Dr John Parkinson, Associate Professor of Public Policy, University of Warwick, UK

The 5% threshold should be maintained. Comparative experience with PR systems in other countries suggests that a lower threshold introduces more instability into parliament, more problems of the 'tail wagging the dog'; but higher thresholds act as a significant deterent to political engagement and, thus, legitimacy of the system. The current threshold was chosen for good reasons, and should be maintained.

The purpose of the electorate seat threshold is to ensure that interests and perspectives that are focused on a particular location get represented in parliament. This is a principle I strongly support, especially in a country such as New Zealand where the population is concentrated in relatively few centres. Countries such as The Netherlands can afford to have little by way of locality-based representation because their population is fairly dense everywhere. That is not the case in New Zealand, where a quarter of the population is concentrated in one city, and population density is very low the further south, and north, one moves. Furthermore, New Zealand has relatively large electorates - and low representative-to-population ratios - by international standards. This means that single electorates already represent quite widespread and diverse populations. Dilution of the one-electorate rule would, therefore, largely remove the value of this particular threshold. It would remove the "local interests" logic of the rule. I therefore support the present system's continuation.

No, there should not, for the reasons given above.

I do not support such a change. Party lists are determined by parties, and it is up to party MEMBERS to determine their lists and rankings which are then offered to the general public. The issue here, really, is about party membership and party decision making processes, and their relative transparency and accountability. One way to address that is through primary elections, but that is outside the scope of the review.

Nothing, because the possible solutions are worse than tolerating the rather minor inconvenience of overhang. Solutions might include docking the party electorate seats, but that would rob a local electorate of their local voice, which cannot be justified. They might include allowing more supplementary list members from other parties to compensate, but that would introduce potentially dozens more MPs into the house, for which there is no public appetite, and no available seating. The problem of overhang really is very minor compared with these.

I do not consider this to be a problem at all. As already noted, New Zealand electorates are already very large by international standards - our proportion of elected politicians per voting age population is one of the lowest in the world - which makes the working life of an electorate MP more difficult, and reduces the direct connection between MPs and voters, lowering voters' perceptions of MPs quality, trustworthiness and legitimacy. Lowering the number of electorate MPs would further diminish these important functions; raising the number would impact adversely on the system's proportionality, which should be protected above all other considerations (see "other issues" below). For my part, I would recommend a dramatic increase in the number of MPs overall, but again, this is not in the review's remit, and there is no public appetite for such a move.