As another Review of Retirement Income Policy in New Zealand gets underway, here are my top 3 questions on KiwiSaver.

Most experts agree that the stress on KiwiSaver in the Terms of Reference for the 2025 Review is well-placed.  The Retirement Commission has already suggested ways to improve KiwiSaver.  There are worries that KiwiSaver account balances are low compared to other countries, and not evenly spread among people.

At the most basic level, pension savings is a simple concept.  You save money while you’re earning and then spend it during retirement.  My top 3 KiwiSaver questions are all to do with helping people understand how to do this: 

1. How much they ‘should’ save – the target pot
2. How to get there – default and mandatory contributions
3. How to think about using their funds in retirement – drawdown

A PDF of this post can be downloaded here.

Help on how much people ‘should’ save

People can find it reassuring to know a realistic figure for the size of the target pot they ‘should’ aim to have saved before they start drawing on it in retirement.  Ideally, advisers, commentators and policy makers would be aligned on how to calculate and communicate the pot size.  However, there are different motivations at play.  An individual may want a simple and safe benchmark that doesn’t mean scrimping now and saving too much, but a KiwiSaver provider has a commercial interest in increasing contributions.

The ‘should’ is in quotation marks, because there is no one-size-fits-all answer.  Even with perfect knowledge of the future, people will have different views on how comfortable they want their financial retirement to be, and how they make tradeoffs between spending less now to save more for later.

In New Zealand there are a couple of suggested benchmarks for how much income people might need in retirement. 

• The long-established Retirement Expenditure Guidelines from Te Kunenga ki Pūrehuroa Massey University’s Financial Education and Research (Fin-Ed) Centre describe how much some current retirees spend in a year.
• Te Ara Ahunga Ora Retirement Commission suggests using replacement rates.  This hasn’t yet been worked into a widely used method and Sorted, the Retirement Commission’s calculator website, uses the Massey guidelines.

Both of these benchmarks for target retirement income can be converted into a target savings pot using the FMA’s inflation and investment assumptions for KiwiSaver projections. Using consistent best estimates for these assumptions is key to avoiding unrealistic targets.   In theory, this should be fine, but there are some silent assumptions behind the calculations which are generally not explained, and probably misunderstood:

• First, for which specific people the benchmarks are intended.  The Massey “No Frills” and “Choices” benchmarks are what current retirees in the second- and fourth-income quintile are spending. So just not applicable to someone who would choose to be (or turns out to be) in a different part of the distribution. 
• Second, how spending is assumed to change through retirement. In both New Zealand’s benchmarks, annual spending – and by implication income requirement – is assumed to increase with inflation during retirement.  But RIIG* shows that spending is likely to decrease in real terms through retirement, as real-life evidence suggests, and many international studies support.   Making this assumption leads to a lower target pot.
• Third, how long retirement is assumed to last. The target pot is often constructed assuming income will be received for a number of years, but it isn’t stated how probable that is (given the rate of growth of the invested fund is uncertain) or it’s assumed the income needs to last only until age 90.  However, the actuaries in RIIG recommend people consider their retirement plans should they live to between age 90 and 95 years.  One in five New Zealanders who reached age 65 in 2023 is expected to live to at least to age 93 (men) and 95 (women).  Understanding when income might run out and what could happen if it does should be part of any communication about target pots.
• Fourth, what is being assumed to stay the same throughout saving for retirement and during retirement.  Generally, it is not stated that people’s own circumstances are assumed to stay the same and that taxation and New Zealand Superannuation (NZS, the public pension) stay as current. Both are unlikely. 

Another problem is that it is difficult to explain the different answers from the benchmarks. The Massey headlines suggest target balances from $120,000 to $1,142,000 across the two spending levels and regions of New Zealand (for a two person household).  In contrast, a replacement rate approach on a median income suggests $605,000 if spending is assumed to increase in real terms, or $375,000 if real spending reduces by 2% a year.

These figures may be correct given their method and assumptions, but how can people understand what is behind the figures and what is right for them?  We need a better way to help people feel safe in setting their own target pot, understanding enough to accept the silent assumptions and the risks they are taking.

