

- All India Radio: https://www.newsonair.gov.in/congress-president-mallikarjun-kharge-reconstitutes-congress-working-committee

- Wikipedia, LDP General Council: https://en.wikipedia.org/wiki/General_Affairs_Committee_of_the_Liberal_Democratic_Party

- Asia Economy: https://view.asiae.co.kr/en/article/2026011113421933802

- Add Wikipedia, Instant-runoff voting: https://en.wikipedia.org/wiki/Instant-runoff_