Policy issues relevant for the savings industry

Since the publication of the last Periodic Report Group (PRG) report in December, there has been little in-depth debate on any of the limitations of the uniquely New Zealand approach to retirement income provision. Instead there has been a unilateral adjustment to the level of the state pension and an announcement of yet another task force. Issues of equity that imply a need for integration of the state pension with private retirement provision, and the role private pensions might play in the future have not been discussed.

The private savings industry needs to press for resolution to the issues surrounding the state pension and understand the consequences of further government tinkering. Furthermore, society needs to consider what is likely to happen to the income distribution next century as fewer and fewer people have annuities or pensions from private provision, as employers withdraw from their traditional role. It is important too to remember the impact of other policies on the overall level of welfare of the retired population. Health care, housing and long-term care policies are intimately involved in any consideration of the appropriateness of the retirement income package.

The New Zealand retirement income system is quite different to those of other OECD countries. There are no separate compulsory savings schemes and no tax incentives for private saving for retirement of any kind. The state provides a flat-rate, taxable, universal pension called New Zealand Superannuation (NZS). This is set at a significant level so that for those with no other financial resources, few additional means-tested supplements are necessary, especially for those retirees who own their own home. [1]

Eligibility for New Zealand Superannuation is based on meeting the qualifying age [2] and residency requirements. An applicant must be a legal resident of New Zealand and have lived in New Zealand for a total of ten years from the age of 20, and five years since the age 50, normally live in New Zealand and be living there when the application is made. The level is linked to net average earnings through the use of a wage band [3] which currently provides a net pension equal to 33% of net average earnings for each married pensioner, and 39% of net average earnings for a single pensioner (60% of the married rate). For those living alone, a living alone rate of 65% of the married rate is paid.

The age for the state pension has always been the same for men and women and is currently being raised from 60 to 65 over a period of ten years finishing in 2001. In the transitional period, a social security benefit called the Transitional Retirement Benefit has been available to those people whose plans have been disrupted by the rapid raising of the age of eligibility for the state pension. The age for this benefit is itself rising so that it will completely phase out in 2004. At this point there will be no official pre-retirement or early retirement benefits.

Expenditure on the net state pension has fallen significantly since the age of eligibility began to rise. The net retirement pension cost approximately 4% of GDP in 1996/97 and is projected to reach around 9% of GDP over the next 50 years under current settings. In comparing this cost with other countries it must be remembered that New Zealand has no hidden tax expenditures for retirement.

New Zealand provides few other non-financial benefits to pensioners. Government spending on the public health system is currently 5.7 % GDP. Medical care is not provided free at the point of use to all New Zealand residents, but a community services card is available to those on low incomes, and a high use card for those with chronic illness. New Zealand Superannuation recipients do not automatically qualify for the community services card, but their relatively low incomes mean that over two thirds hold one. This entitles them to a higher subsidy for visit to the GP and for prescription subsidies. [4] Access to the public health system for many formerly widely available services such as cataract operations has diminished markedly, due in particular to few funds being made available for elective surgery. Many superannuitants now hold private insurance, although the insurance companies have excluded many by sharp premium rises in recent years.

Residents in all long-term residential or hospital care face the same income and asset test. After a review, a maximum personal contribution of $636 per week was introduced in 1994 for care in all long stay institutions including private and public hospitals as long as that care is appropriate to the needs of the person concerned.

Eligibility for a rest home subsidy is determined first by an asset test. The threshold for married couples with one spouse in long stay care is now $45,000 with house, car, personal effects and prepaid funerals (up to $10,000) remaining exempt. A single person without dependent children may retain $15,000 with no exemption for the family home. A married couple, both in care is effectively treated as two single people with a joint exemption of $30,000.

The amount of subsidy for those who pass the asset test, is based on the difference between the fee-for-service rate and the resident's total income from all sources. The government provides up to $29 per week for those who pay full fees but who do not have enough income left for a personal allowance. The income of the spouse is counted on a dollar for dollar basis in the income test. When the spouse is working, the exempt amount for this test is $28,927 of spouse's earnings where there are either no dependent children or only one child. For three or more children, the exemption is $36,553 (Minister of Health 1996). These thresholds were not changed for 1998. Also exempt is income from any benefit, and income from assets below the threshold levels. There is no allowance for older children at tertiary institutions who may also be dependent.

