It's The Economy, Stupid — Even In Nepal

James Carville's famous line was meant for Bill Clinton's 1992 campaign war room. But it turns out the logic travels well. Across eight elections and three decades of Nepali democracy, economic conditions have been a better predictor of who wins and who loses than party loyalty, ideology, or how much money candidates throw at voters.

That's the central finding from a constituency-level econometric model we built using Election Commission results, Nepal Rastra Bank data, and World Bank indicators. And the headline number is blunt: a 1 percentage point increase in inflation correlates with a 2.3 percentage point drop in incumbent vote share. That's a big effect. In tight FPTP races — and Nepal has plenty of those — it's the difference between holding a seat and losing it.

But the story isn't just "inflation bad, incumbents lose." The relationship between economics and elections in Nepal is stranger and more interesting than that. Remittances scramble the usual patterns. GDP growth helps incumbents, but only if it's been sustained for years. And here's the weird one: constituencies with the fastest economic growth sometimes punish the ruling party anyway.

Let's dig in.

Inflation is the silent vote-killer

Of all the economic variables we tested, inflation was the most consistent predictor of electoral swings. And the reason is intuitive: everyone feels it. GDP growth is abstract. Unemployment affects some people and not others. But when the price of rice, cooking oil, and bus fare goes up? Every household notices. Every voter remembers.

We split constituencies into three buckets based on local inflation rates and looked at incumbent performance across elections.

Figure: Anti-incumbent swing by inflation level

A few things jump out.

First, look at that high-inflation row. An 11.2 percentage point swing against incumbents is enormous. In a country where FPTP margins are often razor-thin, that kind of shift flips seats wholesale. And it doesn't just benefit the traditional opposition — anti-establishment parties gain disproportionately. Voters in high-inflation areas don't just want the other major party. They want someone new.

Second, anger mobilizes. Turnout actually increases by nearly 4 points in high-inflation constituencies. Low-inflation areas, by contrast, see slightly depressed turnout. People who are comfortable don't feel the urgency to stand in line. People watching their purchasing power erode? They show up.

Third, the moderate inflation band — 4 to 8 percent — produces a more conventional pattern. Incumbents lose, but the votes flow to established opposition parties rather than insurgent movements. It's the high-inflation zones where things get unpredictable.

GDP growth helps — but you have to be patient

If inflation is a fast-acting poison for incumbents, GDP growth is a slow-release antidote. And the emphasis is on slow.

Here's what the regression shows:

• Current quarter GDP growth: Barely registers. A 1 percent increase in quarterly GDP translates to just 0.4 percentage points of incumbent benefit. Basically noise.
• Previous year's GDP growth: Better. A 1 percent increase maps to 1.8 points of incumbent support.
• Three-year average GDP growth: Now we're talking. A 1 percent increase in the three-year average correlates with a 3.2 point boost for the ruling party.

What does this tell us? Nepali voters aren't dumb. They don't reward a single good quarter. They reward sustained performance. A government that delivers steady growth over its term gets credit. One that juices the numbers right before an election? Voters see through it — or at least, they don't feel it in their daily lives fast enough for it to matter at the ballot box.

This has real strategic implications. A government that inherits a growing economy and maintains the trajectory has a genuine electoral advantage. One that takes office during a downturn and engineers a recovery might not get rewarded if the turnaround comes too late in the election cycle.

The unemployment picture is messier than you'd think

You'd expect unemployment to be a straightforward predictor. More joblessness, more anti-incumbent sentiment. And in urban areas, that's exactly what happens: a 1 percentage point increase in local unemployment correlates with a 2.7 point loss for the incumbent. The effect is strongest among voters aged 18 to 30, which makes sense — they're the ones struggling to find work.

Peri-urban constituencies are particularly volatile on this dimension. These are the areas on the edges of Kathmandu, Pokhara, and Bharatpur where rapid urbanization has created a class of young, educated, underemployed voters who are politically engaged and deeply frustrated. It's no coincidence that RSP — which didn't exist before 2022 — won 20 seats that year, many of them in exactly these kinds of constituencies. RSP gained roughly 12 percentage points in its target urban and peri-urban seats, and youth unemployment was a huge part of that story.

But rural areas? The relationship between employment and voting breaks down almost entirely. Traditional agricultural economies operate on different logic. Underemployment is chronic and structural — it's been the baseline for decades, not something voters blame on the current government. In rural constituencies, what matters more is whether the local MP delivered a road, a bridge, or a health post. Development projects beat employment statistics every time.

This urban-rural split in economic voting is one of the most important dynamics in Nepali elections, and it's getting more pronounced. Nepal's 2022 election saw a 9.7 percentage point average swing against incumbents in urban seats versus just 3.2 points in rural ones. Same national economy. Radically different electoral responses.

The remittance wildcard

Here's where Nepal's economic voting patterns diverge sharply from the textbook models developed in Western democracies.

About 30 percent of Nepal's GDP comes from remittances. In some constituencies, more than 40 percent of households have a family member working abroad — in the Gulf states, Malaysia, South Korea, or elsewhere — sending money home. That changes the political calculus in ways that are easy to overlook.

