Muito bom

O que tenho lido sobre as eleições americanas: uma selecção.

1. O Pollster.com, para os últimos números das sondagens nacionais e estaduais e o que implicam para o colégio eleitoral, e os comentários de Mark Blumenthal (aqui sobre o "post-convention" bounce de Obama).

2. Também sobre as sondagens (e sobre previsões que se podem fazer na sua base), Nate Silver no FiveThirtyEight.

3. Votamatic, idem.

4. John Sides, em dois blogues: YouGov e Monkey Cage. Aqui sobre como a discussão sobre os rendimentos de Romney afectaram a sua campanha, e aqui sobre as "gaffes" e o seu (duvidoso) efeito nas sondagens.

5. Brendan Nyhan, aqui sobre a estranha persistência do mito de que Obama seria muçulmano.

6. Andrew Gelman (também no Monkey Cage), sobre quase tudo o que me interessa, incluindo e para além de eleições. Aqui resumindo por que razão Romney é um mau candidato (e dificilmente outro seria bom tendo em conta aquilo em que se tornou o Partido Republicano).

7. Pollyvote, tudo sobre diferentes formas de prever resultados eleitorais.

8. Jay Cost, ajudando a que os Republicanos não percam completamente a esperança.

9. Lynn Vavreck, que contribui para vários dos blogues acima, aqui sobre os eleitores indecisos.

Tudo disto é um bocado técnico, pouco especulativo e muito colado aos dados empíricos. Mas sucede que isso é precisamente o tipo de coisa que me interessa. Para "grandes reflexões", nem sempre impecáveis do ponto de vista empírico mas sempre impecavelmente bem escritas e inteligentes, leio o David Brooks, aqui sobre as mutações do conservadorismo americano.

More poll results and caveats

Starting with the caveats, note that in a "poll-poor" environment such as Portugal, it's understandable that a single poll gets a lot of attention. But let's not forget that a poll is just a poll. And although this last one somehow "feels right" (unlike this one), that does not make it "right" or "wrong," whatever that means when we talk about measuring public opinion. We need more data.

Furthermore, recall that the last Catholic University poll was conducted in late May/early June. This means that the massive PSD drop cannot be assigned exclusively to the measures related to social security contributions: the "Relvas affair(s)" and the privatization of public television were two very contentious issues that preceded the last austerity measures. Finally, note that the poll was conducted not only at the height (so far) of mobilization against the government but also following a week during which PSD figures, employers, unions, and even the minor coalition partners voiced their disagreement with the government's measures. So, although it is always possible that things can get worse for the PSD, it's probably safe to say that they had never as bad as in that period when the fieldwork took place.

Having said that, a few more interesting results:

* In the previous poll by the same institute, the PM was positively rated (getting 10 or more in a 0 to 20 scale) by 48% of respondents. This figure now dropped to 34%, making him the party leader with the worst approval rating.

* 78% of respondents disagree with the notion that the reduction of companies' social security contributions will help fighting unemployment, while 81% disagree with the notion this will have an impact on consumer prices. This is, in a way, a measure of how unsuccessful the communication of this measure was, as well as of the large consensus against it on the part of the kind of sources (media, expert, partisan) people likely used to form any sort of opinion about it.

* In contrast, there is an absolutely massive support for measures that were barely discussed, such as increasing taxes over luxury items or over high value real estate: support is always above 80%.

* 48% of respondents think now likely that Portugal will face a situation similar to Greece in the short-run, against 34% just three months ago.

*60% of respondents believe that no opposition party would do better than the current government if it was in power. Considering that 77% make a negative evaluation of the work of the current government, this helps explaining why, while the PSD drops in this poll, the PS fails to capitalize.

Católica, 15-17 Set, N=1,132, Face to face

PS: 31% (-2)
PSD: 24% (-12)
CDU: 13% (+4)
BE: 11% (+2)
CDS-PP: 7% (+1)
Others: 3%

Source. The comparison is with the last Católica poll, from May 2012.

Eurosondagem, 10-13 Sept., n=1037, Tel.

PS: 33.7% (+0.7)
PSD: 33.0% (-1.1)
CDS-PP: 10.3% (+0.2)
CDU: 9.3% (+0.5)
BE: 7.0% (+0.4)

Aximage, 3-6 Sept, n=600, Tel.

