Mike Williams: You can't predict the result of any election

It's far too early to predict who will win the election. Photo / File

The Covid-19 pandemic, though uniquely contained in this country, is still raging in the normally well governed and well organised United States and will undoubtedly impact the general election due in that country later this year.

The Economist magazine has developed a formula involving polling and economic factors which predicts that the incumbent president Donald Trump will be heavily defeated by his opponent, Democrat Joe Biden in the November presidential election.

Trump appears to be already lining up excuses for defeat, the most outrageous of which involves challenging the validity of a postal voting option which appeals to many of the states as a way of avoiding the Covid prone queues which are typical of American election days.

Though many will be delighted to these predictions, it is my view that anyone who thinks that they can tell, months in advance, the outcome of an election in the current world political atmosphere is indulging in fantasy.

I well recall a university academic, Claire Robinson, who in these very pages, predicted a National Party win in the 2017 general election a year in advance on the basis of a significant poll lead which went up in smoke with the promotion of Jacinda Ardern to the leadership of the Labour Party within weeks of that election.

This week's TVNZ Colmar Brunton poll amply demonstrated the volatility of the New Zealand electorate which makes any election predictions likely to be unwise.

With National gaining 9 per cent to register a score of 38 per cent and Labour losing the same quantum to tally 50 per cent of the vote, we are led to the conclusion that a touch less than a fifth of the voters changed voting intention in just a month.

A core reality in politics these days is that traditional levels of party loyalty are much eroded to the point that a clear majority of voters have no fixed party preference and are up for grabs at any given general election.

For about 70 years pundits could rely on the National and Labour parties each getting around 40 per cent of the vote, but this has all changed under the MMP voting system that was introduced in 1996 and both major parties have plumbed new depths in their party vote scores in the last couple of decades.

In the 2002 general election National got just under 21 per cent of the party vote and in 2014 Labour scored only 25 per cent of that vote.

It is this volatility which trips up pundits seeking to make predictions, and the same phenomenon which, in many ways, dictates how campaigns are planned and run in the 21st century.

Having run a dozen or more election campaigns over nearly half a century, my advice to campaign managers is always the same – organise your activity as though you are going to win by just a handful of votes.

Mike Williams

Burned into my memory is the outcome in the Taupo electorate in the 1981 general election when I was a Labour Party organiser. In this election Bill Rowling's Labour Party got more votes than National - led by Robert Muldoon - but close results in three seats cost Labour the election.

I was sent to Taupo, lost by Labour by just 36 votes, in preparation for an electoral petition which ultimately confirmed the National candidate's wafer-thin majority.

I discovered that dozens of votes from the solidly Labour voting town of Tokoroa had been disallowed because the people who had attempted to vote were not on the electoral roll. Had the Taupo Labour Party bothered to door-knock a half a dozen streets in that town, the seat would have been retained.

There were similar stories from the Eden and Helensville electorates where National candidates held on by 117 and 210 votes.

This was an unusual outcome for a first-past-the-post-election.

It has become the rule under MMP.

The 2017 poll was decided when National lost a list seat to Labour on the special votes and second to the Green Party.

Had the election night result survived the final count, New Zealand First would have been forced to back National.

I inadvertently reversed a couple of figures in my most recent column and therefore heavily understated the reduction in prisoner numbers achieved by the current coalition Government. The prison population peaked at 10,820 (not 10,280 as I previously reported) and the muster on June 12 stood at 9481, a reduction of 1339 from the peak.

This is a yearly savings to the taxpayer of more than $160 million and means that our incarceration rate has dropped from 216 per 100,000 of population to 191.

This number is still shamefully high when we put New Zealand alongside countries that we normally like to compare ourselves with, but it is at least heading in the right direction.

Mike Williams grew up in Hawke's Bay. He is CEO of the NZ Howard League and a former Labour Party president.