Israel's coalition system explained: why 61 seats decide everything, and why elections keep coming back

One country, one constituency: how an Israeli vote actually works

The first thing to unlearn about Israeli elections is the idea of a local member of parliament. Israel does not divide itself into geographic constituencies where voters pick a named candidate for their town or district. The whole country is a single electoral unit, and voters cast one ballot for a closed party list. The order of candidates on that list is decided by the party, not by voters, so a vote is primarily an endorsement of a party and its platform rather than a judgement on an individual representative's local record.

This design has a clear democratic logic. It ensures that a party which wins, say, 15% of the national vote receives roughly 15% of the seats, and it gives organised minorities — religious communities, ideological movements, and Arab citizens — a realistic route into parliament even if they are spread across the country rather than concentrated in one region. The trade-off is equally clear. Because representation follows votes rather than places, there is no constituency link to anchor a member to a local area, and the Knesset tends to fill with many parties rather than two or three large ones. That fragmentation is not an accident or a failure of the system; it is the system working as designed, and coalition-building is the price it charges for broad representation.

From votes to seats: the threshold, the quota and the Bader-Ofer method

Not every party that appears on the ballot wins seats. A list must clear an electoral threshold of 3.25% of valid votes, which in practice works out at roughly four seats. Votes for parties below that line are discarded for seat-allocation purposes, and that has real political consequences: supporters of a small party hovering near the threshold face a genuine dilemma, because backing their first choice risks wasting the vote entirely, while voting tactically for a larger allied party may help their bloc but weaken the small party itself. The threshold was not always this high. It stood at 1% for the first four decades of the state, rose to 1.5% in 1988 and 2% in 2003, and reached 3.25% in 2014, each rise nudging small parties towards mergers and joint lists.

Once the threshold has done its filtering, officials add up the valid votes for qualifying lists and divide that total by 120. The result is the general quota, effectively the price of one seat — about 36,227 votes in the 2022 election, for example. Each qualifying list receives as many whole seats as its vote total buys at that price, but because vote totals almost never divide neatly, a handful of seats remain. Those leftovers are distributed one by one under the Bader-Ofer method, named after former Knesset members Yohanan Bader and Avraham Ofer, which awards each remaining seat to the list with the highest votes-per-seat figure when its existing seats plus one are counted. The formula slightly favours larger lists, which is why two lists may sign a surplus-vote agreement before election day, pooling their leftover votes for this final round. Such an agreement does not merge the parties or bind them into a coalition, and it only helps if both partners cross the threshold independently — but in a parliament where one seat can decide a majority, it can matter.

Why 61 is the magic number, and why blocs matter more than parties

The Knesset has 120 seats, so 61 is the smallest number that is unambiguously more than half. A coalition generally aims for at least 61 because that gives it a working majority to pass legislation, approve a budget and survive confidence votes. The largest party does not automatically lead the government, however. What counts is which leader can assemble 61 seats of support, and Israeli history offers striking examples of the gap between size and power. In 2009, Tzipi Livni's Kadima won 28 seats to Likud's 27, yet Benjamin Netanyahu formed the government because he could build the broader coalition. In 2021, Naftali Bennett became prime minister even though his Yamina party had won only seven seats, because a rotation agreement with Yair Lapid and a wider coalition put him at the head of a 61-seat alliance.

That is why Israeli politics is usually described in blocs rather than single parties. Before and after an election, parties sort themselves — sometimes formally, more often by declared red lines — into camps that will or will not govern together. A bloc's total is what matters on election night, and a bloc can be arithmetically large yet politically unusable if its members refuse to sit with one another. Conversely, a government does not strictly need 61 affirmative votes at its investiture: it needs more votes in favour than against among those voting, so a minority government can take office if enough members abstain or stay away. Such governments are inherently fragile, because they depend on outside support that can be withdrawn vote by vote, but their existence shows that 61 is best understood as the threshold for a secure majority, not the only route into office.