Between 1993-1996 the Superannuation Accord gave retired people protection against partisan and ill-considered changes and those coming up to retirement could plan with a reasonable degree of certainty. This framework was endorsed by the Periodic Report Group's year-long comprehensive review in 1997. The PRG found the current pension, with the rise in the age to 65 by 2001 and the wage band formula to be adequate, efficient and sustainable. The after-tax cost of the pension as a fraction of GDP is not expected to reach the level of the early 1990s again until 2020 so there is no crisis and no need to panic. From 2015, some well-signalled, moderate modifications could reduce the cost, but the detail of these requires no urgent decisions. Nevertheless, the PRG noted that the abolition of the surcharge was a mistake. It fundamentally weakened the glue that kept the multiparty agreement on track.

Admittedly the surcharge was always contentious. But agreement had been reached on adjustments to it for 1997/98. In that year only 16% of the top income recipients were affected, and only the richest 5% had all their super payment clawed back. The PRG said for reasons of intergenerational acceptability, and because the revenue loss was significant, some way of reducing the net pension as other income rises must be reintroduced. They set out five options for discussion. Critically, change should only occur after multiparty participation and wide public discussion.

The next major review was to be in 2003. If the report of the PRG had not been sidelined, we might have expected by now to have an open process for the conduct of multiparty talks, serviced by a well-resourced secretariat and chaired by the Retirement Commissioner. A number of issues needed urgent discussion such as; funding for new surveys of wealth and saving; tax and disclosure issues for superannuation schemes; the collection of information on living standards of older people; the appropriateness of the 55+ benefit for people unable to work but not yet eligible for NZ Superannuation; the appropriateness of the wage band formula in light of the proposed changes to the Quarterly Employment Survey's calibration of average wages and, the design of a link between private and public provision.

The PRG had recommended a new process for multi-party talks in which political parties would agree on a set of common principles but could have some differentiation on matters of detail. Any new legislation that did not have significant multiparty approval might proceed only after a period of time had passed, including an election. Setting up such processes was never going to be easy, but did the politicians really try?

Just ten months after the PRG report, another large well-funded task force is announced. Its purpose is to make recommendations to be passed into law by late 2000 for a system which is 'adequate, equitable, efficient and sustainable'. It is hard not to conclude that this cuts across the work of the Office of the Retirement Commissioner and involves duplication of the work of the PRG.

Damage has been done to the shape of NZS by the sudden, unilateral change to the indexation of NZS announced in late 1998. The floor for the couple's net rate of superannuation is to fall from 65% to 60% of the net average wage, by indexing to prices alone until the new floor is achieved. There is now no certainty or security that comes from an agreed process of decision-making. It is therefore of no surprise that the major opposition parties will have nothing to do with the new task force. If the change to the wage band floor can be pushed through Parliament under urgency, without discussion, consultation, evaluation or even a rational philosophical basis, the precedent is set for other arbitrary changes. Thus the floor could be lowered again, the age of entitlement raised, or mean tests introduced as the government sees fit. Stability for the long term has been fatally compromised.

The current shape of NZS raises serious questions of equity among members of the retired population. The projected saving of $2.4B over the next ten years from the reduction to the wage-band floor comes at the expense of the living standards of the 50-60% of superannuitants who get almost all of their income from NZS. Over the same period at least $3B from the surcharge removal flows to the wealthiest 16%. Sacrifices have been made by low-income superannuitants over the last decade, but they will now be denied participation in any recovery in the economy's fortunes. Their incomes will be linked only to prices.

The second equity issue is between the well-off retirees and the less well-off of the working age population. Universal pensions and low tax makes no sense in a user-pays world. As a result of the tax cuts, benefit adjustments and the removal of the surcharge the better off two-income superannuitant households have benefited by up to $350 a week A couple going on the sickness benefit now gets $30 a week a less. Low taxes for middle income families are a mirage. Many families face effective marginal tax rates (EMTRs) of 52.4%% and higher over long income ranges, due to tax and ACC, the abatement of family support and loss of other income supplements.

