On May 6, Greece went to the polls for parliamentary elections which yielded a result so unwieldy that the leaders of the main political parties soon discovered that no government with a functional majority could be formed. So new legislative elections are taking place tomorrow, June 17, and Europe's political leaders and businessmen will again be watching with apprehension.
Not long ago, they had managed to persuade Greece's two main traditional parties - the right-wing New Democracy and the left-wing PASOK, sworn enemies of sorts - to approve "the memorandum," an EU financial rescue (bailout) deal which commits Greece to a package of painful budget reforms. Those reforms, however, were blamed by many Greeks for escalating poverty and unemployment, rising prices, and brutal cuts in public services and benefits. As a result, both New Democracy and PASOK experienced some brutal cuts themselves when elections were held in May.
Massacre in May: the elections which took an axe to Greece's political system
In last month's elections, New Democracy lost almost half its support, dropping from 33% of the vote to 19% - the worst result since the party was founded after the fall of the military junta in 1974. PASOK managed to do worse still. It lost over two-thirds of its support, dropping from 44% to 13% of the vote. For PASOK too, this was the worst result the party had gotten since it was founded in 1974. Suddenly, the two parties which had dominated and alternately governed the country for almost four decades, together pooled just a third of the vote.
The coalition government they had formed in a rare collaborative effort last November, which also included the small rightwing party LAOS, had enjoyed a majority of 266 seats in the 300-member parliament. That was gone. Even the quirky element of the Greek electoral system which awards the largest party 50 extra seats - and despite its paltry 19% of the vote, New Democracy was still the largest party in the new, highly fragmented political landscape - couldn't save the "unity government". LAOS didn't make the 3% electoral threshold, and ND and PASOK together got 149 out of 300 seats in the new parliament. Close but no cigar.
Most of PASOK's urban and working class electorate and most of its voters in its traditional stronghold of Crete abandoned the party for the Coalition of the Radical Left, Syriza. That party and its main predecessor, Synaspismos, had so far occupied a rather marginal perch in the Greek political landscape, to the left of PASOK and to the right of the orthodox, unreformed communist party KKE. Between 1991 and 2009, Synaspismos and Syriza had received between 3% and 5% of the vote in national elections. Now, suddenly, they received 17%, thanks in part to the charisma of its young leader Alexis Tsipras. Suddenly they are the country's second-largest party, larger than PASOK.
Another beneficiary of PASOK's swoon was DIMAR, the Democratic Left, a new party which brought together some former PASOK and Syriza politicians, and received 6% of the vote. The communist hardliners of the KKE, however, did not benefit from the political and economic crisis. While they held on to their true believer constituency, getting 8% of the vote, they did not gain any ground.
In its turn, New Democracy predominantly bled votes to a new formation called Independent Greeks (ANEL), which was founded just three months before the elections by renegade, anti-bailout deal MPs of New Democracy. ANEL won 11% of the vote. The other winner of the elections on the right was the fascist Golden Dawn (XA), an extremist party which received 0.3% in the last elections but now surged into parliament with 7% of the vote. Both the Independent Greeks and Golden Dawn are rabidly opposed to the bailout/austerity deal, as are Syriza and KKE.
Faced with the need to find at least two more votes in parliament to form a government, New Democracy and PASOK therefore turned to DIMAR, which takes a more middle-ground position on the bailout. But the left-wing party was hesitant about being offered a poisoned chalice, and insisted it could only join a government if Syriza did too - which Syriza refused. An alternative anti-bailout coalition government led by Syriza, meanwhile, was never really in question, since the KKE prefers revolution over government, the Independent Greeks are loath to take part in what would be an overwhelmingly left-wing government, and the fascists are considered beyond the pale by all. So just nine days after the elections, coalition negotiations were declared fruitless and a new election date was set: tomorrow.
(A far, far more in-depth analysis of the May election results can be found on the ever reliable World Elections website.)
There can be only one: the race for the 50 bonus seats
With the pro- and anti-bailout camps so polarized, and so evenly balanced in support, the quirk in Greece's system that awards 50 bonus seats to the largest party takes on a crucial role in the outcomes of next week's elections. Voters have realized this, and consequently, have been rallying to the two winners of last month's elections: New Democracy and Syriza. That is, at least, what the polls were saying, between early May and two weeks ago. Then, two weeks ago, the curtains were drawn - the publication of opinion polls is forbidden in the last two weeks before the elections.
The many polls that appeared until then, however, showed a number of clear patterns.
The voters of the various parties that didn't pass the 3% electoral threshold in May were the first to shift. LAOS and the Ecologist Greens had both received 2.9% of the vote - just a couple of thousands of votes shy of the threshold. They won't come anywhere near that tomorrow. No poll since May 17 has given either party over 2%.
