JACK VOWLES’ BLOG

 
 

February 21 2017

Willie Jackson and the Electoral Rolls

Aspiring Labour politician Willie Jackson blames the Electoral Commission (EC) for low turnout and low electoral enrollment particularly among the young, Māori, those on low incomes and who do not own their homes. The headline of the story says that the EC is ‘failing’ voters. Willie Jackson is right to identify the problem, although he exaggerates it. He is wrong in assigning blame and in the extent of his proposals for reform.

Jackson claims those who rent their homes are not voting. But many still are, and the Labour Party in particular works hard to enroll them and persuade them to vote, and Jackson himself pledges to help in that process. The housing problem is certainly relevant: many people who rent move around in conditions of housing scarcity and it is hard to keep track of them. The Electoral Commission does its best with the resources available. In 2014 it organized a large conference at Parliament, bringing in international and local experts to discuss strategies to address the problem with those working in the community to promote participation. After the 2011 election in particular the Electoral Commission has intensified its efforts to enhance participation, and has extended its research to better understand the problem.

In theory, enrolment to vote is compulsory in New Zealand: in practice, it is not enforced because this is deemed impractical and perhaps counter-productive. Jackson pays particular attention to enrolment. In 2014, just under 8 per cent of the eligible population were not on the rolls. The bigger problem is that over one in five of those who are on the rolls failed to vote in the last two elections. 

Focusing on the somewhat lesser problem of enrolment, Jackson wants to give everyone who enrolls the right to a higher level of privacy than now. This is not something that the EC can simply alter: the nature of the electoral roll is defined in law, so Parliament would have to make the change. Currently, the names, addresses, and occupations on those on the rolls are available for public purchase and inspection, in the latter case usually in public libraries and electoral offices. Each electorate has its own roll: there is no published version of the roll that combines the names on all of them together. Jackson cites fears among those not enrolled that they could be hunted down by state agencies and debt collectors, and that victims of domestic violence could also be identified by their abusers. In the latter case, there is already a privacy option: those with legitimate concerns can apply for anonymity and will be granted it. 

In theory, use of the electoral rolls is restricted by law. It is not permitted to scan them into a database and use them for any commercial or business purpose, including debt collection. They may be used for social science and health research, but under strict controls. But Jackson has a point: all rolls are available for purchase, and their use is not well monitored. State agencies may be using them to find people, although in many cases for good reasons. The Electoral Commission and the Privacy Commission have recommended to the Government that access to the rolls be reduced, bringing their sale to an end, and providing public access only under controlled conditions. This will require change to the law, and the current Government has put off engaging in that process until after the next election.

Names and addresses of those enrolled are publicly available for very good reasons: to provide a transparent foundation for the detection of possible electoral fraud, open to all citizens to inspect. Electoral fraud is rare in most developed democracies, and in New Zealand rarer still. Openness is one of the reasons for this. On occasion, a close election has led to an electoral petition in which the rights to vote of people who have cast ballots may be reviewed. A few votes are usually disallowed on inspection, and this can change an outcome. It is important that this process is transparent to the public. Public access to names and addresses therefore underpins confidence in the fairness of New Zealand elections. But more controls are needed to ensure that the rolls are not used for inappropriate purposes.

To improve the coverage of the rolls, another important reform would be the adoption of automatic voter enrolment, recently adopted in several states in the United States: the use of government records to identify people not on the roll who can nonetheless be identified as living at a specific address. In 2014 nearly 30,000 people voted but their votes were not counted because they were not on the rolls. Almost certainly under automatic registration many of them could have been counted. Automatic registration would permit counting of their party votes at least: people enrolled under this process would have to be contacted to choose the Maori or general rolls if eligible to do so, and would also have to be informed of their right to be entered on the confidential roll.

When Willie Jackson encounters people who do not want to be on the roll because they fear it will be used against them, he should at least inform them of the existence of the confidential roll. As for those who are seeking to evade collection agents, he should refer them to one of the various organisations that exist to help people whose debts are out of control. If he becomes an MP again, I am sure that the Electoral Commission would appreciate his support in changing the law to enhance the protection of privacy, while maintaining public access to the rolls for the appropriate purposes.