How a government is actually formed: the president, the mandate and the clock

Government formation in Israel is a staged, deadline-driven process set out in the Basic Laws, and election night is only its starting point. Official results are published within days of the vote, after which the president consults the heads of every elected party. Each party recommends a candidate for prime minister, and the president entrusts the task of forming a government to the member of the Knesset judged most likely to succeed — guided, in the High Court's reading, by who has the best chance of assembling a coalition, not simply by who leads the largest party. The chosen candidate then has 28 days to negotiate, with a possible extension of up to 14 days.

If that candidate fails, the president may give a second candidate 28 days without an extension. If that too fails, a 21-day window opens in which 61 members of the Knesset may ask the president to entrust a specific candidate, who then gets 14 days. Only if all of those routes are exhausted is the Knesset dissolved and a new election held within 90 days. The clock matters because negotiations are not just about adding numbers: they involve ministerial portfolios, committee chairs, policy commitments, budget allocations and legislative priorities. A party may agree on the broad direction of a government yet hold out over a single ministry or a single law, and every day of delay increases the leverage of the last party needed to reach 61.

Why small parties punch above their weight

In a 61-seat coalition built from five, six or eight parties, a party with four or five seats is not a footnote; it is a veto. If its departure drops the coalition below 61, it can block legislation, demand concessions or threaten collapse, and its leaders know it. This is why coalition agreements in Israel are detailed, hard-fought documents rather than vague pledges of support. Partners bargain over who controls which ministry, which bills will be advanced, which will be shelved, and how much funding flows to their constituencies — whether that means yeshiva budgets and religious services for ultra-Orthodox parties such as Shas and United Torah Judaism, settlement and religious-nationalist policy for parties on the religious right, or secular and liberal priorities for centrist and left-wing partners.

The same leverage explains why a single defection can end a government. The eight-party coalition formed in 2021 held exactly 61 seats, the bare minimum, and it unravelled after a member of Bennett's own Yamina party quit the coalition in April 2022, leaving it without a majority, before wider disagreements finished it off. Governments built on a cushion of 64 or 65 seats can absorb a defection or two; governments built on 61 cannot absorb even one. The budget is a recurring flashpoint in this dynamic: the Netanyahu-Gantz unity government of 2020 collapsed in December of that year after deadlock over passing the national budget. When a small partner's core demand — on military conscription exemptions for ultra-Orthodox students, on religious observance in public institutions, or on spending for its community — collides with another partner's red line, there is often no compromise that keeps both inside the tent.

Arab parties: from the margins to the kingmaker question

Arab parties represent Palestinian citizens of Israel, who make up about a fifth of the population and are distinct from Palestinians in the West Bank and Gaza, who do not vote in Israeli national elections. Their political history runs through several traditions: early lists linked to the dominant labour party, the Arab-Jewish communist movement that became Hadash, the nationalist Balad founded in the mid-1990s, and the Islamist-rooted United Arab List, known as Ra'am. For much of Israel's history these parties sat in opposition, able to vote and speak in the Knesset but excluded from government, and their voters often debated whether participation in a state that defines itself as Jewish could deliver equality or would merely legitimise the status quo.

The electoral threshold forced a strategic answer. Before the 2015 election, Hadash, Balad, Ta'al and Ra'am united as the Joint List, winning 13 seats that year and a record 15 seats in March 2020, when the alliance became the third-largest faction in the Knesset. Seats alone, however, did not translate into power: when the Joint List recommended Benny Gantz for the premiership in 2020, the effort to build an alternative government around Arab support proved politically contentious and did not succeed, illustrating that being arithmetically necessary is not the same as being treated as a legitimate partner. The breakthrough came in 2021, when Mansour Abbas's Ra'am left the Joint List and joined the Bennett-Lapid coalition, becoming the first independent Arab party to sit in an Israeli governing coalition. Ra'am's pitch was pragmatic — funding and policy change on crime in Arab communities, housing and planning, and access to public services — rather than a settlement of the wider Israeli-Palestinian conflict. The experiment lasted only about eighteen months before the coalition collapsed, and it left open the question it had raised: how far can an Arab party cooperate with Jewish-led parties, and how far will those parties accept Arab partners, before either side's voters conclude the price is too high? That unresolved question is why Arab parties keep returning to the centre of coalition arithmetic whenever neither Jewish-led bloc reaches 61 on its own.