Another 6.5% of Greek voters cast their vote for three liberal parties which each received 1-3% of the vote. One of them (DISY) merged with New Democracy. The two others responded to the result by entering an alliance (DX/DRASI-FS). It may not help: most of the polls in May still showed them failing to reach the threshold, though some saw them squeaking in.
Over 5% of the voters in May, meanwhile, voted for a party that ended up getting less than 1% of the vote. Most of those quickly thought again, the post-election polls showed, with the number of people voting for "other parties" dwindling.
The medium-sized parties, DIMAR, KKE, ANEL and XA, seemed to suffer too - but definitely not all in equal measure. Here's an overall chart (click chart for full size):
The dots on the left present the results of the last elections. In the subsequent month, support for both New Democracy and Syriza bounced upward. Support for PASOK initially grew a bit too, then seemed to drop again. The other parties, at first glance, all suffered.
Looking at the polls in more detail (full table far below), however, reveals some nuances. First off, because of the 50-seat bonus for the largest party, the biggest question of all is: which party will become the biggest?
SYRIZA
(click chart for full size)
Syriza was the first to see a big jump in its popularity - within days of the party receiving close to 17% in the elections, it was polling at 23-28%. But over the next three weeks, the party didn't appear to gain a whole lot more additional ground. The last polls which all the pollsters got to release, on May 29-31, showed the party at anywhere between 24% and 32% of the vote.
(click chart for full size)
When it came to Syriza's numbers, ever since the first polls after the May 6 elections were published the differences between pollsters were bigger than those between the successive polls by each pollster. Taking the eight pollsters who conducted three or more polls in this period, only Kapa saw Syriza grow by another 5% or so after the first post-elections bounce. Rass and Public Issue saw the party add another 3-4% or so, while Alco, MRB, Metron and Pulse RC saw the party add only a percentage point or two after the initial bounce.
In short: did Syriza quickly rally most of the floating left-wing voters once the May 6 elections showed it to be the new top dog on the left, but hit something of a ceiling of about 26-28% after that? The only two pollsters who did not show such a ceiling were VPRC and Public Issue, which gave Syriza 30% and 31.5% respectively. Both those pollsters adjust their results by weighing in the likely votes of undecided voters.
NEW DEMOCRACY (ND)
(click chart for full size)
The pollsters were generally more in agreement about New Democracy's support than about Syriza's. The party did not enjoy quite the same, immediate post-elections rally in the polls after May 6. The first few polls, on May 9-11, gave the party some 22-25% of the vote, compared to the 21.4% it received in the elections. But a week later it was 24-28%, and another week later 26-29%, positioning it right at the same level of support as Syriza's. Such a more steady, consistent growth suggests that ND didn't quite face the kind of natural ceiling Syriza might be dealing with, and may have kept persuading additional voters in the polling-free weeks of June as well in a way that Syriza couldn't. This would point to a ND victory, possibly a decisive one.
(click chart for full size)
It does bear pointing out, however, that ND's growth did seem to level out in the very last week of May as well, as the last set of polls gave the party 26-30% of the vote. Looking at the polling by individual pollster adds an element of uncertainty: while Kapa, Rass and arguably Pulse RC showed a steady and healthy growth of ND support throughout May, Metron, Marc/Alpha and Public Issue saw it levelling off in the later weeks and Alco and MRB never saw it move much at all anymore after the initial post-elections bounce.
PASOK and the others: In the shadow of ND vs Syriza race, who wilted and who won?
When a long-dominant party is suddenly overtaken by a minor rival and relegated to third place in the election results, you might expect an exodus of sorts among voters, now the party is no longer the default choice on its side of the political spectrum. Yet this is not what happened to PASOK - at least not straight away.
(click chart for full size)
Possibly benefiting from an immediate collapse of support for parties that didn't make the electoral threshold, PASOK initially actually clawed back a modicum of support. By mid-May, Alco, Public Issue and Pulse RC published their first post-election polls, and Marc/Alpha a second post-elections poll, which showed PASOK back up by some 2 points at over 15%. Metron even had the party at over 17%. Follow-up polls by Kapa, Public Issue and Rass the week after saw some additional voters flocking (back?) to PASOK as well.
(click chart for full size)
The party was sinking again by the last week of May though, when the final polls were released before the curtains were drawn on public polling. Seven out of ten pollsters agreed: PASOK was once more losing ground as the race for the top spot intensified between Syriza and ND. Considering the enormous losses the party had already suffered to Syriza in the elections, could it be that any PASOK voters switching camps by late May went the other way, to ND, to rescue the bailout deal which ND and PASOK had wrested together? And will that dynamic have continued through the pollingless period, leaving PASOK with even less support than they got in the elections last month?
