The Treasury

Global Navigation

Personal tools

NZ Super

New Zealand has had a state pension system, of sorts, for over 100 years. In 1898 the Government introduced the first publicly provided pension, which was means-tested and available to those over 65. At the time, people who reached 65 could expect on average to live another 12 or so years.

The discussion in this section draws on the Treasury (2013). The Future Costs of Retirement Income Policy, and Ways of Addressing Them. Background paper for the 2013 Statement on the Long-Term Position. Available at

Since then, New Zealand has experimented with a number of different pension systems. We have had a two-tier system (a means-tested pension available from age 60, and a universal pension available from age 65), a compulsory contribution scheme (very briefly), and a means-tested system (via a tax surcharge on other income). But essentially we have nearly always had some kind of universal system. The key elements of the system we have now - NZ Super - were introduced in 1977, although we have changed the age of eligibility and also the payment rates since then.

Population ageing will drive increases in NZ Super in the future. Our projections in the "Resume Historic Cost Growth" scenario suggest that the cost of NZ Super could rise from around 4.3% of GDP in 2010 to around 7.9% of GDP in 2060.[93] We might be able to afford that, but we would have to either cut other government spending or increase taxes. This approach would raise intergenerational questions - is it reasonable for taxes collected from working-age people to fund the costs of NZ Super indefinitely, given the projected expansion in those costs?

Why have a retirement income system?

Different people have different views about the purpose of retirement income systems in general and New Zealand's in particular. Retirement income systems try to achieve a number of aims, which sometimes conflict. The Commission for Financial Literacy and Retirement Income has identified the following eight objectives of retirement income policy (the quotes after the bold headings are the Treasury's interpretations):[94]

Eight objectives of retirement income policy
Eight objectives of retirement income policy.

To adapt our NZ Super system to the long-term trend of population ageing, there are a number of options we could explore. Section I of this Statement set out some specific options, but this section explores a broader range of possibilities.

Could we raise the age of eligibility?

Currently, New Zealand residents are eligible for NZ Super payments at the age of 65. This is not a retirement age. Many people work past 65, which does not affect their eligibility for NZ Super. However, the fact that NZ Super is available from age 65 sends a signal that 65 is the age at which society expects most people will want to stop working. Now that people are living longer than ever, and working longer than ever, there are obvious questions about whether 65 remains the right age.

Section I of this Statement modelled and analysed the option of increasing the age of eligibility for NZ Super to 67. But we could, of course, make 67 a first stop on the way to a higher age. A few countries plan to link increases in their pension ages to increases in life expectancy.

Could we reduce the relative value of payments?

NZ Super payments increase each year, to ensure that they keep up with inflation and that couples receive at least 65% of the net average wage (in practice, in recent years the rate has been set at 66%). Because wages tend to grow faster than inflation, the payments usually grow at the same pace as the average wage, broadly maintaining the relative income position of superannuitants in line with the working population.

Instead, the growth rate could be set at the rate of inflation. The real purchasing power of NZ Super payments should remain the same, while the real purchasing power of wages would increase. Doing this would remove most of the projected increases in the cost of NZ Super. Section I of this Statement explicitly modelled and analysed the option of indexing NZ Super payments to inflation from the 2019/20 fiscal year.

Other, less drastic variations of this approach include:

  • The rate of growth could be set at some mid-point between price inflation and wage increases.[95]
  • The rate of growth could be the same as wages until 2030 (or some other date) but then inflation thereafter.
  • The rate of growth could be set at the rate of inflation, but with periodic reviews to increase its value from time to time.

Should we limit NZ Super to those who need it?

Raising the age of eligibility for NZ Super and changing the way payments are indexed would have the biggest impacts on retirees with lower incomes. An alternative would be to means-test NZ Super so that only some people receive it, based on some calculation of "means" (potentially including an asset test and/or an income test). Different types of means-tested systems are used overseas. Obviously the fiscal savings from this option would depend on exactly how many people would no longer be eligible to receive NZ Super. As two extremes, the test could be set so that:

  • only the 10%, say, of superannuitants in the lowest income decile receive NZ Super, or
  • only the 10%, say, of superannuitants in the highest income decile don't receive it.

There are several ways of achieving a means-tested system. As well as applying a direct eligibility test based on some measure of "means", we could also apply a means-test by using the tax system.[96]

There are some problems with means-testing. One of the benefits of our current universal system is that it involves little disincentive to keep working or to save. Means-testing is likely to introduce such disincentives.

Means-tests are usually complicated to apply in practice, as people sometimes hold houses and other assets in trusts. The Government would have to examine people's private arrangements carefully to get a clear picture of their "means". Further, some people are likely to spend effort and money trying to avoid means tests, which would be both expensive and unproductive. So while some people might find a means-tested retirement income system appealing on the grounds that it directs the state's resources to those who need them the most, the practical barriers are difficult to get around.


Page top