The big problem with the “rotation” deal

The Israeli press, especially Haaretz, has carried several critical opinion pieces since the announcement of the joint list and rotation agreement between Labor, led by Isaac Herzog, and HaTnua, led by Tzipi Livni. (On the agreement, see previous post.) One problem with the deal that has been less directly addressed–with one partial exception that I will cite below–is its impracticality. Or, more to the point, its non-enforceability.

When a Prime Minister resigns, as Herzog promises to do after two years, there must be a new government-formation process. Therefore, the deal between two leaders of one list can’t bind the other parties. (It can’t actually bind Herzog, as the first PM under the deal, but that is another matter.)

Let’s suppose that Labor(-plus) does win the most seats in the election, as several polls now suggest is possible if the election were held now. (It is not till March.) Let’s also suppose that Herzog and Livni operate under the best of faith and fully intend to honor the deal by which Herzog steps aside and Livni replaces him. So far, so good. But then comes the problem. The deal would also require the coalition partners who form the Herzog-led government to agree to it. It further requires the coalition partners to uphold the deal after two years, when the “new government” has to be formed (and there would be a new formation process under the terms of the Basic Law, Government, Art. 19). That is a lot of to expect!

So, while the deal may be good in the polls, it has a flaw in its non-enforceability. Will enough Israeli voters buy it to give it a chance? At the moment, the same polls that put Labor ahead also suggest a right-wing government headed by Likud and with Benjamin Netanyahu, is more viable. Being first in seats does not guarantee a party/alliance the right to form a government. Just ask Livni. She has been there before.

The Herzog-Livni deal draws on the precedent of prior arrangements in which the Prime Minister was changed at midterm. However, these really are not very relevant as precedents. In the two previous rotation agreements, it was two separate parties making the deal, and together controlling a majority of the Knesset. While there still would have been the possibility that the party initially holding the top post might renege, the (potential) problem was only one of commitment between the two leaders and their parties. (For overviews of Israel’s “unity” governments, see the Washington Institute’s useful history; the cases involving rotation agreements were 1984 and 1988.) By contrast, the Herzog-Livni agreement is a pre-electoral commitment regarding their own list, but would depend on consent of various other, post-election, coalition partners. More specifically, it would depend on two rounds of bargaining with those partners.

One article in Haaretz went so far as to quote a “source” inside Labor as admitting the rotation deal was not necessarily viable:

Don’t get too upset, sourpuss analysts. The arrangement for a rotation between the two is good for now, for the start of the election campaign…

It has been well received by young voters and women; it conveys confidence and power. When the time comes and Herzog is asked to form a government, he and Livni will sit down and negotiate with the others all options on the table. If Kahlon demands a rotation, that too will be considered. We’re not naïve, and that won’t be a hindrance to forming a government.

Moshe Kahlon is leader of a just-registered new centrist party, Kulanu.

Yes, there will be a lot to consider. In the meantime, maybe it will help get votes, under the assumption that voters do not work through the next steps and all the pitfalls they put in the way of a rotation promise.

7 thoughts on “The big problem with the “rotation” deal

  1. A key point that was made by Samuels and Shugart (2010) was that under parliamentarism, the party providing the prime minister is generally able to replace that prime minister (usually through a leadership election, as PMs tend to be party leaders), even in coalition governments. Israel has already made at least one exception to that, when Olmert resigned in 2008 but Livni consequently was not able to renew the coalition, leading to new elections soon afterwards. Is this just down to Israel’s relatively strict rules dissolving the government when a PM leaves office and mandating a new government formation (some countries have no such rules, enabling the head of state to simply appoint a new PM without need for a new formation) or is something else involved in addition to that?


    • Not sure, but I think most parliamentary systems require the government-formation process to start again if the PM resigns.

      I don’t think S&S meant the point to apply to parties in coalition changing their leader. If we say it does, we might be wrong! I think we meant, or at least assumed, that the point in question applied to single party majorities, at least if the point is taken to be about the continuation of the government when the party leader/PM changes.


  2. In the 2010-13 Australian hung parliament, it was widely believed (and, I think, may even have been explicitly stated by the key players themselves) that the Independents’ agreement to support a minority Labor government included Julia Gillard as a persona designata, and would not automatically continue if the Labor caucus were to unilaterally replace her as leader, as it had done with Rudd in 2010.
    Ironically this may have fortified Gillard’s position against a splll from within her own party (compared to, say, John Gorton and Bob Hawke, whose caucuses had House of Representatives majorities), although in the end it only delayed the inevitable.
    A lot of constitutions these days seem to require an absolute majority to remove a chief minister in mid-term, especially if the no-confidence vote has to be “constructive” and simultaneously name a successor. It would be ironic if a PrM leading a minority government had more secure tenure against mid-term removal.


  3. Well, the deal did not last very long. Livni “renounced” it the day before the election!

    So my question is, if it was just a cynical deal all along (as my quote above from a “Labor source” implies), why admit it on the day before the election?

    Even more cynically, MK Shelly Yachimovich (former Labor Party leader) is quoted as saying, “The rotation agreement was not very well liked, but was originally considered as one of the concessions needed to be made to gain more power.”

    And now polls say it isn’t helping “gain more power”, so drop the agreement and see if you can… gain more power. This isn’t exactly looking good!


  4. Well, Livni didn’t actually renounce the rotation deal, but rather vaguely stated she would not insist on it it posed problems with government formation. Whatever that means is anybody’s guess. The press rather typicly – and probably encouraged by ZU strategists – depicted this as waiving rotation rights. The funny thing is that way back in December it was thought that this is a genious move that placed the emerging ZU as a viable contender. Now it seemed a liability on the campaign. The campaign in Israel is way too long and it’s politicians way too cinical. Not one of the left’s greatest moments to be sure.


    • Yes, the headline is overly dramatic, I agree. However, this may be one of those cases in which the spin the story receives is more important than the story itself.

      That “rotation” would not be allowed to prevent a coalition deal with other parties that objected–or wanted a piece of the rotation pie–actually only reiterates the statement of the “Labor source” back in December (quoted in the main post).


Leave a Reply

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

You are commenting using your 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.