“Instability” in Israeli politics?

I was asked to offer some remarks on current “instability” in Israeli politics.* Instead, I am going to argue that Israeli politics suffers from too much stability—at least at the level of party and electoral politics. The fact that Israel underwent three elections between April, 2019, and March, 2020, might seem to imply political instability. So might the government recently formed, with its unwieldy power-sharing provisions and the parties’ need to reform constitutional provisions (Basic Laws) in order, at last, to prevent what could have been a fourth election within two years.

However, if we go a little deeper, there are two aspects of fundamental stability that have led the country’s politics to this current situation. First is the fact that a right-wing nationalist bloc of Benjamin Netanyahu’s Likud party and the two Haredi (ultra-orthodox) parties, usually also joined by an ultra-nationalist and religious-Zionist party (currently Yamina), has been unwilling to break up in order to facilitate government-formation. Likud and the Haredi factions function almost like one party nowadays, even though they run as three separate ones in elections.

The second factor is the persistence of anti-Zionist ideology among the Arab parties, who in their alliance known as the Joint List, have emerged as a (potentially) powerful force able to command about 12% of the Knesset seats. One can hardly ask parties representing the Arab sector to be Zionist, but if they would cease being opposed to the very nature of the political system (or be replaced by more pragmatic parties), they could participate in governments and bring benefits back to their voters who polls show care much more about better public services and integration into Israeli society than about ideological goals or the wider Israeli–Palestinian conflict.

Because of these two aspects of excessive political stability (or we might say stasis), it proved impossible after the two elections of 2019 and almost impossible after the 2020 election for a government spanning the political center to form. Yet consider that in three elections within eleven months, the right-Haredi bloc headed by Netanyahu failed to win a majority of seats. It did especially badly, relative to the 2015 election, in the second election of 2019, held in September. It recovered only slightly this past March. Yet at the same time, the Blue & White electoral alliance headed by Benny Gantz and its potential coalition partners also failed to win a majority for an alternative government. Blue & White could have formed a government only with the cooperation of the Joint List—or else it needed the Likud.

In most parliamentary democracies with complex multiparty systems, what happens when no bloc of ideologically similar parties wins a majority of seats is that a coalition of the center forms, leaving the extremist parties of both left and right out. But owing in part to Netanyahu’s preoccupation with his own legal problems, he was unwilling to break up his loyal bloc. Meanwhile, many of Gantz’s secular and left-leaning allies were unwilling to enter a government with the Haredi parties due to sharp differences over issues of religion and state.

Only with the unwillingness of the key parties to go to fourth elections, particularly in the context of the covid-19 crisis, did such a center-right coalition (which excludes the far right Yamina, as well as the most secular components that had been included within Blue & White) finally form. They could have arrived at a similar destination much sooner if the right-Haredi bloc had been more flexible, and alternative governance options would be more feasible if the Arab bloc could be part of a government or of its Knesset support base.

In other words, Israeli politics would benefit from just a little bit more instability.

Before concluding, I want to add two final points—one on the institutional context and one looking ahead:

(1) It is not the case, as some commentators claim, that Israeli politics would be more stable if it would change its highly proportional electoral system. The country has about the number and relative size of parties we expect from the electoral system it has (based on mathematical models I have contributed to developing and testing). But it does not follow that the country would be better served by changing the system to reduce the number of parties. The system suits the country’s social divisions well. A more disproportional system would create more, not less instability, by making it less transparent how popular support would be turned into baragaining power in the Knesset and government.

