Maxim on measuring poverty

March 31st, 2015 at 7:00 am by David Farrar

The Maxim Institute has published a very readable 13 page paper on how best to measure poverty in NZ.

They make several recommendations:

  1. Poverty should be defined as a situation where: a person or family lacks the material resources to meet their minimal needs as recognised by most New Zealanders.
  2. Regularly publish a poverty and deprivation dashboard including income measures, deprivation and outcome indicators. 
  3. Use consensual budget standards to better identify what most New Zealanders think is a minimal acceptable standard of living and potentially derive an income threshold from this process.
  4. Use clustering statistical techniques to target, tailor and evaluate policy by identifying people and families with different combinations of risk factors
  5. There should be some legislative requirement that the measures and indicators above are regularly published.
  6. A poverty-specific legislative framework should not be implemented.
  7. Extend the Better Public Service targets / Results for New Zealanders framework to include reasonable, time-specific targets aimed at reducing poverty.

I strongly support the first three recommendations. Poverty should be defined based on need, not based on comparative income standards.

I also think the recommendation to have a BPS target for reducing poverty is a bold one, but worth doing.

8 ways to measure poverty

July 11th, 2014 at 3:00 pm by David Farrar

The Maxim Institute has a very thoughtful discussion paper on poverty. They pose a number of questions, and also outline eight different ways you can measure poverty. They eight ways are:

  1. Average income threshold – a percentage of mean or median income
  2. Consumption expenditure – a percentage of amount spent on consumption compared to average household
  3. Budget standards – based on a budget judged high enough to avoid deprivation
  4. Component and multiplier – based on amount needed for an essential such as food, and multiplied
  5. Subjective measures – based on self-assessment as poor
  6.  Benefit-based/statutory measures – based on government definition of minimum income
  7. Material deprivation indices – based on actually lacking essential necessities
  8. Multi-dimensional measurement of poverty, deprivation or social exclusion – a mixture of the above and more

No measure is prefect. I don’t like the income measures because they treat all households as identical in terms of needs, and they are more about income spread than actual poverty. They also avoid the effect of tax, as they tend to be on before tax incomes.

My preference is No 7.  Stats NZ already do this – an occasional survey asking if families can afford stuff such as more than two pairs of shes for kids, transport to school etc etc. This is relatively objective, and measures actual deprivation rather than merely equality of income.




Soundbites & Stereotypes

November 20th, 2012 at 7:00 am by David Farrar

Maxim Institute and the the VUW Institute for Governance and Policy Studies have arranged two public lectures/panels discussing the quality of public debate in New Zealand. The panels are:

  • Richard Harman (ex Political Editor TVNZ, produces The Nation)
  • Bernard Hickey,
  • Mai Chen, public lawyer

The details for each are:

  • Wellington – Wed 28 Nov, 6.00 pm – 7.30 pm, VUW Law School GBLT1
  • Auckland –  Thu 29 Nov, 6.00 pm – 7.30 pm, Auck Uni Bus School Level 1 Foyer

If you want to attend, RSVP Maxim by 23 Nov.

More tax tracking

June 12th, 2012 at 1:00 pm by David Farrar

Maxim has done some tax tracking scenarios, for four different people or households. They are:

  • Minimum wage earner
  • Median income earner
  • Median income household
  • High income household

The minimum wage earner pays an average 13.5% tax being $3,320. The high income household pays $33,240 tax. Some of the breakdown of what that goes on is:

  • NZ Super $4,745
  • Family Tax Credits $995
  • DPB $857
  • Invalids $622
  • Accom Supp $580
  • DHBs $5,611
  • Primary schools $1,355
  • Tertiary tuition $1,121
  • Secondary schools $1,010
  • ECE $653
  • Student loans $330
  • Tertiary allowances $289
  • Defence $926
  • KiwiSaver $332
  • Debt $1,748

Maxim on the MMP Referendum

September 20th, 2011 at 7:00 am by David Farrar

Maxim have a handy wee paper by Steve Thomas on the five electoral systems on offer in the upcoming referendum. They don’t say which one is their preference. It’s a good guide to the pros and cons of the various systems, so I’ve embedded it below.

Their summary of the five systems are:


MMP provides well for electorate representation and the representation of interests, and can provide for reasonably stable government.21 The strength of MMP is the flip-side of its drawbacks. It enables more parties to be elected to parliament, which is great for the breadth of representation, but it also gives parties a lot of power. It can also create bargaining instead of debate among parties, and a weakened accountability of the government to voters. It can also encourage interest groups to act in unhelpful ways.