October 17 2016

There is no Hypocrisy in Mt. Roskill

Comparing the Act-National deal in the Epsom electorate with the Green party’s announcement it will not stand a candidate in the Mt. Roskill by-election is just plain silly. Those who are labeling it as hypocrisy are either consciously twisting facts to make a political point or just not thinking straight.

In the Epsom electorate, National encourages its supporters to vote for Act in the electorate vote, while keeping its own candidate in the race. This means Act can cross the threshold for representation in Parliament, perhaps bringing in more MPs, even though it gets less than 5 per cent in the party vote. Meanwhile the National candidate can still promote the National party vote. 

Bringing in more MPs on the back of an electorate win in this way has been given the label of ‘coat tailing’. Labour and the Greens are critical of this strategy. In Coromandel in 1999, the two parties apparently played the same game, getting Jeannette Fitzsimmons elected as local MP. But Jeannette Fitzsimmons has reminded us that there was no direct collusion: Labour leader Helen Clark just signaled that she would not mind Labour party voters casting an electorate vote for the Greens in Coromandel. There was no discussion between the Green and Labour leaderships about this, and therefore no ‘deal’ as such (Listener December 3 2016). Labour and the Greens now say they want to get rid of this option by changing electoral law to prevent it. 

What has this to do with the Mt. Roskill by-election? There is no coat-tailing payoff for anyone. The Green Party has no chance of winning Mt. Roskill. The Greens had no chance of winning in the Northland by-election, and didn’t stand a candidate there either. 

Parties not standing candidates in electorate seats because they know they can't win is standard practice, even in general elections. National doesn't stand candidates in the Maori electorates, presumably helping the Maori Party get a few more votes. In 2014 ACT didn’t stand candidates in many electorate seats, particularly marginal ones where National and Labour were close, again to National’s probable advantage. No one criticised then for that. It doesn’t merit criticism. 

The Mt. Roskill by-election and the Epsom general election arrangements between the parties in question are different, not the same. There is no hypocrisy.

NOTE: This post has been amended to include Jeannette Fitzsimmons’ clarification of what happened in Coromandel in 1999.


September 15 2016

Let’s Talk About Polls…

There is a duel going on over whether the New Zealand Labour Party can currently expect to get the votes of 26 per cent of the electorate or somewhat more, perhaps over 30 per cent. Which of the two polls is a ‘rogue’?  Journalist Vernon Small thinks that the margin of error is supposed to include such rogues. On this, of course, he is wrong: rogue polls are those that go outside that range.  They are the 1 out of 20 possible polls that go outside the margin. Either of the polls could be one. 

But it may be a good idea to take a step back.  The margin of error actually means that 19 out of 20 possible polls will record a number that is within plus or minus the figure in question. But there is not just one margin of error for any poll – the margin varies according to the numbers themselves. In a sample of 1000, at 48 per cent the margin of error will be plus or minus 3.1 per cent.  The poll is actually telling us that 48 per cent actually means somewhere between 45 and 51 per cent.  For 26 per cent, the margin of error is plus or minus 2.7 per cent: so 26 per cent means somewhere between 23.3 per cent and 28.7 per cent.  For a party reported to be on 5 per cent, the margin of error is only 1.4 per cent: so the party in question is somewhere between 3.6 and 6.4 per cent.  When newspapers or television news programmes report changes between one poll and the next that fall within the margin of error, they are misleading you: no change is the safest conclusion.  But that is not news. 