(2) Will the current government last? The constitutional reforms that were passed as a condition of its formation make it likely that it will, at least for a while, by instituting a rotation in the Prime Minister position and making it harder than usual for the leader of one bloc to depose the other and break up the government. However, by August, it must pass a budget. If Netanyahu wants to prevent a budget deal, he probably can, and this would force an early election. Polls consistently show that the right-Haredi-ultranationalist bloc would win a majority if an election were held now. However, it is not clear that he would want such an outcome even if he could engineer it. All of Netanyahu’s coalition governments but one have had at least one party of the center-left in them in order to balance out the ultranationalist parties (and the farther right of his own Likud). The exception was the one formed after the 2015 election, which was a narrow right-wing government. And it was precisely his small right-wing partners who maneuvered to bring that government down, precipitating the first of these three recent elections when no election actually would have been due until November, 2019. Netanyahu may be quite happy with the deal he has now while his legal process slowly plays out. In other words, Israeli politics may be a good deal more stable than it appears to many observers.


* At an on-line meeting of Davis Faculty, scheduled for 5 July. In addition to drawing on several earlier posts on Israel at this blog (some of which are linked in this post), I also draw on my teaching of Israeli Politics at UC Davis, as recently as this past spring quarter.

17 thoughts on ““Instability” in Israeli politics?

  1. Humor me, just hypothetically, if Israel had a system that used a large number of 4-seat districts with both 3-seat Largest Remainder with a Hare Quota and a single-seat Rank Choice Vote and one at-large seat, wouldn’t that tend to favor the biggest and the smallest parties or small parties that are more based on geography than ideology or religion?

    As I understand it, 3-seat LR Hare, if it were actually used anywhere, would be an exception to the Seat-Product Model and tend to increase the size of the biggest party and the effective number of parties relative to 3-seat PR with a less proportional quota… The problems with the Hare quota that traditionally have been rectified by using the Droop quota might instead be offset by using both PR and single-winner at the same time… And it would be easy to enforce anti-clone laws if there were two big parties, one of which has an incentive to make sure such laws are enforced…

  2. There are perhaps parallels between this and Italy’s First Republic, where superficial ‘instability’ (in that case a revolving door of Prime Ministers) masked the fact that fundamentally the country was governed by the same corrupt centrist coalition, which was essentially insulated from electoral competition, for that entire period. Amazingly enough, opponents of PR in the Anglosphere seem to focus on the apparent instability in both cases instead of the underlying instability.

    • Good point. It is almost as if “instability” is just an automatic term to apply to the politics of a country using PR, given that–as you say–the evidence of the alleged problem is quite different in these two cases.

      I really do not know for sure what the organizer had in mind in asking me to write about “instability”. Certainly not the frequency of PM turnover! As I say in the post, presumably it was a reference to the three elections, but as I try to develop there, even that does not work.

      I suppose someone could point to all the break-ups of alliances as the new government was formed as “instability”, but I doubt this is what prompted the proposed title of the talk that I am giving.

    • In a similar way in the French Third and Fourth Republics the same parties tended to be in government (with the obvious exception of the Popular Front period) and the same individuals tended to hold a particular portfolio for long periods despite the frequent changes of prime minister.

      • True. Although to be fair to winner-take-all politics advocates, that’s not much improvement. A multi-party coalition that’s in power for decades years and where the premiership rotates in mid-term among different politicians is actually the worst of both worlds from their point of view. The parliamentary majority keeps removing the premier, but the voters can’t (seem to) remove the parliamentary majority. Put another way, a leader might be removed from the big chair despite (in FPTP terms) “winning” an election (eg, Schmidt in 1982) or might be installed in the premiership despite not winning the mos seats (eg, Ardern).
        I don’t see “PMs and Cabinets get voted out” as an unqualified good in the eyes of the pluralitarians: they (to their credit) want it to occur only or mainly as a result of an immediately prior election loss, or an imminently impending election loss, rather than independently of popular opinion. MPs removing, say, a Nakasone in mid-term despite his popularity is a false positive; an Andreotti hanging on in Cabinet for almost all the entire period 1972-1992, with three stints as Prime Minister, despite the Christian Democrats’ popular support rising and falling,[] would be a false negative.
        [
        Although the DC always had a plurality of votes 1946-1994, so by strict FPTP logic the Italian system gave the “right” results in terms of the Premiership, ironically].
        I’ve heard it said that when PMs in multi-party systems get replaced in mid-term, it’s actually because MPs and party leaders are getting feedback from the grass-roots about their unpopularity; that it’s really a form of informal recall; that it’s (in Australian terms) more like the intra-party ouster of Bjelke-Petersen in 1987 or of Abbott in 2015 than of Rudd in 2010. This may well be true, but the problem is, it’s hard to evidence externally. Losing a lot of of seats in an election is an obvious, dramatic rebuke (even if it’s misleading because your vote share actually went up, like the UK LibDems in 2019). “We heard from a lot of voters” is hearsay. (Not unlike closed lists – “parties usually promote popular candidates who campaigned hard to top slots on the list at the next election” may well be true, but there’s no hard evidence in the form of personal preference votes for candidates to confirm this).

