There were nine (!) posts yesterday, several from a guest poster. The first post yesterday was “Changes (Personal/Professional):”
Some news: I have informed Princeton that I will be retiring at the end of next academic year, that is, in June 2015. In August 2015 I will join the faculty of the Graduate Center, City University of New York, as a professor in the Ph.D. program in economics. I will also become a distinguished scholar at the Graduate Center’s Luxembourg Income Study Center. You can read all about it in the Graduate Center’s announcement.
None of this will have any effect on my work at the New York Times, so unless you’re just curious about the guy behind the column and blog, you needn’t read further.
So, why am I doing this? It is in no sense a commentary on Princeton, which has been a wonderful place for me professionally and personally. In particular, I can’t praise Princeton’s intellectual quality enough: it has been a great honor to be affiliated with a superb public policy school and an equally superb economics department.
Instead, my move reflects some hard thinking about how I can best make use of my time.
Fairly obviously, the center of gravity of my work has shifted over time toward more of a public policy focus; and of course I also have a fairly unusual role as an academic who is also a columnist at the world’s greatest newspaper. Meanwhile, I’m now 61, and I realized that it’s time to take a hard look at where I really want to be at this point.
In terms of geography, the answer seemed clear on reflection:
somewhere near Zabar’s
New York is the best place to pursue my current interests. Given that decision, I sought an academic base.
Here too the answer seemed clear: more and more of my work has focused on issues of income inequality, and nobody does more important work producing the hard data on which all of this work relies than the Luxembourg Income Study, directed by Janet Gornick, professor of political science and sociology at the Graduate Center of CUNY. So I approached Janet about the possibility of some kind of affiliation with LIS that would give me both an office and the ability to interact with the excellent group LIS has assembled in New York.
That is indeed going to happen, but to my surprise the Graduate Center offered me a faculty position as well, which I quickly realized would be a wonderful thing. The Graduate Center is a remarkable locus of public affairs-oriented scholarship, and is in the process of assembling an incredible team associated with LIS. I couldn’t imagine a better place for me at this point; I also, to be honest, like the idea of being associated with a great public university.
I suppose there’s also a bit of coming home here. I grew up in the New York suburbs, and always imagined that “the city” was where intellectuals went to live once they, as my grandmother used to say, reached “mature adultery.” At any rate, I truly do love New York, so that there will be
a lot more good restaurants
a lifestyle gain to complement the professional advantages.
So there you have it. Again, this will have no effect on my work for the Times.
The second post was “Hungary: An Election in Question (Part 1):
This is the first of a five-part series from my Princeton colleague Kim Lane Scheppele, head of the Law and Public Affairs Program.
Hungary: An Election in Question
Part I: The Political Landscape
Kim Lane Scheppele
Princeton University and the Institute for Advanced Studies, Princeton
24 February 2014
Hungary’s parliamentary elections will be held on 6 April. And it is already clear who will win.
Unless something truly surprising occurs, the governing party Fidesz is headed to victory. The only uncertainty is whether it will again win two-thirds of the parliamentary seats, a result that would continue to allow it to change the constitution at will.
Fidesz won the last elections in 2010 fair and square. But this time the election is unlikely to be judged so favorably. The whole election framework – the laws, the institutions and even the new electorate – favors Fidesz because the governing party has used its four years in office with its two-thirds majority in the parliament to redesign every aspect of the electoral system to its advantage.
Fidesz also overwhelmingly dominates the offline media and has closed off almost all avenues through which opposition parties can reach the electorate. New decrees from local Fidesz-affiliated officials around the country and misleading instructions from election officials are creating last-minute campaign obstacles that put the opposition even more on the wrong foot.
Under the new election framework, the allied opposition parties cannot win a parliamentary majority, even if they gain more votes than the governing party. Simultaneously, the changes also make it nearly inevitable that the governing party will keep its two-thirds parliamentary majority even if it gets less than half of the overall vote.
Róbert László of the Political Capital think tank in Budapest shows how Fidesz can win a two-thirds majority with less than half of the party-list vote. His model also predicts that a united center-left opposition would need about 6% more votes than Fidesz to win a simple majority in the parliament.
Central European University Professor Gábor Tóka estimates that, under the new system, a united center-left opposition might get 8% fewer parliamentary seats than Fidesz if both got an equal share of the votes.
Political Capital’s “mandate calculator” permits everyone to try out different models and different assumptions. We tried it here in Princeton and, depending on the assumptions one makes about the nature and shape of the opposition, Fidesz could get its two-thirds majority in parliament pretty easily with only 48% of the vote if the other parties perform as polls indicate they would if the election were held now. If the foreign votes split 85/15 for Fidesz (not unreasonable for reasons I will explain), Fidesz could get its two-thirds with only 44%. If Fidesz wins by the same margin it won last time, with 53% of the party-list vote, it would get 76% of the seats in the parliament instead of the 68% it won under the old system.
In short, Fidesz has designed the election to allow itself to win big, even without majority support. Or, to put it differently, Fidesz has designed the election so that the opposition loses even if it wins.
These effects occur because the way that the districts are drawn and the votes are aggregated. It doesn’t even count all of the other things that Fidesz is doing to help the opposition lose, like monopolizing the media, operating an election office that is giving out misleading instructions and only selectively registering to vote Hungarian citizens who are living abroad.
If Fidesz is reelected under this self-dealing system, then it is hard to avoid the conclusion that the election has been rigged. Prime Minister Viktor Orbán’s “mandate” will be tainted.
It’s serious to accuse an incumbent party of potentially rigging an election, so the evidence needs to be strong. In this series of five blog posts, I will show precisely how the outcome of the election is cooked into the rules even before a single ballot is cast. The rules were designed to look “normal” but to allow Fidesz to win in a very particular political context, which is where we will start.