First public poll since August 17th, and first poll by this company since early July. Fieldwork predates the announcements by the Prime Minister and the Minister of Finance of new austerity measures, so you can probably see this as just giving a first inkling of the likely massacre to come.

PS: 35.4% (+4.6)
PSD: 33.3% (-1.7)
CDS-PP: 7.1% (-0.8)

I'll fill you in on the other parties when I get them.

Transatlantic Trends: alguns resultados

Os resultados do inquérito Transatlantic Trends, apoiado em Portugal pela FLAD, são divulgados em inglês e suficientemente conhecidos lá fora. Logo, passo para português neste post. Os inquéritos cobrem muitos temas, especialmente de política internacional, mas têm dado muita atenção ao tema da crise económica nos últimos anos. O trabalho de campo ocorreu em Junho passado. Destaco alguns resultados:

Em geral, com poucas excepções, há mais pessoas a avaliar positivamente a actuação da UE do que dos seus próprios governos nacionais. É natural: os governos são partidários, dividem as pessoas; a UE não é vista da mesma forma. Dito isto, é curioso que uma crise cuja responsabilidade tem sido atribuída à incapacidade de decisão na Europa permita, mesmo assim, que na maior parte dos casos a actuação da UE seja vista como positiva. Quem paga as favas são os governos: afinal, são eles que podemos castigar ou recompensar, ao contrário de uma distante e politicamente irresponsável UE. Notar também as excepções: Alemanha, Turquia e Suécia.

Uma mini-surpresa: na maior parte dos países, a actuação da chanceler Merkel é vista como positiva pela maior parte dos inquiridos. Aqui, claro, há uma clivagem:

Na Turquia, em Portugal, em Itália, e em Espanha, essa avaliação é predominantemente negativa. E está fortemente relacionada com uma pior opinião da Alemanha enquanto país. Isto é algo que, sendo porventura compreensível na Turquia, me parece um bocado perturbante nos restantes países, devo dizer.

Na maior parte destes países, a avaliação que se faz dos efeitos do Euro (ou dos que seriam os efeitos do Euro para os países que não o têm) é mais negativa que positiva. E onde ele não existe, pelos vistos, poucos o querem. Por outro lado, noutra pergunta, há já 50% de espanhóis que acham que seria melhor sair, valor que chega aos 40% em Portugal...

Que fazer? Uma das perguntas coloca os inquiridos perante três opções: diminuir despesas para diminuir a dívida; manter o nível de despesas; ou aumentar para promover crescimento? O gráfico seguinte mostra as percentagens para a primeira opção:

Apesar da diminuição em relação a 2011, Portugal continuava em Junho o campeão da defesa do corte das despesas do Estado. Claro, a pergunta é colocada em abstracto, sem especificar onde se corta, o que facilita o apoio. Para além disso, cortar despesas é bastante diferente de aumentar impostos ou contribuições para a segurança social, o que nos deve tornar ainda mais cuidadosos sobre interpretações destes resultados como sendo de "apoio" às medidas de quaisquer governos. Mas dito isto, há (ou talvez mais exactamente, havia em Junho) um capital político não irrelevante de apoio a uma redução das despesas do estado nalguns países, especialmente em Portugal. Já o que se fez com esse capital é outro assunto.

O que me leva ao último gráfico. "Algumas pessoas dizem que o nosso sistema económico funciona de forma justa para todos, enquanto que outros dizem que a maior parte dos benefícios do nosso sistema só vão para alguns. Qual destas opiniões se aproxima mais da sua?"

Tal como sucede em todos os estudos que alguma vez me lembro de ter visto sobre este tema, Portugal destaca-se por uma sensibilidade particularmente elevada às questões da justiça distributiva do sistema económico. Num país tão desigual como o nosso e, porventura, com uma cultura política marcada ainda pelo 25 de Abril, não creio que isto seja particularmente surpreeendente. O que surpreende é que seja tão frequentemente esquecido.