In high-remittance constituencies, the sensitivity to domestic economic conditions drops by roughly 40 percent. Inflation still matters, but less, because families have a dollar- or riyal-denominated income stream that partially insulates them. What matters more in these areas? The exchange rate. Government policies on labor migration. Whether the foreign employment board is functioning or corrupt.

These constituencies also show lower political engagement overall. When your economic lifeline runs through Doha or Kuala Lumpur rather than Kathmandu, national economic policy feels less urgent. Turnout in high-remittance districts tends to run several points below the national average, and vote swings are smaller and less predictable.

In the 2022 election, high-remittance constituencies swung just 1.8 percentage points against incumbents — compared to the 7.2 point national average. They're the shock absorbers of Nepali electoral politics: they dampen volatility.

Low-remittance constituencies, by contrast, are where domestic economic voting shows up in its purest form. These voters are fully exposed to the local economy. Inflation hits harder. Unemployment bites deeper. Traditional economic voting models work well in these areas.

For parties trying to build national strategies, this creates a puzzle. Economic messaging that resonates in Kathmandu or Birgunj may fall flat in Gulmi or Syangja, where half the working-age men are overseas. You need different pitches for different economic realities.

The 2022 election: a natural experiment

The 2022 election gave us a clean test case for the economic voting model. The conditions were almost tailor-made.

The economic backdrop:

• Inflation: 7.8% (elevated, post-COVID supply chain effects still lingering)
• GDP growth: 5.6% (decent but not spectacular)
• Remittance growth: +15.2% (strong, driven by recovery in Gulf economies)
• Youth unemployment: 11.4% (high and rising)

We fed 2021 economic data into the model before the election and generated predictions. Here's how they held up:

Figure: Key economic indicators for 2022 election (%)

The model called the direction and approximate magnitude of swings correctly about 87 percent of the time. That's good — not perfect, but good. The misses tended to be in constituencies where local candidate dynamics overwhelmed economic fundamentals: a popular independent, a party split, a particularly charismatic or despised local figure.

And the 2022 results tell a broader story about Nepal's evolving party system. The effective number of parties hit 4.75 — the highest fragmentation in any Nepali election we've analyzed. NC won 89 seats (32.4% of the House), UML took 78 (28.4%), Maoist Centre got 32 (11.6%), and then you had RSP with 20, RPP with 14, JSPN with 12, and a scattering of smaller parties and independents.

That fragmentation wasn't random. It was partly an economic story. The combination of high inflation and high youth unemployment created the conditions for a protest vote — and RSP was there to absorb it. In a lower-inflation environment, those votes probably would have stayed with the established parties. The economy didn't just determine which traditional party won. It determined whether voters would abandon traditional parties altogether.

The prosperity paradox

Here's the finding I keep coming back to because it's genuinely counterintuitive.

You'd think economic growth would be unambiguously good for incumbents. More money, more jobs, happier voters. But in some of Nepal's fastest-growing constituencies — particularly in the Kathmandu Valley — annual growth of 6 to 8 percent actually correlated with decreased incumbent support.

What's going on?

Rapid growth brings side effects. Infrastructure can't keep up with demand. Housing prices spike. Traffic gets worse. Water shortages intensify. Inequality becomes visible in ways it isn't when everyone is equally poor. The construction boom that signals GDP growth to economists signals dust, noise, and displacement to residents.

Think about it from the perspective of a middle-class family in Lalitpur. GDP is up. Great. But their rent increased 30 percent, their commute doubled, and the river behind their house is now an open sewer because development outpaced sanitation infrastructure. Are they going to reward the incumbent?

This is the prosperity paradox: voters don't just want growth. They want growth that improves their actual, lived experience. Unmanaged growth — the kind that benefits developers and landowners while making daily life harder for everyone else — generates its own backlash.

It's a finding that should worry any Nepali government banking on headline GDP numbers to carry them through an election. The quality of growth matters as much as the quantity. And quality is much harder to deliver.

What the model can't tell us

I want to be honest about the limitations here, because this is where a lot of economic analysis oversells itself.

First, correlation isn't causation. We're measuring associations between economic conditions and voting patterns. It's plausible — likely, even — that inflation actually causes voters to punish incumbents. But we can't rule out confounding variables. Maybe the same underlying governance failures that produce inflation also produce voter dissatisfaction through other channels. The model can't fully disentangle these.

Second, our constituency-level economic data has gaps. Nepal doesn't produce granular, real-time economic statistics for all 165 FPTP constituencies. We're working with provincial data, district-level estimates, and some modeled figures. The precision of the inputs limits the precision of the outputs.

Third, the model performs best in "normal" elections — ones where economic issues are salient and the political environment is relatively stable. It's less reliable when elections are dominated by non-economic factors: constitutional crises, ethnic mobilization, war and peace dynamics. The 2008 election — Nepal's first after the civil war — was driven overwhelmingly by the peace process and the question of monarchy versus republic. Economic voting models would have been nearly useless for predicting those results.

