Parliamentary fragmentation in PNG at the elections: is it getting worse?

Political Parties in Papua New Guinea

Parliamentary fragmentation in PNG: is it getting worse?

By Maholopa Laveil
19 March 2021

Papua New Guinea has many parties in parliament. PNG’s average of 19.6 parties elected at each election between 1997 and 2017 is higher than many other countries with multi-party systems. (For example New Zealand has had on average five parties in its parliament in the same period.)

When discussing parties in PNG’s parliament, it is important to realise that some MPs are elected unaffiliated to any party (so-called independents). The share of MPs elected as political party members was at its lowest in the 1990s, but has been 80 percent and above since 2002.

Parliament share of MPs with party affiliations 1977 – 2017Parliament share of MPs with party affiliations 1977 - 2017

As the chart below shows, the number of parties in parliament was 7 in 1977, increasing to 10 in 1987. It stayed at about this level up to 1997, then grew rapidly to 24 in 2002, and has since fallen to 21 in the last two elections.

Twenty-one parties is a very high number, but the picture changes somewhat when party size is considered. Also shown below is the effective number of parties (ENP), a measure of the fragmentation of parliament. The ENP measure was first introduced by Markku Laakso and Rein Taagepera in 1979, and weighs the number of parties according to total number of seats won, thus giving a higher weight to larger parties. As can be seen, the effective number of parties increased from 1982 to 2002, but has fallen since. PNG’s ENP is, however, much lower than the total number of parties. This suggests that, although many parties are elected to parliament in PNG, only a smaller group have had substantial numbers of their MPs elected at each election.

Even though a smaller group of parties has substantial numbers, PNG’s ENP of 6.9 in 2017 is still high by global standards. According to a 2019 study of over 140 countries (not including PNG), only five other countries had a higher ENP between 2017 and 2019: Belgium (9.7), Bosnia and Herzegovina (8.7), Brazil (16.5), Indonesia (7.5), and the Netherlands (8.1).

We can also look at the number of parties that have had more than three MPs elected since 1977. This number follows the ENP closely, except in 2002 and 2007. 19 different parties have had more than three MPs elected in at least one election since 1977. However, only two, PANGU and People’s Progress Party (PPP), have existed since independence and had more than three MPs elected in each election. The PANGU party has been the most successful, averaging 17.7 wins in every election since 1977 (though it only won one in 2012). The PPP has averaged 9.2 wins over the same period.

Parties in parliament 1977 – 2017Parties in parliament 1977 - 2017

When governments are formed in PNG, they usually have large established parties at their core. The table below shows which parties have held the important positions of prime minister (PM) and deputy prime minister (DPM) and their relative sizes, both immediately after elections. Except for 1987, 1992 and 1997, the ruling party (PM’s party) has always been the largest in parliament. The PM’s and DPM’s parties have never on their own commanded a majority in parliament. Governing coalitions have required more parties.

Parties (share in parliament) in coalition 1977 – 2017Parliamentary fragmentation in PNG: is it getting worse?

The average number of years that any PM, after the election, has been able to serve is 3.2 years. Is there a relationship between government tenure and how large the PM’s party is, or between the number of effective parties?

It seems reasonable to assume, first, that governing coalitions with larger parties at their heart will serve longer; and, second, that more fractured parliaments will be more unstable.

The charts below show the relationships of interest.

Years in government, effective parties, and ruling party’s share in parliament 1977 – 2017Years in government, effective parties, and ruling party’s share in parliament 1977 – 2017

Surprisingly, there is almost no relationship between the share of seats in parliament held by the PM’s party and the length of the government’s tenure (r = 0.14). There is a much clearer relationship between how fragmented parliament is and length of government tenure (r = 0.62). However, the relationship is the opposite to expected: when parliament is more fragmented, governments have been more likely to stay in power. This may simply be a product of chance and two unusually long-lived recent governments. But even if these governments are ignored, there is no sign of the expected negative correlation. More fragmented parliaments do not lead to shorter government tenure.

In summary, although PNG’s parliament is made up of many parties, a smaller number of large parties form the centre of political coalitions. Yet surprisingly, greater fragmentation following the elections does not seem to lead to increased political instability.

Notes: The author would like to thank Dr Terence Wood for insights and comments on earlier drafts. Party data were taken from a number of sources: H. Okole (2005) ‘The fluid party system in Papua New Guinea’; R.J May (2008) ‘Political parties in Papua New Guinea’; M. Rooney (2019) ‘Political developments in Papua New Guinea in a historical context’; PNG Elections Database; and Wikipedia ‘Elections in Papua New Guinea’. Effective parties were calculated using the effective parties index by Laakso and Taagepera (1979) ‘Effective number of parties – measure with application to West Europe’, and using number of party wins at elections, excluding seats for which election results were yet to be declared. This paper assesses party membership at the time of the election, counting only elected MPs who campaigned as a party member.  After elections, some independents join parties, and during the parliamentary term, a number of MPs switch sides. These are matters for further research. 


This research was undertaken with the support of the ANU-UPNG Partnership, an initiative of the PNG-Australia Partnership. The views represent those of the author only.

About the author/s

Maholopa Laveil
Maholopa Laveil is the FDC Pacific Fellow at the Lowy Institute, seconded from the University of Papua New Guinea.

Page 1 of 1