As noted above, in the last year of the surcharge (1997/98), only 16% of all recipients were expected to pay any surcharge. Only about 5% found that all their superannuation was clawed back (PRG, 1997a, p.48). A further small percentage of those who are eligible chose not to receive NZS at all, because the surcharge would effectively take it all back. The surcharge as an income test was a rather lenient one, and was always based on individual not joint income for married couples. The abolition of the surcharge in 1998 means that, just as prior to 1985, the pension payment is fully universal. The difference between the early 1980s and the late 1990s is that, the top tax rate has been considerably reduced, so that the very wealthiest can now retain 67% of the gross compared to only 34% in 1984.

So, the state pension now looks more vulnerable than it has since 1991. The process of rational change has been destroyed, and the inequities now evident are likely to provoke more unilateral action. It is important that the savings industry is clear about the implications of these possibilities.

There are few immediate fiscal reasons for making changes to the state pension. Nevertheless some conclusions follow from the above discussion. The removal of the surcharge was achieved for political reasons. The same is the case for the change in the indexation formula. Universal lower level pensions have not been based on any articulated philosophy that might be expected to underpin popular support for them. They are in sharp contrast to other parts of the welfare and user-pays systems. Students, for example, are now facing yet higher fees and debt for 1999 and their promised universal allowances (Coalition Agreement 1996) have failed to materialise. Young families no longer receive free healthcare and a universal family benefit.

Table 1: Distribution of over 65 non-NZS income individuals 1996 [5]

The growing divide between the better-off retirees, often with comfortable employer subsidised pensions, and those with few assets and no additional income is stark and likely to grow. Table 1 shows taxable income (adjusted for tax-free pensions) and shows other income over and above the state pension. The distribution is highly skewed with the top deciles of income having more private pension income (PRG, 1997a, background document).

In the final report (PRG 1997b) the PRG put forward some suggestions as to how to re-integrate the state pension and private provision. The wide debate around these options that was expected has failed to materialise.

The options presented by the PRG need to be read about carefully to appreciate just how and why they have been constructed. Essentially they extrapolate from today's distribution, building in some assumptions about growth in incomes. The figures are in today's dollars.

Some of these options have severe disadvantages. It is clear from the income distribution that 75% of superannuitants do not have significant amounts of privately-sourced income. A fully income-tested welfare benefit approach to NZS (such as advocated by Gareth Morgan [6]) would have a severe impact on the willingness the save and work in retirement. So an option that looks like it saves a lot of money may have some nasty downsides. If taxable income alone is used as the base for abatement there are incentives to save in ways that do not give rise to such an income; capital gains, for example. The effect on work effort is really important with an ageing population; especially an ageing population that is forming an increasing fraction of the total population. We do not want to give the signal that supplementing retirement income with a part-time job is not worth while.

A full welfare income test would reach even into 9^th decile with extreme effect on the incentive to have extra cash income from investment and work effort. Moreover if it is a standard social welfare benefit based on joint income the effects are more severe than those shown in the accompanying slide "A Welfare Approach". Many women for example would find that their pension would disappear because of their husband's earnings, and some men married to younger women still working would lose their right to an independent income. This may be fine in marriages where there is full sharing and caring, but with the greater fluidity in relationships today, joint income testing would be a major step backwards. This option would also make New Zealand look very peculiar in the international context. Essentially we would have a welfare benefit for the poor and no state assistance of any kind for retirement saving. The pressures for tax incentives and compulsory savings might become irresistible.

In other options, part of the pension is universal and part is targeted. One of these versions as presented by PRG (1997b) is shown in the slide "Example: Low price indexed universal or age benefit". If the top-up is one half the total, it takes between $4,000 and $11,500 of other income to abate the pension away. Thus it would worry all those with incomes just under $4000 and 20-30% who get some abated amount. Thus again it would affect the majority of people over a long income range. It certainly would not be worth having a part-time job that pays $200 a week or income from a small capital sum of say $100,000. A part targeted option runs the danger, as with full targeting, of discouraging the very behaviour that should be being encouraged. Such policies trap people at the lower end of the income spectrum, while allowing those at the top end a substantial bonus of a part pension as well as low tax rates.

Who is going to be affected? The bottom deciles are more likely to be older people, more likely to be women and more likely to be Maori. It is these who queue up for the top ups, and many will fail to access their entitlements because of stigma and complexity. Remember the current arrangements already have a raft of special add-ons. A part targeted option greatly expands this and creates welfare beneficiaries out of the bottom 70% of the retired.