FPP is simple to understand and it usually produces clear results. FPP delivers strong, stable, single party majority government most of the time, and there is usually no confusion about which party can form a government. It is easy for voters to dump a government and elect a new one since parties generally do not negotiate together to form a government.20 But, as New Zealand’s experience indicates, instances of highly disproportionate election results weakened the legitimacy of electoral outcomes and the Cabinet’s tight control over legislation and parliament weakened the public’s trust in government.21 It can also be difficult for minorities to be represented, either because safe seats make it difficult to dislodge a popular candidate or because it is difficult for minority candidates to win enough concentrated support in one electorate.


PV provides for strong electorate representation, through the election of local MPs, which usually leads to the election of single-party majority governments. That said, PV gives minor party candidates a fighting chance of winning a seat when second and subsequent preferences are used to help elect a candidate. However, it is still harder for minority candidates and parties to be represented in parliament under PV because it is not a proportional system. Further, PV can sometimes produce electoral outcomes that might not be considered entirely legitimate if the most popular candidate on first preferences does not win—although this point is debatable. While PV would enable voters to more clearly express their preferences for certain candidates it could also introduce some new ways for parties and candidates to engineer electoral outcomes, as parties would advise supportive voters how to vote to give them the best advantage.


STV is an attractive system in principle since it enables voters to indicate exactly which candidates they would like in multi-member electorates. STV enables voters to choose both between and within parties, meaning that parliament ought to reflect a wider diversity of opinions within society.22 The use of multi-member electorates also means that electoral outcomes will be more proportional.

The theoretical advantages of STV have to be weighed carefully against the practical issues with using it and the way voters tend to interact with this relatively complex system. For example, it could undermine the cohesiveness of political parties as candidates from the same party would compete against each other for election. The option of voting above-the-line can also give parties more control over which candidates are elected and in which order. In this case, many voters would not actually end up individually choosing their local MPs. In short, the advantages offered by STV could be eroded by measures to make it easier for voters to understand and use.


In trying to blend two styles of voting system, SM has some of the benefits and some of the drawbacks of both. It is neither a completely proportional system, nor does it guarantee that one party will win a large enough majority to be able to govern alone.

In terms of representation, SM has the potential to achieve a good balance between national and local representation of interests. Electorate representation would be strong, creating good ties between parliament and voters, but a quarter of parliament would also be made up of list MPs who tend to be able to represent minority interest groups well.

Because there would be more electorate MPs under SM than under MMP the major parties would benefit, but there is also a chance coalitions would be needed to form a government and that minor parties would have more representation than they typically do under single-member electorate systems, like FPP.

The document is below.

EMBARGOED Kicking the Tyres

Maxim on Tax

May 10th, 2010 at 12:01 am by David Farrar

Maxim have released a comprehensive 102 page report on tax policy by Steve Thomas. Its aim is a tax system that maximises economic growth. It notes:

Growth is affected by tax, which is how the government raises its revenue to do the crucial things we need it to, like paying for a police force or a public education system, building roads and supporting the poorest when they need it.

However, when we try to take too much money out of the economy in tax to fund government spending, we risk undermining the very source of that revenue. Also, if government spending is misdirected or of poor value, then we hamstring the economy’s ability to produce what we need and the amount of tax the government is able to collect.

This relationship between tax and the economy therefore needs to be carefully considered. We need to design the tax system so that it allows the government to take the money it requires, while doing the least amount of damage to the economy and so too our potential prosperity.

This is absolutely right. It is a balancing act between economic growth and funding Government services. To take two extremes – an economy with tax rates of 95% would end up like North Korea, while an economy with tax rates of 5% would not be able to fund much in the way of defence, health or education.

Maxim propose a number of policies:

  1. A two step personal tax rate system with a top rate of 27%
  2. A corporate tax rate of 27%
  3. Aligning the trust and PIE rates to the personal and corporate rates
  4. Removing tax incentives for KiwiSaver
  5. No land tax or capital gains tax
  6. GST from 12.5% to 15%
  7. An upper limit for central govt spending of say 30% of GDP
  8. A benchmark for core govt expenditure on welfare of around 15% of GDP

For me the key thing is No 7. If one can limit spending as a percentage of GDP, then you get options around tax reform. Maxim note:

A 2001 OECD study found that about one half of a percentage point increase in government consumption (the expenditure to GDP ratio) could cause a 0.6 to 0.7% direct reduction in per capita output.

If we can limit spending so that over time it is under 30% of GDP, then there will be a very significant boost to incomes and jobs.

What I would like is both National and Labour to outline desired limits for spending as a percentage of GDP – then voters could choose between them. The limits probably need to be soft (non legislative) to take account of recessions etc, but a soft limit would still be a huge improvement over no limit.

One can get to a limit without massive spending cuts. If one can retain discipline over new spending so that it grows significantly slower than the overall economy, the ratio will reduce over time.

A very good report.I suggest people don’t just argue the recommendations but read the summaries of research about why such tax changes will be good.