As Fidesz officials are quick to argue, they will win the election because they are the most popular single party in Hungary. Which is true (see graph below). But Fidesz’s popularity has only recently climbed above 30%, a level that would cause analysts in most democratic states to predict that an incumbent party is in trouble, especially given how low Fidesz fell over the last several years. What makes Fidesz look like a winner, however, is that all of the other parties are even less popular.
For the last month, however, Fidesz has been confronted by a more substantial opponent than it has had during its tenure in office so far. Five left-leaning parties calling themselves the “democratic opposition” have combined to form the Unity Alliance (Összefogás). They have put forward a common slate of candidates for the individual constituencies and they are running a joint party list. Their joint strength might just be enough to challenge Fidesz’s domination of the elections – if there were a level playing field. But they were late to the election party, so to speak, announcing their joint effort only on 14 January 2014 just before the election date was set. So they have some catching up to do.
(The five parties in the Unity Alliance are the Socialists/MSzP headed by Attila Mesterházy; Together 2014/E-14 headed by Gordon Bajnai; Dialogue for Hungary/PM led by Benedek Jávor; Democratic Coalition/DK headed by Ferenc Gyurcsány and the Hungarian Liberal Party/Liberálisok headed by Gábor Fodor. Since the coalition was formed, the Movement for a Modern Hungary/MOMA, headed by Lajos Bokros, a conservative MEP, has agreed to support the joint ticket.)
And then there is Jobbik, which its detractors call the “non-democratic opposition.” This far-right party has become internationally known for its anti-Semitism and anti-Roma agitation, its toxic assertion of nationalism, and its ideology so far beyond the edge of the European political spectrum that its three representatives in the European Parliament cannot affiliate with any party caucus. Fidesz might reasonably worry that it would lose votes to Jobbik on the right, which may be why many – including Jobbik’s leadership – claim that Fidesz is “stealing [their] issues and ideas.” For its part, Jobbik’s campaign ads this year portray it as substantially more moderate than its reputation in order to steal voters from Fidesz.
At the moment, Jobbik seems to have the allegiance of just under 10% of the electorate, though some worry that Jobbik’s support may climb again to the 17% of the vote it won in the 2010 election. Jobbik cannot form a government with that vote, but is the only party that can seriously challenge Fidesz’s electoral strategy by dividing the vote on the right.
Just as Fidesz faces a challenge to its base from Jobbik, the Unity Alliance is challenged by a party called Politics Can be Different (LMP) that provides an alternative for its voters as well. In the last year, LMP – a small party to begin with – split so that one fraction joined the broader opposition alliance and the rest remained unaligned. While LMP lost support since the split, it still seems to be polling around the 5% threshold needed for a single party to enter the parliament.
Though Fidesz and the Unity Alliance are the two big parties in this race, polling data show that the largest single voting bloc – a clear majority of the electorate for the last several years – is still “undecided.” That large number becomes even more formidable when one considers that more than half of the Hungarians asked do not answer surveys. Is it hard to know if those who do not answer are still engaged in politics at all, and if so, how.
In past Hungarian elections, the turnout had to reach 50% for the election to be valid. But Fidesz changed that rule too so that there is no minimum turnout required any longer. Low voter turnout, then, is no barrier to a valid election.
But even with the large number undecided or apolitical voters, the results are not in doubt. The governing party designed the system precisely to prevent surprises in this particular political landscape, and they wrote the rules to allow themselves to win almost no matter which way opinion breaks and almost no matter what the turnout is on election day. It is hard to see a realistic outcome for this election that doesn’t put Fidesz front and center in the next government. Fidesz will thrive if there is low turnout because the party has a powerful system for bringing out its voters. If Jobbik surges, Jobbik could not govern unless Fidesz were the dominant partner in a coalition. But, perhaps most importantly for judging the fairness of the election, Fidesz will win even if the “democratic opposition” were to pull ahead of them by a substantial margin.
Why is that? According to election experts, the Unity Alliance could only gain a parliamentary majority if it won by more than a comfortable margin in the popular vote. That is because of the way that the system has been designed. Unless there is swing toward the left that is larger than anything we have seen in the post-communist period or unless Jobbik’s support rises by so much that it substantially depletes the Fidesz vote, Fidesz will surely win outright and is very likely to get its two-thirds back again.
How could Fidesz win under almost any likely scenario for 6 April? I will turn to that next.
The third post yesterday was “Hungary, An Election in Question, Part 2:”
Hungary: An Election in Question
Part II: Writing the Rules to Win – The Basic Structure
Kim Lane Scheppele
Princeton University and the Institute for Advanced Studies, Princeton
24 February 2014
How did the governing party Fidesz stack the deck so much in its favor that the upcoming Hungarian election’s results are not in doubt?
Fidesz started immediately after its election victory in 2010 to reshape the electoral system to ensure its hold on power. The Fidesz parliamentary bloc, which enacted constitutional changes without including or consulting any opposition party, slashed the size of the parliament in half, redrew all of the individual constituencies unilaterally, changed the two-round system to a single first-past-the-post election for individual constituencies, and altered the way votes were aggregated.
Moreover, Fidesz has granted dual citizenship and therefore voting rights to ethnic Hungarians outside the borders who are overwhelmingly Fidesz supporters, while at the same time maintaining a system that makes it comparatively harder for Hungarian citizens living or working abroad to vote.
The media landscape and campaign finance rules overwhelmingly benefit Fidesz and a series of last-minute changes to the law just before the campaign started put the newly united center-left opposition at an even greater disadvantage. In addition, the governing party has captured the election machinery which is now staffed with its own loyalists.