Why governments fall early, and what history teaches

On paper, a Knesset is elected for a four-year term. In practice, early elections are common, because the Knesset can dissolve itself and because a coalition that loses its majority has few ways to govern. The most dramatic illustration is the crisis of 2019 to 2022, when Israel held five elections in under four years — in April 2019, September 2019, March 2020, March 2021 and November 2022. The first two failed to produce any government at all. The third produced the Netanyahu-Gantz unity government, which collapsed over the budget. The fourth produced the broad Bennett-Lapid coalition, which collapsed after losing its one-seat majority. Only the fifth, in November 2022, produced a coalition with a more comfortable margin.

That sequence also shows how personal and ideological blockages interact with arithmetic. In the early phase of the crisis, the central division was framed in Israel as for or against Netanyahu himself: the liberal and centrist parties would not govern with him, Likud would not replace him, and his traditional allies would not govern without him. Neither camp could reach 61 without the other, and the Arab parties' seats sat in the middle of that deadlock. The lesson is general, not personal. Whenever a political system fragments into many parties, allows a small party to veto a majority, and lets governments fall when they lose that majority, elections become the default mechanism for resolving disputes that coalition talks cannot. Frequent voting is therefore less a sign that Israeli voters are unusually fickle than a sign that the system gives them repeated chances to redraw a parliament that no combination of parties has yet managed to hold together.

Caretaker governments: powerful on paper, restrained in practice

Between the fall or expiry of one government and the swearing-in of the next, Israel is never left without an executive. The outgoing government stays on as a caretaker government, and its ministers continue to run their departments, take decisions and represent the country abroad. Legally, a caretaker government has the same powers as a permanent one: it can make binding decisions, conduct diplomacy and, subject to the Basic Law requirement of government or Security Cabinet approval, authorise significant military action. The prime minister cannot simply decide on war alone, and caretaker status neither removes that collective requirement nor grants extra authority.

What changes is the standard of restraint. A series of High Court rulings has established that a caretaker government, lacking a fresh democratic mandate from the new Knesset, must exercise its discretion more narrowly than a regular government. It should keep the machinery of state running and respond to genuine urgencies, but avoid unnecessary, irreversible or narrowly partisan decisions that would bind its successor or use state power for political advantage — particularly on senior appointments and long-term policy. The Knesset itself is constrained during the transition as well: a new Knesset must be sworn in within 14 days of the election, but until its committees are reconstituted through the Arrangements Committee, legislation is difficult, and during an election recess a Consensus Committee of one coalition and one opposition member must agree before committees can even convene. A caretaker period can therefore last for weeks or, if coalition talks fail repeatedly, stretch towards three months — a long stretch in which the country is governed, but deliberately not redirected.

What to watch: the fault lines that will decide the next coalition

If the mechanics above are the grammar of Israeli politics, the coming coalition talks will be decided by a handful of recurring fault lines. The first is personal and bloc-based: which leaders and parties refuse to sit with whom, and whether a leader who cannot build a coalition steps aside for someone from the same party who can. The second is the ultra-Orthodox question — military conscription, religious exemptions and national service — which pits the demands of Shas and United Torah Judaism against secular parties for whom equal service is a matter of principle. The third is the Arab-participation question discussed above: whether Jewish-led parties will govern with Ra'am or the Joint List, accept their outside support through abstentions at an investiture vote, or rule both out and accept deadlock instead.

Beyond those, three structural signals deserve attention. Watch the threshold casualties: parties that fall just short of 3.25% take their votes out of the seat calculation entirely, and the loss of a few hundred thousand votes can shift the balance between blocs, as it did in 2022 when Meretz fell short by only about 4,000 votes. Watch the budget: a coalition that cannot pass a budget has lost the most basic test of governing capacity, and budget fights have ended governments before. And watch whether a security crisis pushes rivals towards a broad unity or emergency government, as has happened in past crises, or hardens the divisions that made a coalition impossible in the first place. None of these signals predicts a winner on election night, because in Israel election night never does. The winner is decided in the weeks of bargaining that follow, when arithmetic meets red lines — and when a single seat, a single defection or a single party's refusal to sit with another can send voters back to the polls.

Editorial disclosure

This article was researched with AI assistance and reviewed by the publisher before publication. Report factual errors through our contact page.