I don’t have a full answer, but it might include:

• A settled view on what method to use to calculate and describe target pots, to reduce confusion.  My personal view is that the replacement rate method is more useful than the Retirement Expenditure Guidelines, because it is a forward-looking estimate for an individual rather than an interpretation of survey data from the very diverse group of current over-65s.
• Some Good Practice Principles for choosing and communicating the assumptions in the calculations. In the Good Practice Principles for Retirement Modelling in Australia, the Actuaries Institute lists the assumptions which retirees should:
  • be able to choose,
  • trust as expert decisions behind the calculations, but be able to see and understand the implications of the choices made, and,
  • trust as expert decisions behind the calculations, and elect to be able to see them if interested.
• A policy statement from political parties about their long-term intentions on KiwiSaver and NZS policy and a commitment of the timescale on which they would make any changes.  The same can also be said of other public sources which people might need to call on in later life such as the costs of health care and aged care.  The aim must be to give people time to adjust their savings plans if needed, and to avoid changes to the finances of the oldest retirees with little or no scope to adjust.

Default and mandatory contributions

Once we have better clarity on what target pot someone is aiming for, the contributions required to get there can be calculated, again using best estimate consistent assumptions. 

KiwiSaver is a voluntary scheme, allowing individuals to save as much or as little as they want at any time.   This freedom is good, because people have times when saving is hard.  There are different ways to reach the same target balance by age 65.  Not everyone (anyone?) will have a career with steady salary increases and regular KiwiSaver contributions exactly as assumed in savings calculators.

Policy can nudge people to stay in KiwiSaver and guide savers on how much they ‘should’ save to reach a target pot by setting default contribution rates for individuals and employers.  The Retirement Commission worked through examples in their recent KiwiSaver report using replacement rates.   They recommended a higher default contribution rate of at least 4% (with employer matching at this level), and keeping a minimum contribution rate of 3%.

The 2025 Review is a chance to examine this approach and test all its assumptions, including the silent ones.  But the recommendation looks on the right lines as it distinguishes between people of different income levels and does not call for an unfeasible change in behaviour or a magic money tree.

There are other detailed issues in how contributions are set by age or employment type.  A clear principle such as 4+4 default/3+3 minimum should help to guide these issues to a solution.

How to think about drawdown

RIIG has provided a framework to help people planning retirement to think about how to take income from their savings pot through retirement.  This helps avoid the competing risks of taking too much from savings and running out of money or taking too little and not enjoying retirement.  The framework will be on Sorted soon, so there is the potential to make it New Zealand’s go-to way of thinking about drawdown.

It’s wonderful that RIIG’s drawdown framework will be on Sorted, but that is not the full answer to how best to help people.  There is more work to do to settle in how the framework is used and explained – another set of Good Practice Principles, perhaps.

There is also policy work to do to hold the line against complicating drawdown from KiwiSaver.  Drawdown in New Zealand is simple compared to other countries because of tax structures (KiwiSaver contributions are taxed on the way in so no tax is payable on the way out), lack of means-testing (KiwiSaver withdrawals do not affect how much NZS is payable), lack of product choices (there are no annuities for sale), and a single KiwiSaver account which follows the member through working life. 

We should be wary of implying too much about the benefits of post-retirement products used in other countries when they have been designed to fit around complexities which New Zealand doesn’t have.  Simplicity means it is easier for people to make rational drawdown decisions with no constraints, a real advantage for New Zealanders.

And we must acknowledge the nascent size of KiwiSaver compared to for example, the Australian superannuation industry which is 37 times larger**.  For the next 20 years, most New Zealanders reaching age 65 will have helpful but modest balances in their KiwiSaver account.  RIIG calculated in 2021 that three-quarters of contributing KiwiSaver members then aged 45 would have less than $250,000 in their KiwiSaver balance when they reached age 65.  Older members have even less.  

For now, the most efficient and helpful policy option in New Zealand is to make drawdown work.  It’s important not to let commentary relevant only for the largest KiwiSaver balances drive the agenda for the majority.

******

* RIIG is the Retirement Income Interest Group of the New Zealand Society of Actuaries, of which I am a member.

** KiwiSaver assets NZD111.8 billion at 31 March 2024 (FMA) when Australian superannuation assets were AUD3,852.1 billion (ASFA)