The sum total of all of these changes makes it virtually inevitable that Fidesz will win.
The devil is in the details, so let’s walk step by step through these various ways that the governing party has changed the rules in its favor.
As one of its first acts in office, on 25 May 2010, the Fidesz parliament amended the constitution it inherited to cut the parliament’s size in half. This was a move lauded by all sides of the political spectrum, as the old 386-member parliament was widely perceived as too large to be effective and too expensive for a small country in debt. The new 199-member parliament that will be seated after the 2014 elections will represent new electoral districts that had to be newly drawn to accommodate this new, smaller parliament. Redrawing the districts was not only widely welcomed, but also required by the Constitutional Court, which had ruled (first in 2005 and again in 2010) that the old districts had become too unequal in population size to give all citizens an equal vote.
The old districting system already favored Fidesz because the larger districts were in the urban strongholds of the left and the smaller districts were in the rural districts of the right. As a result, rural conservative votes were given more weight because it took fewer of their votes to elect an MP. But the way that Fidesz redrew the districts for 2014 gave their party an even greater advantage than they had before.
Without any consultation with opposition parties, Fidesz enacted a new “cardinal law” in 2011 that simply set the boundaries of the districts (Law CCIII/2011). While most election laws provide principles for drawing districts and assign some neutral or at least multi-party body to actually draw the boundaries, the borders of the districts in Hungary are now written directly into the law. Moving a district boundary by even one block requires a two-thirds vote of the parliament. The districts are therefore heavily entrenched and were not the result of either a public or an inclusive process. No justification for these districts was offered by the governing party.
Of course, not all districts in any electoral system have identical numbers of voters. But how much can districts vary before they deny equality of the vote? The Commission for Democracy through Law (the Venice Commission), recommends no more than 10% variation as the international standard. The Venice Commission is not terribly clear about what this means, but given that the Venice Commission is working with a principle that demands that votes be weighted as equally as possible, one can guess that this means that districts should not vary by more than 10% in population overall.
The Hungarian law is fiendishly clever in appearing to come close to that standard while being miles away from it. The Hungarian Election Law (Act CCIII of 2011 – section 4(4)) mandates that the districts should not vary by more than 15%. The Venice Commission was not thrilled with the difference, but let it pass. They shouldn’t have.
A closer reading reveals the trick. The Hungarian law requires that districts vary by no more than 15% calculated from the mean number of voters in the district. This is not an overall 15% deviation, as the Venice Commission presumed, but is instead a standard that permits districts to vary by 15% below the mean and 15% above it.
An example demonstrates what a huge difference this makes. To aim at an average district containing 100 voters, a 10% overall deviation would permit districts to vary between 95 and 105 voters. (Divide 95 by the 10 voters that separate the largest and smallest districts and you get about 10%). The Hungarian law would permit districts to vary between 85 and 115 voters – 15% above the mean and 15% below the mean of 100. The gap between 85 and 115 voters in a district would be 35% overall! (Calculated the same way as above: 30/85 = 35%.) This is a huge difference that the Venice Commission did not seem to see.
In the actual districts were constructed as a result of the new election law, the variation became even larger than that. As you can see in the chart below, the smallest districts in Hungary now have about 60,000 voters while the largest districts have nearly 90,000 voters, roughly a 50% gap. (The horizontal axis shows the number of eligible voters in the new constituencies based on voter data from 2010, and the vertical axis shows the number of districts in the new scheme with that number of voters.) Not only are the actual districts highly unequal, but this variation has no apparent justification.
The Size of Parliamentary Districts in Hungary after Redistricting. Source: Calculations by Gábor Tóka, Central European University
Hajdú-Bihar County, in the eastern part of Hungary, provides a case in point. A last-minute amendment to the 2011 election law divided the city of Debrecen into two districts of highly unequal size. Now, one district has 87,278 voters and the other, right next to it, has 60,125 voters. These are very nearly the largest and smallest districts in the country, side by side, without official explanation.
The government may have given no reasons for its districts, but this huge variation in district size is not random. As Political Capital shows, the left-leaning districts are systematically 5,000-6,000 voters larger than the right-leaning districts, which means that it takes many more votes to elect someone from a left-leaning district than to elect someone from Fidesz.
The borders of these new districts also appear to be drawn to Fidesz’s advantage, since they just happen to break up the areas where the opposition alliance voters have traditionally been strongest and they scatter these opposition voters over a new Fidesz-majority landscape. Historically left-leaning districts were partitioned and blended into historically right-leaning districts, creating fewer districts where left-leaning candidates are relatively certain to win.
One of the most obvious gerrymanders occurred (again) in Hajdú-Bihar County. In the 2006 election, which went nationally by a wide margin to the Socialists, the county voted three of its nine districts for the Socialists and six for Fidesz, as you can see in the chart below, on the left. If the results from the 2006 election were tallied in the newly drawn six districts for that country, as shown on the right, Fidesz would now win every district. The map reveals that this all-Fidesz result was accomplished by drawing the districts to divide up the compact concentrations of Socialist voters so that they would become minority voters in Fidesz-dominant districts. Examples like this one can be found all over the country, as left-leaning districts were partitioned to break up clusters of opposition voters to mix them with even more conservative voters from neighboring areas.
The US may have invented the gerrymander, and so it may seem presumptuous for an American to complain about the new districts. But the Hungarian gerrymander is different from the (also outrageous) American type. In US national elections, gerrymanders occur at the state level, which means one party cannot redistrict the whole country at once. In the US, districting plans are also subject to judicial review to check the worst self-dealing. In Hungary, however, the whole country was redistricted by one party all at once so the Hungarian gerrymander is far more decisive. And there is no judicial review to correct excesses. In addition, unlike in America where the governing parties in the states get a new shot at gerrymandering every 10 years, after each census, it will take a two-thirds vote of the parliament to change any district in Hungary’s future.