      • I once spoke to an engaging but not-very-political European student who was visiting Australia during one of our elections.
        “You have a good system here. Your first-past-the-post [sic], means you can vote out a government you don’t like. And your MPs are all the most popular individuals”, she said. “At home, we use the party lists. The parties decide everything, the voters have no say. Small parties dictate to the majority. We should copy your system!”
        She probably expected me, as an Australian, to be flattered by this. Instead I asked, “So [your prime minister], his party only has 35% of the votes. if you don’t want him beholden to minor parties for support, why don’t you all vote for him and give him 55% of the votes? Then he’d have an absolute majority, even with party lists. He could tell the minor parties to go away.”
        No, she said, that would never do: “We don’t trust him to have all the power on his own.”
        “Well, what about [leader of the largest opposition party]? Bump them up instead, from 30% to 51%?”
        No, she said, her people didn’t completely trust him either.
        I’m not sure how widespread her attitude is but it does start to explain why occasionally Israelis hold rallies to demand more majoritarian electoral reforms while eschewing a simpler, already existing remedy – viz, throw popular support behind one of the two largest parties even without Maurice Duverger’s cattle-prod forcing you to do it.

      • I think having a president shd ideally abet a transparent regular circulation of the elites which can be good, but the trick is to mitigate the power of the way that favors the emergence of two (or one) big party.

  3. An additional problem in Israel could not be the existence of a party simultaneously “hawk” and secularist (Ysrael Beitenu), who, because of that, also hard to put in a coalition?

    Sometime ago I read and article about the oppostion between the Israeli Jews from Europe and the Israeli Jews from the Midde East, and one of the points was that new immigrants from ex-USSR could be culturally “european” but are politically “middle eastern” [https://www.timesofisrael.com/nobody-hijacked-israel-its-just-not-what-its-pioneers-thought-theyd-created/].

    But perhaps the immigrants from ex-USSR are not so in the “middle eastern block” like it was supposed, and many are a kind of third (or fourth, counting with the Arabs) force, not easily aligned?

    • Yisrael Beiteinu used to be easy to put in a bloc–it was firmly on the right. However, it is one of the right-wing parties I mention in the post that led to the premature breakup of the government in late 2018. Only after the April, 2019, election did it become clear that YB was no longer a member of that bloc, by emphasizing the secular side of its agenda.

      My read is it would have been willing to join a coalition as long as (1) the Haredi parties made some concessions, or (2) an alternative government was not reliant on the Joint List for its majority. But neither outcome was in the cards.

      The main issue with XSSR Jews is that a significant minority of them do not count as Jewish in the eyes of the official rabbinate, even though they obviously qualified under the Law of Return. Those in this grey area are unable to marry in Israel or obtain various other services for which the rabbinate has a monopoly. However, they tend to be quite hardline on security, in part because many of them live near the Gaza Strip and are disproportionately affected by the regular terrorist rocket fire. The leader, Avigdor Liberman, lives over the Green Line (in Nokdim), although I don’t think very many of his voters do and the party has tended to be more favorable towards a “two-state solution” than the other parties conventionally classified as right-wing.