Independent Greeks (ANEL)
(click chart for full size)
Paradoxically, the biggest loser of the post-elections landscape in May was one of the parties that had done very well in the elections: the Independent Greeks (ANEL). Angry about the 'Memorandum', these politicians had broken with their party, the ND, and ran a populist campaign. Not without success: 10.6% of the vote is not bad at all. But no sooner were the results of the elections known, or the party started falling in the polls. All pollsters agreed too: ANEL was losing a fair share of its newly won voters. By the time the last polls were published, all pollsters had the party a 6-8%.
Communist Party (KKE)
(click chart for full size)
The unreconstructed, unyielding communists of the KKE, who received a respectably but unexciting 8.5% in the May 6 elections, are also expected to suffer in these follow-up elections. Whereas Syriza's Alexis Tsipras has drawn up an ambitious "National Recovery Plan" for a Syriza-led, left-wing coalition government to enact, which would substantially increase public expenditures as share of the GDP, the communists are refusing to consider any coalition government with anyone. Basically, they're waiting for the revolution, and the loyal comrades on the island of Ikarias – a well-known ‘red island’ where many communists from the Greek Civil War were exiled, the World Elections website explains, and where the KKE received 41% of the vote in May - will probably wait with them. But the polls suggest that about a third of the party's voters from May 6 skipped over to Syriza as soon as the contours of the new political landscape became clear.
(click chart for full size)
That said, it should be added that most of the polls also showed the KKE stabilizing or even (in the case of 5 of the 10 polls) recovering ever so slightly in the last week of May. Most of those closing polls had the KKE at 5-7%. Should a certain defeatism have crept in, during these last days before the elections, about Syriza's chances to beat ND into first place, I would not be surprised if some communists stay with the good old KKE after all, and the party's score is closer to what it received last month. I suppose that the rumours of a poll showing a safe pro-bailout majority that apparently had the stock markets rallying on Thursday could create such defeatism.
Democratic Left (DIMAR)
(click chart for full size)
DIMAR, the Democratic Left, is the only of the smaller parties that did not seem to suffer at all in the post-elections landscape of May. The party pursued a carefully calibrated political course during and after the attempts to form a new government, walking the tightrope between looking obstructive and irresponsible and being roped into an impopular pro-Memorandum government. Like Syriza, DIMAR wants to undo the Memorandum - but only with a guarantee that Greece will remain in the Eurozone. Unrealistic as this middle road may seem, it must have bolstered the image of the party as one that will not sell out Greece like PASOK is seen to have done by many on the left, but is more realistic, constructive and cautious than Syriza. The last polls that were published uniformly pitted DIMAR at 5-8% of the vote, right in line with its May 6 election result of 6.1%.
Golden Dawn (XA)
(click chart for full size)
Greece's fascist party the Golden Dawn (XA) is not, in either program or political conduct, comparable to the kind of far right populists that have met success across Europe, from France and Austria to the Netherlands and Scandinavia. A better comparison would be with the neo-nazi bootboys that make up the rank and file of Germany's NPD. The formerly minute party getting 7.0% of the vote was a fairly shocking illustration of the degree of electoral drift in the country.
If this point needed any more illustration, it was provided when an XA politician, in a live televized election debate, threw a glass of water at a female Syriza politician -- and then hit the KKE politician at the debate, repeatedly, when she rose to stop him. The man's ire was apparently raised when the Syriza speaker recalled police investigations against him over a case of armed theft in 2007. With the cameras still rolling, the man then fled the studio, and his arrest was subsequently ordered.
The polls, before they stopped appearing, had already seen XA support slipping. The party had benefited from discontent about illegal immigration in exurbs and working-class suburbs, but has mostly been seen to have received many general protest votes from people who are just fed up of the whole political system. One would hope that the party will not again draw those votes now, though you can never be sure. The other question is which party would benefit from a demise of Golden Dawn. The party won its votes in May with fiercely anti-Memorandum politics, so you wouldn't expect XA voters to go (back?) to New Democracy now. But it also seems unlikely that XA voters will now switch over to Syriza on the far left. ANEL might be a respectable alternative, but doesn't seem to be faring too hot in this new election campaign itself.
Recreate Greece – Action – Liberal Alliance (DX-DRASI-FS)
(click chart for full size)
Will the new fusion list Recreate Greece – Action – Liberal Alliance (DX-DRASI-FS) make it over the 3% electoral threshold? If it does, that could come as a relief of sorts to New Democracy and PASOK, in particular New Democracy. The liberal parties that make up this list are in favour of economic reforms, market-friendly, pro-EU. If they make it in, they would dilute the strength of the anti-bailout parties and make for a convenient third coalition government party. Ironically, the thing that might keep them from crossing the electoral threshold is the dynamics of the race between New Democracy and Syriza to become the largest party, which is vacuuming center-right votes into the ND camp. As the chart shows, all the pollsters saw the chances for DX-DRASI-FS as touch and go.