Hungarians don’t just cast votes for individual representatives in districts of the sort we have just seen, however. Hungarians cast two votes in national elections. In addition to casting ballots for representatives in the voters’ individual constituency, voters cast second ballots for party lists. Those votes are aggregated across the country and additional parliamentary seats are awarded to parties based on these results, above and beyond the seats won in the individual districts.
In the new parliament as in the old one, MPs elected both ways sit together with equal status. While this dual system of MP elections appears to mitigate the effect of the gerrymander, the new parliament, unlike the old, allocates more seats to the individual constituencies than to the party-list mandates. The new parliament features 106 district mandates and 93 party-list mandates. Since individual constituencies are awarded on a winner-take-all basis, this tilts the system toward an even more disproportionate distribution of mandates than in the prior also-disproportionate parliament.
Individual constituencies in Hungary were allocated from 1990 to 2010 in a two-round run-off system. Unless a candidate won 50% or more in the first round, a second round would be held between the highest vote-getters to determine who won the mandate. This system meant that many political parties would field candidates in round one, and then form coalitions before round two after the relative viabilities of the individual candidates could be assessed. Hungarian political culture grew up around this system so that parties were not accustomed to bargaining before any votes were cast.
The new electoral system in Hungary eliminates this second round, benefiting Fidesz, as the largest single party. It can now win districts outright without needing majority support because it only has to get more votes than any other party on the (single) election day to capture the constituency. Given that the districts have been drawn to give Fidesz an advantage overall, one can imagine other parties will have a hard time winning constituencies which have been constructed precisely so that Fidesz is the largest party.
The design of the new system means that the democratic opposition would only have a chance to win individual constituencies if the various opposition parties of the left could create a grand coalition before the election so that they didn’t run candidates against each other. But this was a result that everyone familiar with politics in Hungary knew would be hard to accomplish. The parties in the “democratic opposition” (excluding Jobbik) are sharply divided both by ideology and personality. But unless these parties could set aside their differences to unite, they would surely lose.
The announcement on 14 January that five parties in the opposition had managed to agree on a single list of candidates for the single-member districts as well as a common party list was therefore something of a political miracle.
But can the party leaders of the Unity Alliance bring all of their voters along with them? Many voters for the smaller parties on the left often don’t trust the larger Socialist Party which now dominates the coalition. And some personalities in the mix are popular only within their own parties and unattractive to the others in the coalition. As a result, it cannot be assumed that votes for the five parties can simply be added together to produce a united whole that is the same size or even larger than the sum of the parts.
Because voters cast two ballots on election day, the individual constituencies are only part of the story, though they are the largest part. Parties will also run national lists to compete for voters’ second votes. The new conditions that came into effect since the last election actually make it easier than it was in 2010 to nominate candidates for the individual constituencies and to register parties with national lists, something that is consistent with a dominant-party strategy to divide up the opposition as much as possible.
But the party-list system also builds in incentives for small parties to join together to form a larger alliance. To be approved to run a national list, parties must field candidates in at least 27 individual constituencies in at least nine of the 19 counties plus Budapest. While this guarantees that parties are truly national, it also aggravates the problems created by the loss of the second-round runoff in the individual constituencies. Any new national list adds to the “clutter” of individual candidates in the individual constituencies and further fragments the vote.
So it makes sense, under these rules, for small parties to form a common national list. To avoid competing head-on and perhaps pushing each other below the 5% threshold for entering the parliament, small parties on the same side of the political spectrum are pushed by the logic of the system to join forces. But as soon as they do so, they run into another problem. In all elections since 1994, parties have had to meet a 5% threshold of the popular vote to gain a fraction in the parliament. For two parties that run together, the threshold rises to 10% and for three or more parties, the threshold is 15%.
If the smaller parties were going to unite for 2014, then, they ran the risk of together missing the higher threshold required of joint party lists. The rules of the game have therefore pushed the small parties of the “democratic opposition” to do what they did – which was to join with the Socialists to form Unity. Only an alliance with the larger Socialist party guaranteed that these smaller parties would be able to enter the parliament given the higher thresholds for joined lists. Because many of the smaller parties were created precisely to distance particular groups of voters from the Socialists, however, this is an uneasy alliance at best.
So that is where we were as the campaign was launched, witnessing a democratic opposition alliance whose members do not like each other much but who have to work together if they are to have any hope of ousting Fidesz given the way that the rules are structured. The public squabbling that occurred as the grand coalition went together belied the name of Unity Alliance and weakened their electoral position. They have the campaign period to convey a new unified message, but – as we will see – that is going to be very hard.
Next up, the fourth post yesterday was “Hungary, An Election in Question, Part 3:”
This is the third of a series of posts by my Princeton colleague Kim Lane Scheppele, head of the Law and Public Affairs Program.
Hungary: An Election in Question
Part III: Compensating the Winners
Kim Lane Scheppele
Princeton University and the Institute for Advanced Studies, Princeton
24 February 2014
Election analysts have predicted that the democratic opposition in Hungary cannot win a majority in the parliament if it produces a tied vote or even pulls somewhat ahead of Fidesz in the final vote. Instead, the allied opposition parties will have to get as many as 6-8% more votes than Fidesz to gain a simple parliamentary majority, mainly because of gerrymandered districts. But that’s not the only Fidesz-friendly element of the new electoral system.
If Fidesz wins these new single-member districts by substantial margins, Fidesz’ parliamentary representation will then be boosted even more by a novel system of “compensation votes.” To understand how this works, we need to understand how proportional representation (PR) systems are typically structured.
