The Question
This question was posted by Mikael Fridell, International IDEA

The May 2012 parliamentary election in Greece is proving to be a very interesting event. The current electoral system has a 3% threshold that a political party needs to surpass to be able to win seats in the parliament. Moreover, out of the 300 seats in parliament, the largest party receives a bonus of 50 seats, as a way to enhance the likelihood of a majority government. Two questions related to this:

1. Do you know if there are further regulations in Greece that mitigate this provision of a bonus of 50 seats?
2. Do you know if there any other examples in other countries of this type of bonus system?

Summary of responses
There are no additional provisions in Greece that mitigate the awarding of 50 bonus seats to the largest party.

Italy also has a similar system, with the bonus awarded to the pre-election coalition with the most seats. This is applied on a regional level for senate elections, on a national one for lower house elections, and at the local level for municipal elections.

Although not explicitly bonus systems, there are numerous examples of countries whose electoral rules and regulations are nevertheless skewed to favour larger parties. This is clearly the case in Turkey, where there is a ten per cent threshold. And in Malta – which uses the Single Transferable Vote (STV) electoral system – there is a constitutional provision that states that the party with the most votes must have the majority of parliamentary seats, with bonus seats awarded accordingly if necessary. Changes in Iraq’s electoral law in 2009 and regulations regarding seat allocation also dramatically benefited larger parties.

A small district magnitude and the manner in which remainders are allocated in List PR largest remainder systems may also favour larger parties.

While there are very few contemporary examples of explicit bonus systems, there are a number of past cases. Mussolini, for example, introduced measures in 1922 to guarantee a majority to the largest party. More recently, a 30 per cent bonus was introduced in French Polynesia in 2004 (and since abolished). In Bangladesh, it used to be the case that the reserved parliamentary seats for women would be awarded as a bonus to the party that commanded a majority following an election.

Examples of related ACE Articles and Resources
Encyclopaedia:
• District magnitude 
• The threshold 

External Resources
• Wikipedia (Italian) Premio di maggioranza 
•  State, Society and Governance in Melanesia: Political Consequences of Pacific Island Electoral Laws, Jon Fraenkel

Names of contributors
1. David Jandura
2. Jack Santucci
3. Michael S.R. Kinnear
4. Stina Larserud
5. Andrew Ellis
6. Jon Fraenkel
7. Velko Miloev

I'm unaware of any similar system used elsewhere although there are plenty of cases of laws that make an ostensibly proportional system perform very unproportionally (See Hungary). 

In addition to the bonus 50 seats (up 40 from the last election) Greece awards its regular seats among 56 constituencies. That creates a district magnitude of only about 4.5, which doesn't create the most proportional seat allocation. The 50 bonus seats, however, are calculated based on the national vote total.

Small degrees of over-representation are consistent with most nationwide PR systems but Greece's tends to highly exaggerate the top party, while mitigating the natural bonus that we would see for the other parties. Invalid votes are disregarded for seat allocation purposes as are votes for parties that fall below the 3 percent threshold.  Therefore, we should normally expect to see most parties that enter parliament get a slightly higher percentage of seats than their raw vote total would suggest.  For example, in this election, it seems all parties that entered parliament would have been over represented if you did not factor in the bonus seats (so a parliament of 250).  Syriza would have received nearly 21 percent of seats for getting only 17 percent of the vote and the ND would have still received 23 percent for getting only 19 percent of the vote.  

With the bonus, however, the ND got a 17 percent boost (36 percent of all seats with only 19 percent of the popular vote), while Syriza still got a one percent boost. The other parties, however, got around one percent less of the seat total than their popular vote would have indicated. 

Italy uses a similar system. A key difference is that the guaranteed majority goes to the pre-election coalition with the most seats. This allocation is regional for Senate elections and national for Chamber ones. If you want to read or translate some Italian, you can construct some historical case-study material from the links in <a href="http://it.wikipedia.org/wiki/Premio_di_maggioranza">here</a>.

I believe that Mussolini introduced a system which would guarantee a majority to the largest party, in the elections held shortly after his "March on Rome" in October, 1922.

Greece and Italy are the most straight forward List PR bonus systems, but it’s important to mention that several rules that are not explicitly phrased as bonuses also advantage large parties/plurality winners. Leaving all the plurality/majority systems (and even the parallel ones) out there aside, rules relating to legal thresholds (as alluded to by David) – which are so popular in Europe – obviously put small parties at a disadvantage. Turkey with a 10% threshold being the most striking example. Small districts (again, as mentioned by David) can also lead to disproportionate results. For example, even though Chile is often referred to as a plurality/majority system, it really is designed as a List PR (D’Hondt) system using multi-member districts that are as small as they can possibly get, returning only 2 representatives from each – leading to a very high effective threshold for representation. In countries with List PR largest remainder systems it is also possible to favour larger parties by only letting remainders be allocated to the parties that have received a full quota in the first round of seat allocation (though this is likely to have significantly smaller effects than what we see in Greece). Another interesting set of rules relating to bonus seats/disproportionality is that of Cameroon (leaving aside the political/electoral realities of the country). Cameroon elects some members through FPTP, and others from lists in multi-member districts. In the MMDs, if a list wins more than 50% of the votes, the entire list is elected (one party gets all the seats). If no absolute majority of votes is gained by any list, the top one still receives half the seats, and the other seats are allocated proportionately to the other lists (excluding those falling below 5%).

'Mitigating provisions' do not exist in Greece: a look back at previous Greek election results shows both Pasok and ND benefiting from the 50 seat bonus to build parliamentary majority.

In addition to the cases already quoted, Malta, which uses STV, is an example where the bonus provision can and does change control of parliament. Malta has  two very different parties which win most of the vote and have roughly equal support (incidentally also conditions which contribute to very high turnout).  In the last election in 2008, the Labour Party polled 48.8% and won 34 seats: the Nationalist Party polled 49.3% and won 31 seats, because the transferred votes of the minor parties went more heavily to Labour.  However, the constitution contains a provision that the party with the plurality of votes must have the majority of seats: the Nationalists therefore got an extra 4 top-up seats to give them a parliamentary majority.  While it can be argued that this rather defeats the purpose of STV, I'm not aware of any challenge to the legitimacy of this provision in Maltese political debate.

I believe that the Italian provisions apply to municipal elections as well and have made a significant difference on occasion to control of local authorities.

One other historical example: I recall that the reserved seats for women in the Parliament of Bangladesh used to be a bonus for the party that could command a majority at the election.  However, they are now allocated to parties proportionately to their vote in the election, and are thus no longer a 'winner's bonus'.

Just another example of deliberately creating hidden bonus seats – and an illustration of something Stina mentioned already.
For the 2005 Council of Representatives (CoR) elections Iraq used simple quota and largest remainders applied at electoral district level, followed by the distribution of a large number of compensatory seats - 45 out of the then 275 total for the CoR.
The amendments to the Election law from December 2009 and the new IHEC Regulation on Seat Allocation changed the system dramatically in favor of the larger parties or coalitions. In the 2010 CoR election a political entity, which did not obtain the electoral divider (quota) in a district, was excluded from the race, i.e. largest remainders were not applied. Also the reduced (seven) national “compensatory” mandates did not anymore serve their alleged purpose (to compensate for disproportionalities  created in the seat allocation on district level), but became just an additional bonus for the so called winning lists. These
“compensatory” seats were distributed to political entities in proportion to their seats already won in the governorates.
One problem with such technical or mathematical forms of electoral engineering is that they remain hidden, since they cannot be easily understood by many
stakeholders – common voters, media, smaller parties or individual candidates.