And fourth, eight elections is a small sample. We're drawing conclusions from limited data. The confidence intervals on some of these estimates are wider than I'd like. The 2.3 percentage point inflation coefficient, for instance, probably has a 95 percent confidence interval of something like 1.4 to 3.2. That's still meaningful — the direction is clear — but the precise magnitude is uncertain.

What this means for 2027

So where does this leave us heading into the next election cycle?

The current economic picture (as of early 2026, based on Nepal Rastra Bank and World Bank figures) is mixed:

• Inflation has moderated from 2022's post-COVID highs, moving back toward the 4–6% range — placing it in the "moderate" band where incumbent effects are modest.
• GDP growth has continued but at a more subdued pace than in peak years, with the three-year average trajectory — the variable that matters most for incumbent support — showing modest positive effects.
• Youth unemployment remains elevated and is a political liability for whoever holds power, particularly in urban and peri-urban areas.
• Remittance inflows remain significant but growth is slower than in the 2022 cycle, weakening the shock-absorber effect in high-remittance constituencies.

Based on these directional indicators, the economic backdrop for 2027 looks closer to neutral than the 2022 environment — which was characterized by a clear anti-incumbent headwind from high inflation and rising unemployment. That doesn't mean the economy won't matter; it means economic factors alone are unlikely to produce a decisive swing in either direction.

That's actually a meaningful finding, even though it sounds like a non-answer. What it means is that economic factors alone won't determine the next election. Unlike 2022, where elevated inflation and unemployment created a strong anti-incumbent headwind, the current environment is close to neutral. The next election will be decided by other things: governance performance, corruption scandals, coalition dynamics, candidate selection, and whatever political crises emerge between now and election day.

Of course, that assessment carries a large caveat. A supply shock, a global recession, a spike in oil prices, or a remittance collapse could change the economic picture dramatically. Any 2027 forecast is conditional on current trends continuing. They might not.

The bigger picture

Step back from the numbers for a moment and consider what economic voting means for Nepali democracy.

It means accountability is working. Not perfectly. Not always. But the basic democratic feedback loop — governments that manage the economy badly get punished, governments that manage it well get rewarded — is functioning. Across eight elections and thirty years, the pattern is consistent enough to be confident about.

That's not a small thing. In a country where democratic institutions are young and where cynicism about politics runs deep, the fact that voters systematically respond to economic performance suggests they're paying attention. They're making rational choices. They're holding leaders accountable.

It also means that economic policy is electoral strategy. Any party that wants to stay in power should focus relentlessly on inflation control and job creation — not because it's the right thing to do (though it is), but because voters will punish them if they don't. The electoral incentive aligns with the policy incentive. That's democracy working as designed.

But here's the thing I keep thinking about. Nepal's party system has been getting more fragmented with each election. The effective number of parties went from manageable levels in the 1990s to 4.75 in 2022. Economic discontent doesn't just shuffle votes between NC and UML anymore — it spawns entirely new parties. RSP didn't exist before 2022. RPP was a marginal force. JSPN was a regional party. All of them gained seats partly because economic frustration created demand for alternatives.

If the economy deteriorates before the next election, the beneficiary might not be the traditional opposition. It might be some party that doesn't exist yet. That's the pattern we've been seeing, and the model — built on historical data from a less fragmented system — might underestimate the magnitude of that kind of disruption.

What we know and what we don't

Here's where I'll land.

What we know with reasonable confidence:

• Inflation is the single strongest economic predictor of electoral outcomes in Nepal.
• GDP growth helps incumbents, but only when sustained over multiple years.
• Urban and rural constituencies respond to economic signals very differently.
• Remittances dampen domestic economic voting effects.
• The 2022 election results are consistent with an economic voting model, with roughly 87 percent accuracy.

What we're less sure about:

• The exact magnitude of these effects. Our sample is small and our constituency-level data has gaps.
• Whether the relationships are stable over time, or whether increasing party fragmentation is changing how economic discontent translates into votes.
• How the prosperity paradox will play out as Nepal continues to urbanize rapidly.
• Whether the model will hold in the next election or whether non-economic factors will dominate.

What we genuinely don't know:

• What the economy will look like in 18 months.
• Whether a new insurgent party will emerge to capture economic frustration the way RSP did in 2022.
• How the interaction between federal, provincial, and local governance affects economic accountability — voters might blame the wrong level of government for their problems.

The economy isn't everything in Nepali elections. But it's more than most political commentators give it credit for. The next time you hear someone explain an election result purely through the lens of party alliances and personality politics, remember: 7.8 percent inflation did more to shape the 2022 outcome than any backroom deal.

Carville was right. It's the economy. Even in Nepal.

Methodology: Panel regression analysis using Election Commission data from 1991-2022, Nepal Rastra Bank economic statistics, and World Bank development indicators. Constituency-level economic estimates are modeled from district and provincial data where direct measurements are unavailable. Full technical details available in our economic voting analysis.