In PR systems, compensation is typically awarded in the calculation of final results to ensure that the distribution of seats in the legislature is as close as possible to the distribution of votes cast by the electorate. That’s what makes them proportional. For example, German parties are compensated by gaining extra votes in their party list totals when their candidates win a lower share of individual constituencies than the popular vote would predict.
Hungary’s system for awarding compensation based on the results in the single-member districts used to be quite similar to the German one, but no longer. The new electoral system now bizarrely compensates not just the losers, but also the winners. This new system increases the winner’s victory margin to create even more of a “winner take all” system. This will most likely ensure that the final tally of votes moves farther from the distribution of votes in the population as a whole rather than closer to the overall distribution, as PR systems typically ensure.
In short, majorities are magnified into super-majorities under this new system.
The new system of compensation is complicated and counterintuitive, so let’s start with the basics.
What is a “lost vote”? A lost vote is a vote for a candidate who loses. If, for example, you are voting in a district that has Red, Green and Yellow parties on offer and you vote for the Yellow candidate, who loses, your vote is considered “lost.” But that lost vote is used instead to help the party you voted for get an extra boost when party-list mandates are calculated. As a result, you are compensated for having “wasted” your vote for the individual candidate by having your vote supplement the party-list totals instead. This is the system Hungary had for compensating lost votes in the individual districts from 1990-2010.
Let’s take an example. Suppose the Red party wins a district with 500 votes while the Green party gets 200 votes and the Yellow party 100 votes. Under the old Hungarian compensation scheme, 200 votes for the Greens and 100 votes for the Yellows would be added to the Green and Yellow party-list votes so that those parties gained in strength when party-list mandates were determined.
Under the Fidesz reforms, however, not only do the Green and Yellow parties get compensation, but now the Red party also will be deemed to have “lost votes” in this election despite having actually won the seat.
How did the winning party “lose” votes? Some Red votes are counted as “lost” because the mandate could have been won with only 201 votes and yet the Red party got 500, exceeding what the party strictly needed to win the mandate. So under the new Fidesz system, 299 votes – the number of votes beyond those necessary to win – are considered lost and are added to the votes for the Red party when party-list seats are awarded.
Under Hungary’s new election system, then, the party winning an individual constituency will be awarded not only that particular mandate, but also extra points in the party-list calculations when it wins by more votes than needed. This is another reason why the electoral system in Hungary is even more highly disproportionate in 2014 than it was before.
The reason for having a proportional representation system is to enable representation to be proportional to the vote. But the Fidesz system makes representation less proportional overall. This innovation puts Hungary out of line with all PR systems in Europe.
The winner compensation system was designed at a time when Fidesz was clearly the plurality party, with all other parties trailing at a distance even though, combined, they would have been more formidable. So Fidesz designed a system in which it would maximally benefit in that fragmented political landscape. If Fidesz won by large margins in the individual districts against a divided opposition, it could have gotten its two-thirds back even with substantially less than half the vote.
The system of winner compensation is therefore another reason why the opposition had to form an alliance, even if only to narrow the gap between the first- and second-largest vote-getter in each individual constituency.
An example shows why. If Fidesz won 500 votes in an individual district and four smaller parties obtained 100 votes each, Fidesz would get 399 votes in winner compensation. But if Fidesz won 500 votes and the Unity Alliance combined the votes of the four smaller parties to gain 400 votes in that district, then Fidesz would only get 99 compensation votes added to its party-list votes. With a unified opposition, the effect of winner compensation is blunted.
So when does winner compensation actually benefit a political party facing a united opposition?
A party would benefit from the winner compensation system if it could encourage a host of new challengers on the “other side” to chip away at the difference between the first- and second-place candidates in each district, throwing additional votes to the winner. And in fact, the new electoral rules make it easier in 2014 than it was in 2010 to field new parties and new candidates, by requiring fewer supporters to endorse them before they can be registered. While we don’t yet know the number of parties that will actually run lists and field candidates, already there are 92 parties that have registered with the National Election Commission. If there are many small “anti-Fidesz” candidates in a particular constituency, for example, they could divide the vote and increase the margin by which Fidesz wins – and therefore increase Fidesz’s likelihood of getting its desired two-thirds majority.
Of course, if the united opposition could sweep the individual constituencies by large margins, then they could also win a disproportionate victory on the party list side as well. But that is why it matters so much that the individual constituencies are drawn in a way to make that maximally unlikely. There are very few safely “left” districts remaining that the united opposition could win by such large margins. So while it is possible in theory for the united opposition to win a disproportionate victory under the rules also, the facts on the ground and the way that the districts have been matched to those facts make it virtually impossible in reality.
But this is not the only “winner compensation” system on view for the 2014 Hungarian election. The fact that Fidesz so decisively won the 2010 election has given it the power to remake and staff the institutions that will run the election this time. In fact, the whole election machinery itself is in the hands of governing party allies for 2014. And we are already seeing worrying signs that these offices are not neutral.
Twice since the 2010 elections, the Election Commission was reorganized and all members of the Election Commission were fired before they completed the ends of their terms. First, the members of the Election Commission elected by the previous parliament were fired when Fidesz passed a law in 2010 that required all Election Commission members to be reelected after each national election, effective immediately (Law LXI of 2010). The old members of the Commission, which included a mix of opposition and Fidesz members with opposition members in the majority, left office immediately and were replaced by a new Commission elected by the Fidesz parliamentary majority which included no members from the political opposition.