It gets worse.  Polls conducted for newspapers and television are done quickly. Robust polls should take longer. Responses rates have declined dramatically. In the UK, telephone polls get less than a 10 per cent response rate and most pollsters are going online.  But online respondents are not randomly selected, as people who answer surveys should be.  In fact, calculation of margins of error is based on the assumption than response rates are 100 per cent and every person in the population has an equal chance of being sampled. Instead, pollsters massage their data to try to make it representative. So margins of error are themselves questionable guidelines for the range of certainty. More often than not, pollsters succeed in getting their numbers more or less right: as far as one can tell.  But poll failure is increasing elsewhere.  New Zealand’s population may be too small for effective online panels, as they require very large numbers for purposes of selection and weighting. Those engaged in polling day-to-day will have a better sense of the current state of play in New Zealand. But poll consumers should under no illusions: buyer beware!

June 26 2016

The Worst Way of Running a Referendum

A booster to democracy’s critics: that was the tenor of a remark made to me by a colleague as the result of the UK’s EU referendum became apparent as the votes mounted. Only to direct democracy, I replied. But there is more to it than that. I have always been sceptical about referendum democracy: on the other hand, I have supported the principles and practice of referendums on constitutional issues where elites in power in a representative democracy have an obvious conflict of interest. On the face of it, this is a justification for the two referendums held in the UK on EU membership. However, membership of the EU is a Treaty matter, albeit with constitutional implications. The self-interest of elites in either outcome is not very obvious, although no doubt a story could be told. But there is no point in debating the wisdom of a referendum on such a matter: one has been held, with momentous results.

The UK has had very few national referendums. It has devised perhaps the worst possible method of running them. Extensive state funding is given to groups contesting the question. Virtually nothing is provided to inject balance and unbiased information into the debate. The campaign groups are free to disseminate misinformation and outright lies. There is nothing to stop or even dissuade them. As a result the debate becomes mired in accusations of dishonesty that if anything compounds its influence. 

There is a cynical response to such criticism of UK referendum practice. There is no such thing as balance and absence of bias in such debates. Any attempt to provide ‘objective information’ authoritatively would be shot down by the tabloid press. 

There is no doubt that introduction of such a process would be resisted and aggressively attacked. For that reason its foundation would need to be robust and enforceable in law, to the point of requiring media to publish retractions in places as prominent as the claims themselves. 

A Panel of highly respected citizens with significant expertise and no prior record of expressing a position on the matter would need to be appointed, and the persons speaking on its behalf would have to be excellent communicators prepared to stand up to attack. The mass media would have to be required to give them space. The Panel would be sufficiently well resourced to be able to provide background material that could provide sources for balanced and informed commentary by others. While many statements with which one could take issue in referendums are opinion, others can be clearly demonstrated as being untrue. A Panel should therefore pick its battles carefully, targeting only the most obvious untruths. 

It would be equally important that no state funding be provided for campaign groups except access to some broadcasting time. Reasonable but tight expenditure limits should apply, including funds used for public opinion research. The voice of the Panel would therefore stand out, perhaps as clearly as the voices of the campaigners.

Of course, no one could guarantee that referendums run within this kind of framework would produce different results than those under the UK model. But the conduct of those involved would have to change in ways that would enhance the quality of the debate. There is experience of regulatory regimes for referendums elsewhere that successfully approximate something like this model. I doubt very much that the sclerotic processes of normal British politics will make a space for serious consideration of reform, even after such a major shock. But someone has to make the argument, even if from the other side of the world.

May 31 2015

John Armstrong on Green ‘toxicity’

John Armstrong (Herald, May 30 2015) wrote:

'The big worry for the Greens is that the last election result amounted to a rejection of them, suggesting that getting roughly 10 per cent of the vote is the best they can do and that 90 per cent of the electorate regards them as toxic or mad or both'.

I wonder on what basis of knowledge he ‘suggested’ this. 

The 2014 New Zealand Election Study asked people the extent to which they liked or disliked parties on an 11-point scale.  
In order of ‘toxicity’, the parties from most disliked/most toxic (from points 0-4 on the scale) were: Internet (73% disliked it), Mana (62%), Act (53%), conservative (49%), United Future (45%), New Zealand First (42%), Maori, Green, Labour (all 39%), National (27%).