Leader approval

A few days ago I gave you the long-run view on voting intentions. Now for the long-run view on leader approval. Pollsters in Portugal ask this question in very different ways, rendering frequencies not comparable. There are ways to go around this, but for now I decided to focus on a single pollster that makes their data publicly accessible in a very nice way: Marktest. So this is is percentage of respondents who make a "positive" assessment of the performance of the leader of the PS, the PSD, and of the President of the Republic (there's just another response option, "negative") since 2005. The lines are lowess smoothers (80% bandwidth):

So, a few things:

1. The most remarkable phenomenon is the decline in the President's approval. One would need to display a much longer series to show this fully, but I think it's probably not too bold to say that Cavaco is now the most unpopular President Portugal has ever had since polls have measured this sort of thing regularly (since the mid-1980s). The fact that he is now as popular (or, better put, unpopular) as the two main party leaders is, I think, particularly striking, given the nature of his office.

2. In spite of the decline in the former Prime Minister Sócrates' popularity, PSD never had, until Passos Coelho, somebody who could match him. Furthermore, Passos Coelho started at a level well below the popularity of Sócrates at the beginning of his own term. Having said that, the political and economic circumstances are, of course, dramatically different from those in 2005/2006.

3. A good piece of news for Coelho is that his level of approval has stayed relatively stable until July 2012, unlike his party, which has suffered a decline in voting intentions. Of course, we'll have to see how the horrifying political summer the government has been experiencing will play out in the next polls.

4. Seguro, the new PS leader, has been rising, although having started from very low levels (for PS leaders' usual standards).

Martin John Callanan: Letters 2004-2006

An artist called Martin John Callanan sent letters to all heads of state of the world telling them: "I respect your authority." An example:

Quite astonishingly, or maybe not, he got a lot of replies. Here's the Portuguese presidency:

Here's Mubarak:

Even better was when Callanan send a letter to dozens of religious leaders asking them "When will it end?":

Here's the Archbishop of Canterbury:

Saw this today at Whitechapel Gallery and thought this might amuse political junkies. There's a book with a selection of the letters, which I duly purchased. Here's Callanan's website, where I'll certainly try to find out more about his work. Like this, for example.

Polls in Portugal

All the voting intention polls since the 2005 elections. The vertical lines are the election dates (2005, 2009, and 2011), the lines in color are lowess smoothers (25% bandwitdth). Data taken from here and here (and from my own data until 2009).

On the coalition parties, the PSD is at its lowest level since mid-2010, and the CDS-PP has also been also dropping since the 2011 election. But note how the CDS-PP rose as the 2009 and 2011 elections approached. Predictably, the CDU (the Communist+Green coalition) is the most stable party. Consider, however, that almost all polls since the 2011 election have come from only three polling institutes: the Catholic University has published only two polls since that election and Intercampus hasn't publicly polled since June 2011.

Photo by Greg Bull, AP.

Economy, Ideology, and the Elephant in the Room

A few years ago, in the aftermath of the Lehman Brothers' collapse and already after two consecutive quarters of negative economic growth in Europe, The Economist asked a few questions about the possible electoral consequences of the economic crisis:

"First, will any overall swing of the pendulum between left and right be clearly detectable? Will, for example, the fall-out from the current financial crisis benefit parties that favour more government involvement in the economy? Second, will voters tend to reject incumbents (perhaps because of the gathering economic troubles), or tend to stick with the leaders they know in uncertain times?"

As far as I can tell, there have been two types of answers to these questions. The first has been to suggest an anti-leftist "swing of the pendulum", often illustrated with the defeats of Social Democratic parties in countries such as Britain or Spain. This account has even been espoused by some social-democrats themselves (such as David Milliband, cited here), and even the reactions to the recent victories of the Socialists in France - "Is Europe's left poised for a comeback" - assume the existence of some sort of "anti-leftist wave" that only now may possibly be receding.

The second type of answer has been to suggest that the electoral performance of incumbents in Europe and other industrialized nations has been "business as usual": retrospective voting, with voters punishing or rewarding incumbents on the basis of economic performance. In a widely circulated paper, summarized here, Larry Bartels looks at incumbent parties' electoral performance in the OECD from 2007 to 2011, and makes two arguments. First, that "election outcomes over the course of this 'Great Recession’ have provided little evidence of meaningful judgments on ideology or policy" (Bartels 2012: 44). Second, that economic performance, namely economic growth, accounts well enough for electoral outcomes in this period: "in periods of economic crisis, as in more normal times, voters have a strong tendency to support any policies that seem to work, and to punish leaders regardless of their ideology when economic growth is slow” (Bartels 2012: 50).

