Spain is not Denmark (in yet another sense)

29 Dic, 2015 - - @politikon_es

Spain and the strategic dilemma of coalition government

Since its transition to democracy in 1978, Spain has not had a single coalition government. This, together with the distance between the policy positions of the four big parties now in parliament has led many observers to argue that Spanish political culture is ill-equipped for pacts, that no coalition agreement is feasible, and that the result of the last election is likely to lead to a chaotic policy landscape.

Reacting to the lack of comparative perspective in which this suggestion rests, many political scientists commenting in social networks have dismissed this assessment as not supported by the evidence. Multiparty governments, to start with, are frequent in parliamentary democracies, such as in the case of Denmark or the Netherlands; they are even frequent in Spain, at the subnational level and, in none of these cases, has the result been chaos.

In this piece I will take the opposite stance. As I said in my previous piece, I believe that we are likely to witness a short political term in Spain, that the pundit’s view is after something here, and that more consideration is due to it.

The framework to keep in mind is what I will call the strategic dilemma of coalition government: coalition partners are jointly accountable, but compete against each other in elections. This affects agreement and the sustainability of coalitions. When members of the coalition are far apart ideologically, finding a common ground for an agreement is difficult- for obvious reasons. On the other hand, when they are close to each other, they usually compete for the same voters and this will prevent cooperation within the government- feeding opportunism in the form of blame shifting, position taking, or just sabotage etc. This dilemma is one of the main forces preventing agreements and cooperation within agreements and leading to coalition termination.

Bottom line: the room for agreement is small

The bottom line of the analysis is that policy positions are too far apart to leave room for an agreement. There is no genetic cultural reason that makes inevitably coalition agreements infeasible in Spain, but there are reasons to believe this is so in the current scenario.

As I explained in the previous piece, any agreement requires a bargain between the socialdemocrats and either Podemos (the anti-austerity left wing newly emerged party), or the conservatives. I refer the reader to that piece for the key points of the analysis. In a nutshell, the strategic context is such that there are four national parties more or less evenly distributed in their vote share in a system in which the natural number is most likely to be around two. As a result, in the absence of constitutional reform, all parties are competing for their survival and any deal is largely zero sum game.

Take stock of the fact that the strategy of Podemos (in the few speeches and declarations) since then has been to propose conditions for an agreement that the socialdemocrats can not afford. An example is the emphasis they put on the red line of holding a referendum in Catalonia -a line the socialdemocrats have long avoided to cross in the past because of the perceived electoral consequences in the rest of the country. In turn, this responds to the importance of the support that the left wing party got in Catalonia (with an independent federation that may dispute its leadership).

If Podemos’ goal is to attract the socialdemocrats’ voters (and that is the only direction in which they can expand), its winning strategy is to impose extreme (like the referendum) or unrealistic conditions (“putting social rights inside the constitution”). If these are accepted, they will be able to take credit for it; if they aren’t, Podemos will blame them for not being progressive enough. This “position taking” and “blame shifting” game is the engine of public disagreement. A final point has to do with the fact that if elections are repeated, Podemos is likely to absorb the votes of the former Communist party -half a million that due to the malapportionment of the electoral system translated only into two seats, but if transferred to Podemos would put the socialdemocrats and Podemos in a virtual tie.

Spain lacks policing institutions

If an agreement was reached, what would ensure that it is actually enforced? The heart of the problem is that coalition agreements can only be drafted in the form of very broad guidelines. Contingencies -sudden slowdown in the economy; actual cost and details of a reform; revenue raised by a tax increase, etc- are hard to anticipate and verify.

Whenever an unenforceable contingency takes place, coalition agreements have to be renegotiated on the spot, and in renegotiation the strategic dilemma of coalition government shows its ugly face again: since coalition partners compete against each other in elections they are likely to act opportunistically in order to reap the position-taking benefits, and, on top of that, disagreement is fueled by asymmetric information.

Drafting legislation and budgeting are both costly processes that require expertise, and any agreement within a cabinet is problematic because the party in charge of the drafting the bill (the minister in charge) is likely to use its privileged position to condition the process -for example, pretending that the bill drafted is the only one that is feasible, which may, or may not, be true. Note that the problem has two faces. Internally, coalition partners may or may not know if the ministry is being a faithful trustee drafting the piece of legislation. Externally, even if they have all the relevant information to verify the deal, it may be electorally advantageous to act opportunistically, and walk from the agreement blaming the other side if the electorate does not.

Martin and Vanberg (CUP 2011) formalize and test this idea in their book on multiparty governments. Their thesis is that parliamentary institutions have a prominent role as “policing institutions”. In multiparty systems, legislative institutions do much more than rubber-stamping bills, they scrutinize them -and have the power and resources to do so.

A different but related idea is put ahead by Hallerberg and his coauthors in their work on budgets 2004 and 2010. They take stock of how political institutions tackle the problem of the “common pool” in fiscal performance (that fiscal performance is common, but political credit from expenditures is individual). They suggest that fragmented political systems can address these problems with certain institutions that make fiscal goals an enforceable part of the coalition agreement. Their empirical work takes stocks of the institutions that make this possible. Such systems are typically equipped with independent institutions that work as watchdogs of the goals, the coalition agreement often has force of law; budget committees are equipped with personal resources and substantial autonomy, etc.

Spain, however, perhaps lacks the institutions that may help to overcome the strategic dilemma of coalition government. Nor are voters expectations shaped to accept substantial concessions- at least Podemos’ voters. Surely, there is no reason to believe that these will not emerge if given sufficient time. Perhaps in the new scenario, parties will find to their advantage appointing MP’s with skills and actual government experience to scrutinize bills; perhaps coalition agreements will acquire more institutional form; perhaps new independent institutions will be created to ensure independent assessments. But any of these developments will take time to build and a learning period -in particular, public opinion will have to get used to it.

Subnational scenarios may not be comparable

While it is true that minority and coalition agreements have been common at the subnational level in Spain, I claim these are not comparable. One should take into account the context I described in the first section and the previous piece -the room for agreement particularly small.

Moreover, the stakes at the national and subnational level are very different. We may distinguish two cases here. Consider, first, regions, such as Catalonia, where nationalist parties concur to elections (these are those in which coalition governments are most frequent). The catch with nationalist parties is that the degree of competitiveness between coalition partners is limited by the salience national/territorial axis. If voters are distributed across the four quadrants, there is room for agreement along one axis while leaving the other suspended. This has been the case in Catalonia ((I and II )), where parties have aligned themselves along the national of left/right axis several times to form mutually beneficial coalitions.

Consider the case of regions without nationalist parties. The Spanish party system is highly nationalized Simón 2013 (national and subnational party systems work similarly). And, I would argue (although I may be wrong since the identification problem is hard to tackle), the causality goes from support at the national, to the subnational level. Regional elections are not politically salient and are perceived as a way to monitor national parties as a whole. Since the degree of accountability of parties for their management at the local level is reduced, competitiveness is reduced and leaves more room for office holding concerns to be taken into account when reaching an agreement. Also, to the extent to which the real game is being played at the national level, it is rarely the case that two parties in direct competition in national elections (the conservatives and the socialdemocrats) have reached an agreement.

In both cases -in regions with and without regionalist parties- the key point is that there are reasons to expect that the strategic dilemma of coalition government will be attenuated. In non-nationalist regions, the game is played elsewhere; in regions where the nationalist dimension is prominent, the dimensionality of the strategic space reduces competition between coalition partners while making them closer in terms of policy.

As the piece linked above suggests:

Let’s try to avoid looking at regional elections from the prism or the experience of national elections. Spain is not Madrid.

Indeed, and the other way around.