Then, in 2013, Fidesz changed the system yet again (Law XXXVI of 2013). This time, the law created a newly structured Election Commission and a newly structured Election Office. The new Election Commission now has seven core members nominated by the President of the Republic (himself a former Fidesz vice-president). They were elected for a term of nine years by a two-thirds vote of the Fidesz-dominated parliament. Not surprisingly, all of the new members of the Commission appear to be allied with the governing party. The Election Office is staffed by civil servants, but the head of the office now is a former deputy state secretary for the Ministry of National Development in the Fidesz government.
While opposition parties report good relations with the new head of the Election Office, one might well still worry about a system in which all of the key players who will make the decisions about the election framework were assigned to their jobs by the governing party, in a system where the governing party just rewrote all of the rules.
Even though the Election Commission has only government-friendly members among its permanent members, once the campaign starts, each party running a national list is able to delegate one person to sit on the Election Commission for the duration of the campaign. These party delegates are able to vote on all matters along with the seven permanent members, which raises the possibility that the permanent members could be outvoted depending on how many and what sorts of groups run national lists. Recently, in a press briefing to the Hungarian International Press Association, András Patyi, the head of the National Election Commission, said that he expected the Election Commission to increase to 20-25 members during the campaign, which means that he anticipates at least a dozen or more national lists. (In 2010, there were 10 lists on the ballot.)
In run-up to the campaign, however, Fidesz allies dominated the Election Commission. The Election Office will remain the key location for information about the election but it gains no members from opposition parties to assist in its operation during the campaign. Already important decisions have been made about how the election will be administered under these new rules. This is why, as we will see in the next blog post, the proliferation of inaccurate and misleading information about the election given out by election officials is especially worrying.
The fifth post yesterday was “
Hungary: An Election in Question
Part IV: The New Electorate (in which Some are more Equal than Others)
Kim Lane Scheppele
Princeton University and the Institute for Advanced Studies, Princeton
24 February 2014
Hungary’s governing party Fidesz didn’t just rewrite the rules for the upcoming Hungarian election. The governing party changed the electorate as well. Different categories of citizens can now vote in different kinds of ways, which creates the very real possibility of unjust discrimination.
The 2014 election features two new voting systems that restructure the electorate and its options.
One permits each major nationality (i.e. minority) group in Hungary to elect a representative of its group to the parliament on a “preferential” basis that requires only one-quarter as many votes to claim the mandate. This system of positive discrimination may look admirable, but in practice limits rather than expands voting options of minority populations, as we will see.
The other gives the right to vote to ethnic Hungarians who never had permanent residency in Hungary. These are people to whom the new constitution has given a route to expedited citizenship upon application. New Hungarian citizens can now register and vote more easily than citizens who have permanent residency but who are abroad on election day. As a result, new dual citizens with the most tangential relationship to Hungary can more easily influence the election than can long-standing citizens whose primary political identity rests in Hungary.
In both cases, these voters with new options are being herded toward Fidesz-friendly results and away from support for the united opposition both because of the new rules and because of the confusing and misleading communications issuing from the offices in charge of running the election. Let’s take these new sorts of voters one by one.
In a move welcomed by the Venice Commission, the new election framework lays out a system in which members of 13 designated ethnic minority groups may vote for a “nationality list.” Though it is called a list, in practice it consists of one person because each minority group can only elect one representative in this new “preferential” way, while all subsequent representatives from the group are elected according to the more demanding conditions necessary to elect a representative on a party list.
While Germans, Romanians, Ukrainians and other registered groups possess the right to elect a minority representative in theory, the Roma constitute the only group who are likely to be able to muster the numbers to elect such a representative in fact.
This new system of nationality representation, however, comes with a number of catches.
First, members of minority groups who want to take advantage of this possibility must sacrifice their ability to use their second vote for a party list when they use their second vote to elect a nationality representative. This system therefore limits the incentives for political parties to court minority voters since minority voters cannot vote for parties if they vote for the nationality representative, further marginalizing them.
Then, minority voters must register in advance to take advantage of this option. According to the Electoral Procedure Law (Law XXXVI of 2013), minority voters must register at least two days before the election. Once they register, they cannot change their minds on election day itself to vote for a party list instead. (They can change their minds before the registration deadline.) The only choice that the registered minority voters have when election day comes is to vote for the representative of their group on offer, or to fail to cast their second ballots. This system, as a result, locks in the minority vote before the end of the campaign. Unlike the situation for any other voter, minority voters cannot decide in response to the full campaign whom to support.
Finally, and most consequentially, the specific candidate chosen to stand for election as a representative of the minority group must be, by law, selected by the national minority self-government, a body that was elected by each minority group in a special election four years ago. (These self-government organizations have been elected periodically since the mid-1990s to ensure representative decision-making bodies for minority affairs.) But the national minority self-government for the Roma at the moment is run by a group called Lungo Drom, whose leader, Flórián Farkas, is a Fidesz MP.
In short, if Roma choose to vote for a nationality representative, they cannot vote for a political party and their only choice is to elect a Fidesz MP, using their second votes that could have been used for any party list. Registering to “vote minority” therefore gives Roma no party choice at all. They must vote for a governing party representative.
Roma don’t have to register to vote for the nationality list if they don’t want to. But a letter sent in January from each local Election Office to all voters announced on the first page that Roma would have to register if they wanted to vote, and only on the second page explained in not-entirely-clear prose that Roma had to register only if they wanted to vote for the minority representative. In even more confusing language, the letter revealed that in doing so, Roma would lose the ability to use their second vote to for vote a political party.
When the letter went out, Roma started to register to vote in substantial numbers, largely unwittingly, for the minority representative. So far, the Election Office has not issued any correction, raising questions about what it was doing with its initial letter telling Roma to register to vote. Given that Roma who registered would find themselves excluded from being able to vote for the party lists on election day and would only have the option of voting for a Fidesz MP instead, this mix-up is worrying, especially when the governing party staffed the new Election Office.