We need to ask: are the top 20% the ones for whom savings incentives really matter? In any case, if the receipt of a universal pension has a strong income effect, high income recipients may save less because of it. The combination of high marginal tax rates and the prospect of income and asset testing of long term care is bound to encourage yet more formation of family trusts and inappropriate gifting.

The need for integration must be debated. It is the younger generation that are going to be most adamant that universal pensions for the wealthiest New Zealanders do not make a lot of sense. Any system however must have integrity and must not be piecemeal in implementation. Thus a consistent and wide definition of income for an income test is critical. The interaction with income and asset tests for long term care, and the need for encouragement of private pensions are broader policy issues that must also be considered in any targeting design.

In light of the severe and damaging effects of tight targeting and a welfare approach to NZS, the tax credit option suggested by the PRG merits closer attention.

The tax credit option (shown in the slide "The tax credit approach") can be designed to bear a close resemblance to the effect the surcharge had in 1997. The bulk of retirees are therefore unaffected. Most of those who are subject to a clawback would pay effective tax rates of only 46%, with a 58% applying only over short income ranges for the wealthiest. This is not nearly as much a disincentive as the greater than 90% rates of welfare- targeted approaches.

It would operate through the tax system, but would have a much more positive image than the old surcharge. It would emphasise that older people are treated favourably by the receipt of a special tax credit, that offsets their tax on other income, rather than punitively, by a clawback of something that was theirs by right.

An attached spreadsheet shows how it might work. Because there is a strong case for bringing the he married and single rates together (PRG, 1997a) the married rate is used. The example is based on the current married gross rate of $9,932. Thus its net rate of $8,424 for those with no other income is the value of the tax credit. A person living alone would receive an additional tax credit.

The vast majority of people would receive a tax credit of $8,424 per annum as a fortnightly payment from the IRD. For each superannuitant on income of over $7,605 (the old surcharge exemption) the tax credit would operate much like Family Support. Like the old NZS surcharge, it could abate at 25% until it disappears.

NZS recipients would not be also entitled to the low-income earners rebate. In other words all income below $38,000 would be taxed at the statutory rate of 19.5% .The tax credit option as demonstrated in the spreadsheet is slightly more generous than the old surcharge. The tax credit is fully abated at around $41,000 of other income.

The 20-25% of people on income of over the exemption could opt to do one of three things:

• Pay tax on other income at a zero rate with a reconciliation at the end of the year. This is a good option for those whose income is around $25,000. They may find a small tax rebate is due at the end of the year.
• For those with incomes $7600-15,000, a reduced fortnightly NZS payment based on the previous years income could be made with a simple reconciliation at the end of the year. Some superannuitants may like the idea of taking it all as a tax refund at the end of the year and using it to pay the rates etc.
• Either claim the rebate as an offset to tax at the end of the year (good for all over $25,000), or select a zero tax rate for income up to $24,000 and a 41% rate above that.

Example, supposing the exemption is $7500 (approximately the case in 1997/98):

Earned income = $20,000

Tax at 19.5% = $3,900

Tax credit = $8,424- 12,500 x 0.25 Tax credit = $5,299

Either pay tax at 19.4% and receive weekly NZ Super tax credit payments of $100;      Or pay tax at zero rate and get $1,399 refund at the end of the year.

It should be noted that none of this is to minimise the problems of the integrity of the surcharge. The issues of what to count as income for these purposes needs wide discussion and cannot be addressed in this paper.

Saving for retirement is largely considered a matter for individual choice. The tax-funded nature of the state pension itself can be regarded as the core compulsory arrangement for most workers. They can then choose to supplement the state pension with saving in a wide variety of ways including repaying the mortgage on their own home or investing in their future earning capacity by undertaking education.

The theory has been that the form of saving should not be determined by differential tax treatment between savings products. Neither is it deemed to be in the best interest of savers to compel them to save at certain times in their lives at certain prescribed rates, in narrowly defined products. A fairly flat income tax scale was a necessary part of this regime.

There are some special features of the New Zealand system that are operating to the discouragement of the traditional employer-subsidised schemes. Private pension provision in occupational schemes is now the preserve of a relatively small fraction of the working age population. Thus few retirees have a private pension from an occupational plan, and fewer are expected to have one in the future. Men are much more likely to have occupational pension income than are women, especially married women. Overall, only around 12% of all individuals over 65 have income from an occupational pension scheme or a private pension.( Table 2)

Table 3 for 'older' households which are either single people over 60 or couples where at least one is over 60 shows that over time private superannuation provision including dividends and interest has risen to 27.9%. But confirming the data from Table 2, job superannuation is a small share (under 10%) of income. Table 3 shows how occupational superannuation is by far of most importance to the top 2 quintiles of income.