Maxim on Three Strikes

May 5th, 2010 at 4:00 pm by David Farrar

The Maxim Institute has a paper by Professor Warren Brookbanks and Dr Richard Ekins on the Three Strikes law. It sets out arguments against the law, the main being a third strike of a maximum sentence is not a proportionate response.

Now I disagree with them in opposing the law. I think it will e a welcome step towards stopping the idiocy that we keep letting offenders out after minimal sentences for repeated serious offending.

But Brookbanks and Ekins have also proposed some amendments to the law, which I do partially agree with, and the Government could consider at the committee of the house stage. They are:

Authorise judges not to impose the maximum sentence on strike three if this would be manifestly unjust (this amendment would bring the legislation into line with the assertions being made by the ACT Party).

At the moment the law directs a Judge to impose the maximum sentence without parole for a third strike, unless this is manifestly unjust in which case they can give the maximum sentence with parole.

So long as the threshold for “manifestly unjust” is very high (and case law suggests it would be) and not used frequently, I support this change. There may be times when a 14 year sentence for aggravated robbery, even with parole eligibility, is manifestly unjust.

Retain presumptive eligibility for parole, or if this is not done, authorise judges not to order the sentence be served without eligibility for parole on strike two if this would be manifestly unjust.

I don’t agree with this proposed change. I regard parole as generally being a failed experiment, and the certainty of no parole for a second strike is important. Possibly could live with a “manifestly unjust” exception again but I worry some members of the Judiciary would interpret that to apply to every case as they don’t agree with the law.

Modify what counts as a strike from a conviction for a qualifying offence alone to at least a custodial sentence for a qualifying offence and preferably a custodial sentence of some length, say at least two years.

This doesn’t catch enough people, and it means that if Judges keep giving light sentences for serious violent offences, the offender never comes under the three strikes regime.

Make provision for strikes to lapse over time (perhaps after ten years).

I think this can be reasonable. Maybe a bit longer period than ten years, but I have sympathy for someone who does a first and second strike at 18 and 19 and a third strike at say 55, and they automatically get the maximum sentence. One could argue this can be dealt with under manifestly unjust exceptions, but I think an incentive for a strike to lapse is a good thing.

However I would make it a condition for a strike to lapse, that the offender is crime-free for that entire period of 10+ years. Not just of serious offences, but of all offences.

Make specific provision in strike three sentencing to recognise a guilty plea, allowing judges to discount the maximum sentence by up to 25 percent, depending on when in the trial process the plea is made.

At present, an offender gets a “discount” on their sentence of 5% to 33% for an guilty plea, depending on how early they plead guilty.

There is a potential problem that there is no incentive to plead guilty to a third strike offence. Some discount for an early guilty plea could solve that problem, and not undermine the regime overall.

Authorise the courts not to impose a life sentence for murder and manslaughter if this would be manifestly unjust.

Again, so long as manifestly unjust holds a very high threshold, I could love with that.

Specify that some instances of manslaughter (most notably accidents arising out of gross negligence) do not constitute a qualifying offence.

I suspect very few people convicted of that type of manslaughter have criminal histories, and it would normally be a first strike anyway. However I think the proposed change has some merit, in differentiating between types of manslaughter.

I’m not sure if the Government is open to changes, but it would be good to hear debate on them. Maybe Opposition MPs can move them as amendments.

Broken Boughs

October 13th, 2009 at 3:00 pm by David Farrar

An interesting 68 page report from Maxim on the role of effective family interventions. They have actually done an evaluation of the dufferent family intervention programmes.

Their conclusion is:

We conclude that the preferred approach to family intervention is to develop and implement programmes that are family-focused, intervene early and are targeted to children and families that are at risk or experiencing actual problems.

They evaluate seven programmes:

  • Family Start
  • Parents as First Teachers
  • Early Start
  • Whanau Toko I Te Ora
  • The Parenting Programme;
  • Incredible Years
  • Triple P.

And their recommendations:

  1. Three programmes should be further implemented and funded: Early Start, the Incredible Years and Triple P.
  2. A comprehensive review of exemplary, evidence-based home visitation and parent management training programmes should
    be undertaken to identify other programmes that might be suitable for implementation and funding in New Zealand.
  3. Funding should be carefully allocated. Existing funding from unproven and/or ineffective programmes like Family Start and Parents as First Teachers should be reallocated to evidence-based programmes.
    Continued funding should be conditional on programmes continuing to demonstrate that they are effective and delivered with fidelity.

NZ Votes

October 30th, 2008 at 1:00 pm by David Farrar

Maxim have put together a nice election resource at NZ Votes.

They have a useful graphic guide to the various sites such as decision08, TVNZ, Scoop, Voteme, etc with what you can find on each site. Also links to all the political parties, and details of upcomings forums.

Well worth a check out.