  4. What do the Joint List parties advocate? Post Zionism? Bi-nationalism? I see that the Swedish minority of Finland are part of every government with rare exceptions. The Turkish minority of Bulgaria has participate and tolerate some governments. The only parallel that I can think of is the Russian minority of Latvia. Are there any countries where governing parties where a small minority has never participate, support, or tolerate a government? Some European countries are keeping some populist parties out, but this may not be the same analogy, and the following questions may be the same supposition as well. Would the Israeli Arab parties be seen as a testimonial party like the Dutch SGP (Dutch Reformed Party) or some faction of the Joint List could be?

    • Rob, there are four parties in the Joint List, so they may not all advocate the same thing. But they are definitely anti-Zionist, which is to say they oppose the Jewish character of the state. (Change the anthem and flag and presumably the country’s name; repeal the Law of Return for Jewish immigration, etc.) This is the sense in which they should be considered “anti-system” parties, and hence unlikely to be brought into government or into formal support-party status. Note that “unlikely” does not mean “impossible”; but a lot of reassurances would be required on the part of the Zionist parties in any such government, and rightly so.

      The Turkish minority party in Bulgaria is not opposed to the state of Bulgaria. In fact, it skirts the ban on ethnic parties by styling itself a party for “rights and freedoms”. If the Arab parties in Israel would take a similar path, I doubt there would be much of an obstacle to deeper cooperation with them.

  5. New talk of Netanyahu seeking an early election, and as foreshadowed in the original planting, it would be provoked by refusing to allow a budget to pass.

    The budget needs to be passed in August, contrary to what I originally wrote (now corrected). The coalition deal calls for a two-year budget, but Netanyahu is trying to insist that due to the covid crisis, it should be one year. (Never mind that the alleged basis of the “unity” government was to deal with the covid, crisis, and thus presumably so was the agreed idea of a two-year budget.)

    If the Knesset does not pass a budget (whether one year or more), then it is automatically dissolved, and a new election could come as soon as this November. In other words, referring to the election earlier this year as “2020a” still may turn out to have been appropriate.

    • But I should add that early elections are threatened in Israel far more often than they actually happen. The polls have very much ceased being good for the Bibi bloc. So this could be bluster (although there is, again, no obviously viable alternative majority). Anyway, worth keeping an eye on a possible “2020b”.

  6. “A more disproportional system would create more, not less instability, by making it less transparent how popular support would be turned into baragaining power in the Knesset and government.” Why would such lesser transparency increase instability?

    • JD, as a PR supporter I understand the “dis-PR” argument to be that political strength is a function not only of electoral size but also of median position. As (I think?) Michael Laver pointed out, a 49-2-49 result is more favourable to a minor party than a 51-25-24 result. The UK LibDems polled 25% in 1983 but had no presence in Cabinet, whereas in 2010 they polled 22% but got Deputy PrM and 5 out of 21 Cabinet seats.
      Inasmuch as some element of disproportionality is seen to reduce the chance that a very small party will end up in the lucky median position, as “kingmaker”, those who advocate such measures would argue that they bring political power closer to, not further from, alignment with popular voting support.
      Of course disproportionality – or departures from pure proportionality – is a wide spectrum. Handing all 55 of California’s Presidential Electors to the ticket with a plurality is one extreme, but it can include even Israel setting a threshold of 1% (an increase over “a quota of 0.833% or one of the largest remainders”) with nationwide PR. All countries except for – I believe – South Africa and Cambodia have some disPR deliberately built in to a greater or lesser degree. STV-PR is closer to the Israel end, FPTP in the UK is closer to the California Electors end.

      • If California was to use 55 districts as a List System to elect the president, the threshold 1/(55+1)*100=1.78…..

        So it is a 1.78% to win 1 seat.

        A country that uses a list proportional representation system regardless of the openess of the lists, the maximum district magnitude ideally could be no more than 15 seats; the threshold is 1/(15+1)*100=6.25% assuming that their is no regional and/or national thresholds.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.