Labour and the Greens were almost equally 'liked', by about 35 and 33 per cent respectively. National were liked by about 56 per cent.  Only 15 per cent are neutral or indifferent to National, about 28 per cent to the Greens and 26 per cent to Labour. None of this adds up to a 90 per cent toxicity rating for the Greens: Greens and Labour are in much the same position. This does show that National has a considerable ‘pull’ factor and, until that ebbs, all the Opposition parties are up against it.

A  comprehensive set of tables from the 2014 NZES can be found elsewhere on this website.

May 27 2015

Neo-liberalism in New Zealand: Half Full or Half Empty?

Chris Trotter attacked Matthew Hooton for declaring ‘the end of neo-liberalism’ in New Zealand after the 2015 Budget. Chris asserted that a neo-liberal settlement has been in place for the last thirty years and that neo-liberalism has been progressively advancing under successive governments, both National and Labour.

He provided a list of what he sees as defining features of neo-liberalism. Slightly reorganised, they are:

Price stability and the independence of the Reserve Bank;
Labour market flexibility and the downsizing of trade unions;
An open competitive economy;
A broad-based, low-tax structure: the ‘steady elimination of progressive taxation’;
Government surpluses and debt repayment;
Continued public sector shrinkage and privatisation;
Creating incentives for beneficiaries to move off benefits and into training and/or work.

However, we need a definition of neo-liberalism, not just a list of features, many of which, listed above, are not necessarily neo-liberal at all.

Neo-liberalism is an ideological extrapolation of neo-classical economics, with a bit of the post-war Austrian school thrown in here and there. It stands for market fundamentalism. While most economists take the assumptions behind much of their work as theoretically useful abstractions not necessarily to be found or achieved in the real world, neo-liberals do have a conviction, close to a religion, that markets do everything better, and if they fail it is the fault of state intervention. In New Zealand, most neo-liberals are in the Act party, now consistently polling less than one per cent of the electorate. 

This ideological position is not to be confused with that of market pragmatists who acknowledge, as Marx did, that markets are very good at generating economic growth when properly regulated and, in which context, corrections for market failure can be applied through judicious public policy. Well before neo-liberalism became influential, social democrats between the 1930s and 1960s understood the power of markets as well as their limitations.

Neo-liberals got their chance in the 1980s to apply market fundamentalism and, into the 1990s, those principles were increasingly applied in the liberalization of domestic and international capital markets. There was indeed a shift to the right in economic and social policy, and some aspects of that rightward do shift continue into the present.

Neo-liberals were successful because the world was changing and some of their policy prescriptions addressed the changes. They consequently attracted the support of pragmatists. The GST, for example, part of the process of flattening the tax structure, is an excellent means of gathering government revenue and corrected an excessive dependence on income tax. Trade liberalization, of course, is an objective promoted not only by neo-liberal but also Keynesian economists, a goal strongly established after World War Two at Bretton Woods. As a nation highly dependent on exports, when New Zealand began losing its guaranteed market in Britain in the 1970s, a failure to develop a more competitive economy would have been disastrous. 

Chris assumed that the reduction of government debt and a return to government surplus are features of neo-liberalism.  But well before neo-liberalism became fashionable, social democratic governments were often fiscally conservative, and for good reasons. Government debt is a constraint on policy independence, as Walter Nash found in 1938 when he negotiated to extend New Zealand’s debt payments to financiers in the City of London to help pay for the beginnings of the welfare state. Government debt enhances the risks associated with external policy shocks such as the global financial crisis. Michael Cullen’s responsible handling of the economy under Labour from 1999 to 2008 provided the means for National to meet that crisis with a classic Keynesian fiscal stimulus, one of the highest in the developed world in relation to the size of our economy. So much for neo-liberalism in this particular case, although one only needs to follow events in the UK to see what happens under that course of action.