In a research note entitled "Economy, Ideology, and the Elephant in the Room," I look at the electoral performance of incumbents in Europe (the EU27 plus Iceland and Croatia) in recent years, covering the period of the Great Recession (early 2008-mid 2009) and the three years since (mid-2009-mid 2012), with data on the 30 legislative elections that occurred in that period, i.e., from Malta in March 2008 until France in June 2012 (replication data here). My starting point was the notion that economic voting theories do not necessarily pose the irrelevance of "ideology". One such argument concerns "policy-oriented" voting, originally proposed by Kiewiet (1983): to the extent that leftist parties emphasize the creation of employment and that increases in unemployment are likely to make voters more concerned with the problem, a worsening of the unemployment situation may actually result in better performance for parties on the left, even if they are the incumbents. A symmetric argument applies for inflation and rightist incumbents (see also Carlsen 2000). Another argument is the "luxury parties" hypothesis: Durr (1993) hypothesized that, while increases in “collective wealth” made citizens more likely to accept policies involving taxation and redistribution, in “times of economic uncertainty or insecurity, citizens will be less supportive of the liberal domestic policy agenda” (p.159). He found that economic downturns in the United States did shift citizens’ policy preferences towards conservatism, while periods of economic expansion drove preferences to the left. Stevenson (2001), Markussen (2008), Kayser (2009), and De Neve (2010) reach similar conclusions: lower economic growth shifts policy preferences away from the left and voting intentions away from “luxury” parties, i.e., those who support generous social programs. What these approaches have in common is the fact that although they see economy decidedly mattering for electoral outcomes, they also suggest that voters make meaningful distinctions between parties in terms of their ideology.

So what did I find? The dependent variable is the change, in percentage points, in the share of the vote for the Prime Minister's party in each election in relation to the preceding one. I regress this on GDP growth in the four quarters before the election's, the change in the unemployment rate in the preceding year, and inflation in the preceding year. Two controls: Time in office (the number of years the same PM party had been continuously in power at time of election) and Previous election result (taking into account the fact that, keeping other things equal, parties that have obtained larger shares of the vote in previous elections are liable to lose more in subsequent elections). Finally, I regress incumbents' electoral performance on dummies capturing party ideology, measured either in terms of expert placements of parties on the left-right scale (Left, source: Parlgov) or in terms of the party families to which they belong (LeftFamily,1 for Social-Democrats and Communists).

The results are disappointing, to say the least. Most variables have the expected direction, including all the economic variables and previous vote share. However, all are far from conventional statistical significance. In what concerns the possibility of an “anti-leftist wave,” above and beyond what might be dictated by economic performance, the results are not supportive either: the coefficient for Left is positive while the coefficient for LeftFamily is negative, both also far from significant.

What if, however, we consider the possibility that different parties may have performed differently under the same economic conditions? Or to put it in another way, the possibility that the same economic developments affected parties differently? If we add interaction terms between incumbent party ideology and the economic indicators, what do the new results tell us?

Basically, they tell us that under conditions of low growth, leftist incumbents did significantly worse than rightist, and the the opposite happens in conditions of high growth (results are similar if LeftFamily is used). We can also take a look at the results from a different perspective: did economic developments affect all parties in the same way? The answer is no:

Leftist governments' electoral performance was sensitive to GDP growth, rightist governments' were not.

In sum, the answer to the question of whether there was a “anti-leftist wave” during the Great Depression can therefore be stated in a different way: instead of a simple “No,” what the results tell us is that, indeed, conditions such as the ones that prevailed in 2008-2009 in Europe – economic recession – seem to have resulted in particularly worse performances for leftist incumbents. However, those worse performances should probably not be mistaken by any fundamental change in ideological preferences in Europe irrespective of economic developments: in those countries that experienced economic recoveries in the last few years and where elections were preceded by solid growth, leftist parties ended up doing significantly better than rightist ones. Thus, the data supports the "luxury parties" conjecture, not a mere "retrospective voting" hypothesis. In both figures, the remaining results are also suggestive of "policy-oriented" voting, but the estimates are too imprecise.

