KEY POINTS:
Winston Peters' value to both Labour and National has become abundantly clear. Both parties are pulling their punches over the donations allegations for fear of alienating him as an ally or future ally.
MMP creates hostage situations. Remember Alamein Kopu and her pull over Jenny Shipley?
In this most recent row Sir Robert Jones has unexpectedly been firing most of the bullets at Peters. He probably summed up the view of many when he said, "I belong to a different era. I don't like it now under MMP."
John Key has said National will, if elected, hold a referendum into MMP.
Key's referendum will first ask voters: are you satisfied with MMP? If the majority says no, then a second referendum will be held pitting MMP against some other unspecified alternative.
But is this what we need? MMP was itself born out of a referendum, and voter frustration at the unbridled power of first-past-the-post governments. First Sir Robert Muldoon, then Sir Roger Douglas proved if you could control the Cabinet you could control the country.
But one wage freeze and an unadvertised rapid economic transformation later, voters realised they wanted their leaders on a tighter leash. They wanted them to have to work harder, and more consensually, to get their own way. Which is what MMP delivers with its minority or coalition governments, its requirements to consult and its generally slower pace of change.
Plus, of course, for anyone younger than 32, two-tick voting is voting.
So why would we ditch it? Because MMP has flaws which undermine the legitimacy of our parliamentary system.
For instance, under MMP, Gordon Copeland, a United Future List MP, can abandon his party, switch voting allegiances and still stay on as an MP. Copeland has no mandate to be in Parliament, yet the rules allow him to continue as if he did.
Electorate MPs like Labour's Rick Barker can be voted out at home and yet still make it back on the party lists. In Barker's case he even made it to Cabinet.
If that is justified by the sanctity of the party list, then what about Mike Ward and Catherine Delahunty? Both Greens and both higher placed on the list than Russel Norman, yet both pushed inelegantly aside when Nandor Tanczos's early retirement offered the co-leader the chance to get to Parliament in time for some pre-campaign publicity.
All these inconsistencies create unfairness, though not so much as the threshold rule itself.
Under MMP a party must win 5 per cent of the party vote or an electorate seat. A win in an electorate, where the party scores lower than 5 per cent, still gets a proportionate top-up. So Rodney Hide's win in Epsom gave Act two MPs even though the party won only 1.5 per cent of the party vote.
By comparison, in 1996, the Christian Coalition won 4.33 per cent of the party vote, a hair's breadth from the magic threshold. But it failed to win in any electorate - so bad luck, no MPs.
There are two issues. Firstly, is the 5 per cent threshold too high? The commission that recommended MMP preferred 4 per cent, but the two major parties argued for a higher threshold. Those fears have proved unfounded. In fact, as much as MMP has delivered a more diverse Parliament, only one new party (Act) has broken in since the switch to MMP. The others have all been created around a sitting member.
But is the electorate threshold too low? In Germany, a party must win three electorates before qualifying for list seats. Adopting a three-electorates or 5 per cent criterion at the 2005 election would have seen five parties able to get in list MPs.
United Future and Act would have been restricted to Peter Dunne and Rodney Hide. As Jim Anderton couldn't bring in a list MP under current arrangements, the Progressives would have been unaffected. Since none of those three parties attracted more than 2.6 per cent of the party vote, is that an unfair result?
And then there is the Maori vote. Last election, the Maori Party won 2.12 per cent of the party vote and four electorates, hence it has four MPs. This coming election it may win more electorates even though polling indicates its party vote will be no higher.
Since the number of Maori seats grows in accordance with the number on the Maori roll, it is entirely possible that over time this disparity between the number of MPs elected and the party's proportion of the party vote will grow. That will mean a larger and larger over-hang and the leading party will need to garner not 61 votes to govern, but 63, 64, 65. Is this what we want?
These are all valid issues in need of debate. But they do not fit the yes-no format of a referendum. Nor do they provide evidence that MMP itself is beyond repair.
What they point to is the need for a considered review of the electoral system. Learning the lessons of the Electoral Finance Act, this should be conducted in a non-partisan way with a clearly stated purpose of seeking greater fairness.
In the spirit of fairness, perhaps such a review should also look at the Prime Minister's prerogative to set the election date. Or the length of the political term; four years might be more productive.
The problem is that these changes require MPs to vote against their own interest. History tells us MPs don't do that. Which is why a simplistic question in a referendum is so appealing. It looks as if something substantive is being done, even if it isn't.
But concerns about MMP's peculiarities are genuine and a more considered review would be more constructive.
* Andy Nicholls is a partner in Chapman Tripp and a specialist in public law.