(Source: New Zealand Household Economic Survey 1987-88 - 1995-96, Statistics New Zealand, Krishnan, (1997).)

For those people with private pensions, lack of indexation has in the past been a significant eroding factor. Currently the rates of inflation are very low so that it is not an issue attracting much attention at present.

Of the current workforce, membership of occupational schemes has been declining and new schemes have tended to be defined contributions schemes reflecting a shift away from defined benefit schemes (PRG 1997a, p.184). Table 4 shows a breakdown of who is contributing to occupational and private schemes. It is clear that men are much more likely to make contributions, and when they do, are much more likely to make larger contributions than women. Table 4 does not however give information about the nature of the schemes, nor the contribution that may be made by the employer on an employee's behalf. It is save to assume however that the higher income contributors are more likely to have matching or greater contributions from employers.

The Government Actuary (1998) reports that there are only around 23,000 members of defined benefit employer schemes. The average age of members is 43, average length of membership 8.4 years, with average salary $54,000. In retirement the average pension is $7,500. Thus few New Zealanders are beneficiaries of significant private employment subsidised pensions. The PRG (1997a) noted that while many employers are likely to play some role in the provision of retirement planning 'there has been some question about the extent to which they will continue to offer superannuation itself' (p.183). Government Actuary figures on membership of all occupational schemes shows that there has been a reduction in membership since 1990. Membership of private sector employer schemes has dropped from 21% of all employed people, to 17%. While total assets have increased from $9.5 billion to $9.8 billion after adjusting for inflation, real assets have fallen 8%.

Including the Government Superannuation Fund, which closed to new members in June 1992 [7], total membership of employment based schemes was 25% of all employed people in 1990, dropping to 19% in 1997. Total assets, inclusive of the GSF, were $11.6 billion in 1990, rising to $13.1 billion in 1997 (PRG, p.183).

The reasons for the fall-off in membership and assets have been explored elsewhere (St John, 1998b). The question for society remains: the demise of private pensions is socially desirable? In the light of the long-term health care needs of an ageing population retirement saving in the form of a steady income stream may be an advantage. Annuities or pensions provide a way of sharing the costs of living a long time. Their widespread use is however unlikely in the present New Zealand tax environment. While it may be heresy to state it, but the development of the annuities market will require the support of and regulation by government. It may for example, be necessary to rethink the tax regime described by T/T/E, (contributions taxed, fund earnings taxed, and the final withdrawals exempt or tax free). Perhaps the recommendations of the Brash committee in 1988 for a different, but still neutral tax regime should be reconsidered (Report of the Consultative Committee, 1988). Another possibility is a careful consideration of how the tax credit option discussed in this paper could be used to encourage retirees to translate their lump-sums into annuities. The options are many and it is past time for the debate to begin.

1. To give some indication of the low level of participation in additional means-tested programmes, only 5% of women and 3% of men on New Zealand Superannuation (NZS) receive an accommodation supplement. Just 14% of NZS recipients also receive a small disability allowance (Periodic Report Group, PRG, 1997, pp.36-7)
2. Currently 63 years 6 months, to rise to 65 by 2001.
3. The floor is now 60% of the net average wage for a married couple.
4. For example, a typical charge for a consultation with a GP might be $18-$25 instead of $33-$40. Specialist services are excluded from the subsidy.
5. Taxable income as reported to the IRD, and sample of non filers (PRG 1997).
6. For example, in his column syndicated to the NZ Herald in 1998. An example of Morgan's views is "Scrap the dole, bring in the poor person's benefit", 14 July 1998.
7. Public sector employees may or may not be offered the opportunity to join new defined contribution schemes, subsidised by their organisation. Increasingly, those on employment contracts are not offered such schemes following the introduction Total Remuneration packages. In some cases, Unions have successfully fought the inclusion of an employer subsidy in total remuneration, but these victories may be short-lived as they have depended on interpretation of trust deeds, which can be changed.

www.geocities.com/nzwomen/SusanStJohn/IIRtext199811.html