One of the features of the neo-liberal turn was a greater priority on controlling inflation than on reducing unemployment. But of course high levels of inflation are not particularly desirable either. Chris complains about government policies to provide incentives for beneficiaries to move into work. Despite neo-liberalism, there is a continued policy focus on reducing unemployment. While this approach is much criticised on the left, all the evidence suggests that except for a small minority, people are better off in work than on benefits. Nor does one encounter much New Zealand evidence that these incentives are being applied in draconian ways as, for example, in the UK. Advanced welfare states like those in Scandinavia have traditionally offered more generous benefits, but they are policed very strongly. Such tight controls on eligibility are likely to generate more confidence among the public that beneficiaries truly do need support, and have probably made it easier for the Key government to increase benefits without significant political costs. 

There is more that could be discussed but this is already too long. Privatisation has not been progressive. We all know of the significant reversals under Labour. The Key government has just put in another big funding boost to Kiwirail, It continues to prop up Solid Energy, not necessarily a good idea. The partial privatisation of the three energy companies was a burst of ideology and a grab for cash, but the government is aware that New Zealand voters have little taste for any more of this. Neo-liberals do not approve of government subsidies to business, but we have the Sky City Convention-Casino deal, questionable arrangements with Saudi sheep-breeders, and the pay-out to keep the Tiwai Point smelter going, a remaining feature of the old industrialization strategy of the 1960s that has not yet come to its probable end. The neo-liberal glass in New Zealand is either half-full, or half empty, depending on your orientation.

Elements of the policies of the 1980s and 1990s remain, but there have been reversals, and they are overlaid with other incompatible initiatives, such as Working for Families. New Zealand Superannuation survives, seemingly untouchable, despite neo-liberal attempts to reduce and marginalise it. But these should not be confused with attempts to reform it to make it more sustainable into the future as people live longer and the proportion of those over 65 in the population continues to grow. 

We should not confuse neo-liberalism with responsible economic management that acknowledges that government cannot tax and spend without any apparent constraints. The implicit alternative to neo-liberalism implied by many on the left is simply not feasible in the 21st century. Indeed, it probably never existed even in the supposed heyday of the welfare state in the 1950s and 1960s, when much less was spent on welfare and its scope was much smaller. Many on the left therefore think in terms of a construction of neo-liberalism to blinds them to the alternatives that do exist to promote the more inclusive and egalitarian society they seek. In the worst case, they find themselves locked into a political ghetto in which everything is wrong, and there is no hope. By communicating this despair to others, they weaken their own cause.


May 17 2015

Reply to Roughan

The following takes issue with John Roughan’s recent opinion piece in the New Zealand Herald (May 16 2015).  
Small parties in government tend to get punished more than big ones, that’s probably true, if punishment is in order. In the cases of Britain and New Zealand, that’s almost certainly because of lack of experience of handling such relationships. Nevertheless, small parties survive, and bounce back, as witnessed by New Zealand First. And the trend within party systems in most of the  ‘old’democracies is towards more small parties, not less of them.

Roughan says that, contrary to the expectations of those who promoted MMP, New Zealand voters want strong single party governments, and that most voters retain an allegiance to a major party. But there is absolutely no evidence to support those assertions and a good deal to refute them. Loyalties to any parties are now very weak, as in the case in most other democracies. After the 2011 election post-election survey data from the New Zealand Election Study tell us that only 32 per cent of New Zealand voters wanted a single party government. The National Party is well aware of this preference, and its agreements with United Future and the Maori Party have been a consequence.

Roughan asserts that electoral reformers wanted governments to be ‘bargained with marginal players’. But this was the fear of the opponents of MMP, not the hope of its advocates.  Indeed ‘marginal players’ have rarely made much difference in deciding which of the major parties governs, except for 1996. Instead, we have a ‘two-bloc’ party system, within which parties compete on a broadly left-right ideological dimension. So far, it seems to be working fairly well.  Roughan is right in one respect. Retaining two large parties is better as it anchors the system.  With its 5 per cent threshold, MMP was designed to discourage party system fragmentation and has largely done so despite misuse of the electorate threshold. 