Is this all? There is an "elephant in the room." Take a look at incumbent performance since early 2008:

What is striking about this is the fact that, even after the "end" of the Great Recession (mid-2009), incumbent performance continued, on average, to decline, and variations between two different "groups" of countries are clearly visible. I speculate that, in the face of a protracted and yet unresolved financial, currency, and political crisis in the Eurozone, voters may be turning their dissatisfaction with the European crisis to the most obvious and accessible target: domestic governments. The final model tests the hypothesis that part of the divergence in the electoral fate of incumbent parties we see in the figure above is attributable to the governments of Eurozone countries suffering increasing electoral punishments as the crisis remains unresolved, above and beyond the (conditional) effects of the economy. The figure below shows the effects of "time" (Years since January 2008) in the performance of incumbents, depending on whether elections took place in Eurozone countries or not.

By each year elapsed since the beginning of 2008, incumbents in the Eurozone countries have lost 3.3 additional percentage points in the comparison with previous election results (p-value=0.027). For the other countries, the point estimate of the marginal effect is -1.7 and is very far from statistical significance. In other words, there has been a decline in incumbent performance experienced since the beginning of the Great Recession, above and beyond what could be expected on the basis of other factors, and that decline seems to be most clearly a phenomenon that has taken place in the countries of the Eurozone.

That the national governments of the Eurozone seem to be bearing the electoral brunt of what can only be seen as a much broader and systemic failure may be seen as a fundamental problem of democratic accountabily, given the highly complex network of economic factors and political actors at play. However, we should also not forget that European voters and governments have been playing this flawed accountability game for a very long time now. In a sense, there is even a sort of poetic justice in these developments. National goverments have been notorious for engaging in extensive blame-shifting to the EU level when faced with negative domestic outcomes and in credit-taking for the sucesses and benefits of integration. Now that the failure is, perhaps for the first time, a truly “European” one, the chickens are coming home to roost.

"I regret nothing"

Gore Vidal, 1925-2012.

Out (very) soon

Click the image to enlarge. Click here for more info on the book and the IntUne series.

Já que falamos em produtividade científica...

Da análise da produtividade científica na Sociologia

"As the sociology’s founding institute in Portugal, ICS-UL retained its leadership position for some time; however, as far as sociology is concerned, it has been overtaken by CIES-IUL. While both institutes occupy a leading role in terms of SCG, CIES-IUL is presently ahead of ICS-UL because it has a much larger and more productive team."

Fernando Luís Machado, Director do... CIES-IUL, num artigo publicado numa revista científica.

Eurosondagem, 12 Jun, N=1022, Tel.

PSD: 34.3% (-0.5)
PS: 32.1% (+0.9)
CDS-PP: 11.6% (-0.9)
CDU: 9% (+0.2)
BE: 6.9% (+0.4)

Here, basically suggesting that the last known poll was not an outlier.

Subtleties: polls in Spain and Portugal since elections

Tidbits

Interesting tidbits from the last Catholic University poll (Portugal, registered voters, N=1366, face to face, 49% response rate):

- 67% evaluate the government's performance as "bad" or "very bad."And yet, 58% believe that no opposition party could do better.
- Passos Coelho and Paulo Portas, the leaders of the coalition parties, have the worst evaluations of all party leaders.
- Miguel Relvas is judged the worst member of cabinet, by far. Vítor Gaspar, the Finance Minister, the best (or least bad, so to speak.).
- Cavaco Silva, the President, has the worst evaluation ever, for himself and for any President before him.
- 72% support staying in the Eurozone. 20% support return to Escudo.
- 59% expect need for further EU/IMF support.
- 58% think it unlikely Portugal descends into a Greek scenario.

Portuguese polls

A new poll out today. Below, an update of our usual graph (the lines are local regression smoothers, 75% bandwidth) already including it. One result of that poll that may surprise some people is that the Left (the Communists + the Left Bloc) adds up to 18% in voting intentions. But as you see below, it's not even the first time that happened.