Choosing the relevant axis
The main issue in both last month's elections and those of tomorrow is, of course, the Memorandum, the bailout deal, the austerity measures that come with it. Government coalition negotiations floundered on this issue last month. So did polling show any gradual changes in the proportions of support for the parties who in principle defend the deal (even if PASOK and ND also vow they will try to renegotiate some of the terms) and the parties that oppose it?
(click chart for full size)
What the polls showed, in May, was some further polarization between the camps, though this seems to have been concentrated in the days right after the elections, when people found out how the results had reshuffled the party landscape and government negotiations quickly failed.
With increased support for New Democracy offset by decreased support for PASOK and DX-DRASI-FS, the Memorandum 'defenders' (so to say) went from 38.6% in the elections (or 41.5% if you count LAOS, which became too small to include in this chart) to the lower-to mid-40s by mid-May - and then stayed there.
Among the anti-bailout forces, Syriza gains were offset by ANEL, KKE and XA losses. Together, the four parties had pooled 42.9% of the vote on May 6. As soon as polling kicked off a couple of days after the elections, it showed these four anti-Memorandum parties at around 46% - and if anything, a percentage point or two was shaved off again on average since.
In a system based purely on proportional representation, DIMAR would hold the balance. In practice though, it will be the 50 seat bonus for the largest party that will be decisive. Together, New Democracy and PASOK were polling distinctly better by the end of May than how they did in the May 6 elections - so considering that the two parties fell just 2 seats short of a majority even after those disastrous elections, a majority is in easy reach for them this time as long as ND beats Syriza into first place.
Should Syriza become the largest party - and at this point I would not bet money on that - options for coalition government are much more unclear. While one can place ANEL, Syriza, KKE and XA on one end of a chart based on their anti-bailout positions, it is hard to imagine any of these parties forming a coalition government with any of the other ones at this point.
(click chart for full size)
For the same reasons, I also have trouble imagining a lot of voter movement between the anti-bailout parties. Will Independent Greeks voters or Golden Dawn voters really shift to Syriza? So I also created a parallel chart, based on the good old left-to-right standards.
This chart shows the right-wing parties roughly stable at 40-45% of the vote in the elections and throughout all the May polling. Whatever ANEL, XA and the liberal list lost in the weeks after the May 6 elections, the ND won. It's not quite as neat as the chart suggests, since the yellow space for "Other parties" includes the right-wing, conservative party LAOS, which received 2.9% of the vote on May 6 and whose support has since dwindled, presumably to the benefit of ND. So overall there must have been some net cross-over from right to left, but it won't be much. That's net, though - if my above speculation that PASOK may have started to lose some votes to ND in that last week of polling holds true, for example, there must have been movements of equal size from right-wing parties to Syriza or DIMAR.
The main shift that catches the eye in this chart, instead, is the depletion of the share of "Others" (and the KKE), which is filled out by the increasing share for Syriza. In that light it's useful to list the parties that are included in the Others category: LAOS (conservative, as mentioned above), the Ecologist Greens (2.9% on May 6), the Anticapitalist Left party ANTARSYA (1.2% on May 6), Social Agreement (a party established by ex-PASOK politicians, 1.0% on May 6), the "I Don't Pay" movement (0.9%), the Union of Centrists (0.6%), the National Unity Association (0.6%), the Greek Pirate Party (0.5%), the Political Party of the Successors of Kapodistria (0.5%), and three radical communist splinter parties pooling another 0.4%, one of them bearing the intimidating acronym KKE(M–L) / M–L KKE. So if the vote for these various mini-parties does collapse as the polls suggest, it is not surprising that this would primarily benefit Syriza. The predominantly centrist-to-leftist nature of these smallest parties also means that the left, overall, received over 50% of the vote on May 6, and must have been polling in the mid-50s by the end of May. That does not, however, remotely mean that we can necessarily look forward to a left-wing government after tomorrow's elections.
All the polls
In this post, I have used the polling data that is made available on the Wikipedia page Greek legislative election, June 2012. However, while the results of some polls are provided in numbers of decided voters and the results of others are provided in numbers of both decided voters and all voters (including undecided and non-voters), there are also results of polls provided only as percentages of all voters (including undecided and non-voters). All depending on the data the Wikipedia editors could find in the media and on pollster websites. This is very scrupulous, but makes comparing results from pollster to pollster difficult. I have therefore been so free as to recalculate the last category of polling results into data of only decided voters. The result is this table:
(click table for full size)