Roughan also states that in a multi-party government ‘both parties know only one of them is the recognised legitimate government’.  This is a highly questionable generalisation. In a confidence and supply arrangement, with junior party Ministers outside of Cabinet, those Ministers are not subject to collective Cabinet responsibility, but in the areas of their portfolios they are part of the government and act on its behalf. In a full-blown coalition, with junior party Ministers subject to Cabinet collective responsibility, they are equally ‘legitimate’ and will claim to be so.  As Roughan points out, because small parties are more vulnerable to attrition of support when in government, it is in their interest of junior coalition partners to make sure their interests are as well-represented in that government as possible.


September 23 2014

The 2014 Election

The Party Vote/Electorate Vote Gap

Labour’s losses in the party vote in some of its safest electorate seats were the subject of media commentary after the election. Of course, this was another way of saying that Labour lost the party vote badly across the whole country, and needs to address that problem. But it also carried the implication that Labour’s electorate candidates were more popular than the party, and their greater relative successes are something the party can learn from, perhaps by way of a change of strategy or personnel. As the overwhelming majority of Labour MPs are now returned from the electorates, many of them were likely to be making this argument. 

The right way to look at this is the other way about. Why did Labour’s electorate candidates outpoll the party Labour party vote? Standard political science gives us the first clue: in a single member district (electorate) the number of candidates that have the most chance of winning is two, and votes tend to consolidate behind them. Many of the votes Labour’s electorate candidates received over and above the party vote in their electorate came from people who cast party votes for the Greens and New Zealand First. One assumes that for the most part these people’s party preferences were reflected in their party vote, but they preferred a Labour MP to a National MP to represent them personally. There is not much Labour can learn from this basic pattern of vote splitting in terms of the party’s political appeal or lack of it. However, the remaining electorate votes for both the Greens and New Zealand First indicate that many people did not think of transferring their electorate votes to Labour because it had more chance of winning. With more strategic vote splitting, Labour could have won more electorates. Why was there not more vote splitting in favour of Labour candidates? That is another question worth asking.

The split voting data reported by the Electoral Commission can be analysed to address these questions . Almost certainly, some National party voters might have cast an electorate vote for a popular Labour incumbent electorate MP, but equally, some Labour party voters might have done the same for a similarly popular National Party MP. 

Some Labour candidates may have campaigned more energetically for the electorate vote than the party vote. 
There is little doubt that some people in some electorates may find some Labour candidates to be more ideologically appealing than their party. Damien O’Connor on the West Coast is probably better able to marshall a more conservative Labour vote than the party in general. Jacinda Ardern may be better at attracting a more liberal vote in Auckland Central than the party as a whole. But of course these two contrasting examples provide no basis from which to infer that the Labour Party should shift its overall strategy or personnel in either direction.
 
Systematic analysis, electorate by electorate, using split voting data, is needed to learn more.

Two more unrelated points. National’s increase in vote share compared to 2011 was overturned by the special votes. And while vote and seat changes between 2011 and 2014 were significant in terms of the substantive outcome, they were marginal in terms of net vote shifts. Estimating those, the 2014 election was almost a replay of 2011. For a closer replica of a previous election, we have to go back to 1963-1960. 

In 2011, many New Zealanders decided to give National a second term, on the basis of a much more significant vote shift. In 2014, they renewed their subscription.http://www.stuff.co.nz/auckland/local-news/manukau-courier/89580371/willie-jackson-the-health-of-our-democracy-is-at-riskhttp://www.nzherald.co.nz/john-armstrong-on-politics/news/article.cfm?c_id=1502865&objectid=11456945http://www.jackvowles.com/FrqNZES.htmlhttp://bowalleyroad.blogspot.co.nz/2015/05/tricky-customer-why-is-matthew-hooton.htmlhttp://www.nzherald.co.nz/politics/news/article.cfm?c_id=280&objectid=11449399shapeimage_2_link_0shapeimage_2_link_1shapeimage_2_link_2shapeimage_2_link_3shapeimage_2_link_4