The Election Office seems to be contributing to the confusion over the system for Roma voting in other ways as well. While the law clearly says that the nationality voters clearly have until two days before the election to lock in their vote for the nationality candidate (Law XXXIV of 2013, section 249), Ilona Pálffy, the head of the National Election Office announced in a press briefing to the Hungarian International Press Association on 29 January 2014 that nationality voters would have to register no later than eight days before the election and could not change their minds after that.
In fact, when I was interviewing officials and party representatives in Budapest about the new election framework recently, I often got different answers from different people about what the law required. When one gets an answer from the head of the National Election Office that differs so strikingly from the plain wording of the law, however, that is especially alarming. Will Roma be told, if they try to change their minds in the last week and “unregister” from the nationality list, that they can’t do so even though the law says otherwise? I hope that the National Election Office clarifies just what they believe the rule is – before the election.
It’s not just the Roma who have new rules about voting this time. The other newly registered group of voters consists of ethnic Hungarians living abroad who were given the right to apply for citizenship under the new Fidesz constitution. For historical reasons, the only Hungarians whose ancestors lost their citizenship en masse were living in the territories that had been part of historic Hungary but that were allocated to neighboring states by the Treaty of Trianon in 1920. (Hungarians who left Hungary for other countries before or since retained their citizenship unless they explicitly renounced it.) This constitutional change made millions of ethnic Hungarians eligible for expedited citizenship, the vast majority still living in the neighboring countries.
As a result of the new citizenship law, about 575,000 Hungarians, primarily from the Trianon territories, have become citizens in the last year. (I’ll call them the Trianon Hungarians.) And as of mid-February, about 150,000 of them had registered to vote. But the deadline for registering to vote is 22 March so only then will it be clear how many of the new citizens will be new voters as well.
In three of Hungary’s neighbors, Ukraine, Slovakia and Austria, dual citizenship is not permitted. Ethnic Hungarians from these states who acquire Hungarian citizenship would lose their first citizenship if a second citizenship were discovered. (There is an exception for Austrian-Hungarian dual nationals who were refugees in 1956 and whose dual citizenship is specially protected by a treaty, but other Hungarians are not included under this treaty.)
To protect its nationals in the neighboring states, then, the Hungarian government has decided that the non-resident citizenship rolls should remain a state secret. As a result, the associated voter list remains secret as well. But how can a government run a fair election with secret voter rolls?
After opposition protest, the government agreed to allow members of the National Election Commission (including representatives of the parties running national lists) as well as international observers to see the foreign voters’ registration list (Law XXXVI of 2013 on Electoral Procedure, amended by Law LXXXIV of 2013). But the opposition parties and international observers are not permitted to take notes on the list or reproduce it in any way. Given these limitations, however, how anyone apart from the election officials check the list against the voters who actually vote or and how can political parties outside the government locate these voters to send them election materials? One cannot memorize hundreds of thousands of names and their identifying characteristics. So it is not clear if this level of access to the secret voter lists will be enough to ensure a fair vote.
The logistical problems raised by the non-resident voters don’t end there. In particular, there are few checks on either the process of voter registration or on the actual voting so as to ensure that those casting ballots are who they say they are, or that the ballots faithfully reflect what these voters decide. While applying for citizenship requires an appearance at a consulate or embassy, registering to vote does not. In fact, nowhere in the process does any official have to see precisely who it is that is either registering to vote or voting.
Instead, Trianon Hungarians can register on the internet, filling in a form that asks for information that is quite widely known about a person, especially in tightly knit communities. All you need to provide to register are your name, your maiden name (where applicable), the town and district where you were born, either your date of birth OR your personal ID number, and your mother’s name.
How does the National Election Office that registers the applicants know if the person actually named on the form was the person who filled out the registration request? Nowhere in the process is there an official check of identification or even the requirement of a signature, photograph or other validating evidence. (The general problem is captured in that famous cartoon where a dog sits at a computer and says “On the internet, no one can tell if you’re a dog.”) And, as we will see, the information doesn’t even have to strictly match what the Election Office has on file for that person.
Ballots will be sent out to whoever registers in the name of a citizen without any way to definitively tell whether it is the citizen herself who registered or whether the address to which the ballot will be sent is in fact the address of the voter. Given that voting will reveal that one has taken out dual citizenship in some countries where it is illegal, a voter might well want the ballot sent somewhere other than her home address in any event.
In fact, the Trianon Hungarians are the only ones allowed to vote by mail ballot, which longtime elections observers know is always the easiest place for fraud to sneak into an election operation. Hungary plans to use the usual double-envelope safeguard – where a voter fills in an attestation of identity attached to an outer envelope while the ballot itself is sealed in in an anonymous inner envelope that can be separated from this attestation once it is confirmed. So far, so good.
But there is precious little control over the envelopes themselves as they make their way to be counted. Not only does the ballot not have to be actually mailed, but the law permits bundlers to go around collecting ballots and then delivering them en masse to an embassy, consulate or other designated location. There are no checks on what these bundlers do with the ballots in their care and nothing to check whether they in fact they turn in all of the ballots they were given. There is even no way to tell whether bundlers who may well know the personal details of voters are filling in the ballots themselves or changing what they were given. Self-appointed bundlers can show up at any of the designated locations and deliver votes in unlimited numbers.
The number of ballots delivered to or cast at the polling places in the neighboring states must by law be registered each day in the run-up to the election, which means that consulate staff must tally the number of votes each day without anyone present from an election committee to supervise the opening and checking of the ballot boxes. Given how few checks are in place to check potential foul play in the foreign votes (or simply to give assurances that no foul play was attempted), this could be quite serious.
But surely these foreign ballots can’t really influence a national election? In Hungary, perhaps they can. Hungary has about 8 million registered voters, but only 5.1 million voters actually cast ballots in 2010. If most of the 500,000+ new citizens register to vote and actually vote, Trianon Hungarians could account for up to one-tenth of the electorate. These voters can only cast one ballot for the party list and cannot vote in a single-member district, which limits their impact on the overall result. (And it is another site of inequality.) But given that so much of this process of foreign-voter balloting is unverifiable in any rigorous way, even a modest effect on the election casts some doubts on the process.
The fairness of this system for counting foreign votes is made worse when one considers the other group of foreign-based voters who are treated differently from the Trianon Hungarians. Citizens who still have permanent residence in Hungary, but who are living abroad, must cast their vote in a decidedly more onerous way. Let’s call this latter group the Expat Hungarians.
Rather than permit Expat Hungarians to vote by mail, as the Trianon Hungarians are allowed to do, the government has insisted on sticking with the old system in place since 2006 for such voters: they have to vote at embassies or consulates. As a result, Expat Hungarians living or working in the UK, for example, must go to London, no matter where in the UK they live. Ditto with German-based Hungarians who have to travel to Berlin, Dusseldorf, or Munich. Expat Hungarians living in the US must travel to Washington, New York or Los Angeles. How much easier (and less expensive) it would be to vote by mail! But they are not allowed to do so.
Moreover, unlike the Trianon Hungarians, Expat Hungarians are not allowed to vote unless they show up in person and present ID (a passport, for example). Since Trianon Hungarians can vote without ever seeing an election official, no in-person identification is ever required of them. But such identication is required of the Expat Hungarians.
How many citizens are in the Expat Hungarian group? The government says at least 300,000 – but other estimates say as many as 500,000 – Hungarians are living or working outside the country without having given up their official permanent residence in Hungary. This, too, could be a substantial voting bloc, especially as their status gives them the chance to cast two votes just as if they were in the country. (One of those votes goes for the party list and the other for the constituency in which they are still registered.) But they have a much harder time casting their votes because they have to travel, often long distances, to do so.
Not surprisingly, however, the two groups of Hungarians living abroad have different political profiles. Hungarians in the Trianon territories would cast their votes overwhelmingly for Fidesz, if the polls are to be believed. A recent poll said 80% of ethnic Hungarians in Romania, for example, would vote for the governing party.
By contrast, Expat Hungarians are more likely to support the united opposition, or at least so the united opposition believes. While Expat Hungarians are no doubt a diverse group, the people most likely to move are probably the Hungarians who know languages and have networks, which implies that they may be younger and/or better educated. While young people are divided in their political views, the better educated voters are much more likely to vote for the united opposition. Either way, the sheer number of Expat Hungarians and the onerousness of the procedure for voting combine to depress voter turnout, which as we have seen, will benefit Fidesz.
The Hungarian Civil Liberties Union working with Együtt/PM (Together-Dialogue for Hungary, now part of the Unity Alliance) has challenged this disparate treatment of the two groups of foreign voters before the Constitutional Court. But even though the petition was filed in November 2013, the Constitutional Court has not yet decided. (A reminder: The Constitutional Court now has a solid majority since the government was able to name the 8th judge out of 15 in April 2013.) So it appears that the election will go forward with this double standard for Hungarians living abroad.
As the election nears, there are reports of worryingly bad advice for these foreign voters coming from election officials. Consulates in the US were given flyers prepared by local election offices that provided voting instructions for Expat Hungarians in the US. But these flyers specified the wrong election day. While election day in Hungary is 6 April, Hungarian voters in North America have to cast their ballots on 5 April, because of the time difference, in order to meet the deadlines set out in the law. If they followed the instructions they were given by their election office, they would be disqualified from voting.
Expat Hungarians in the UK were sent letters by their local election offices that gave them the wrong location of the London polling station. It turns out that, even though Expat Hungarians are generally supposed to vote at embassies and consulates, in some places (like London) voters actually have to go someplace else. But they were not told the correct location.
The Hungarian Civil Liberties Union has protested these flyers and asked that they be recalled.
The head of the National Election Office admitted that mistakes were made. But she did not apologize. Instead she said, rather oddly that she “simply does not trust some of her colleagues.”
Coming on top of the confusing letter sent by the Election Office to all voters in Hungary about Roma registration, a letter that seemed to imply that all Roma had to register to vote at all, these flyers misinforming US and UK voters about when they need to cast their ballots causes particular concern.
The Election Office website doesn’t even appear to be neutral. On its site, the Election Office features a video from an unclear source, containing much nationalist imagery – and not so coincidentally Prime Minister Viktor Orbán himself. It tells voters that “the nation” (meaning ethnic Hungarians) can vote on 6 April – a thinly veiled appeal to voters who overwhelmingly support the governing party (when they are not supporting Jobbik). Just why the Election Office has such a partisan message on its website has so far not been explained. The link is here to a website less likely to take the video down under criticism so you can see it for yourself.
From anecdotal evidence, the Election Office seemed to be making it easier for Trianon Hungarians to register to vote than for Expat Hungarians to register to vote abroad. Expat Hungarians were reporting that their registration was refused if they missed a diacritical mark, omitted some details of their home address, and failed to match the exact form of their mothers’ name that was in the official register. In fact, the complaints from Expat Hungarians were becoming so numerous that it caused us to go back and look at the law.
And sure enough, right there in paragraphs 84 and 92 of the Electoral Procedure Law (Law XXXVI of 2013), we see the reason. Election officials were explicitly told in this law to ignore typos, spelling mi