Issue: December 2012 / February 2013


Action plans

Turn them this way or that, agree with some proposals or disagree with others, but there’s no way that National Treasury can be faulted either for the detail in its discussion papers on promoting household savings and retirement reform; for the research and analysis that lie behind them, and for the lucidity of their presentation.

Make no mistake, however, that they have the makings of a small tsunami for retirement-fund service providers and customers. For all the regulatory interventions that await, few seem undue or unnecessary conflicts with free-market principles: to enhance consumer protection and improve the user-friendliness of financial products so that, through a carefully-crafted mix of carrots and sticks, the levels of household and hence national savings will strengthen to the benefit of individuals and the country as a whole.

The basis has been laid for exhaustive consultation and debate on the four papers so far released, providing meat to the bone of the Strengthening Retirement Savings: Overview of the 2012 Budget Proposals document put out in May. The period for comments on it closed at end-July. These papers are the sequel. Constant themes relate to product costs and ways to improve the net after-charges amounts that ultimately reach the pockets of product purchasers.

The four papers are on enabling a better income in retirement (Paper B); preservation, portability and governance for retirement funds (Paper C); incentivising non-retirement savings (Paper D), and improving tax incentives for retirement savings (Paper E). A fifth (Paper A), on retirement-fund costs, is scheduled to follow.

Industry leaders will doubtless respond vigorously, not least because the modus operandi and bottom lines of service providers stand to be impacted for better or worse. But the papers deserve also to be studied by a broader range of industry participants, notably fund trustees (who’re supposed to help members receive advice) and financial advisors (whose livelihoods can be affected), simply for a clearer understanding of retirement issues.

And with these papers being quantum steps for near-term reforms, they set the path for more holistic reforms over the longer term. That said, the four papers (available on National Treasury’s website) should be read as a whole rather than nit-picked. Otherwise, context is lost and the landscape obscured.


On better retirement income

National Treasury reviews the annuities market, and finds it wanting.

When most members of pension funds retire, suggests this paper, they often choose (on bad advice) an inappropriate annuity product that leaves them even more vulnerable as they age and are no longer able to earn their own incomes.

The value of the annuities market has grown from about R8bn in 2003 to R31bn in 2011. In 2003, 50% of single premiums were used to buy conventional life annuities (the only products offering longevity protection). By 2011 this number had fallen to 14%.

The paper looks at the products’ respective merits and flaws because, it notes, many employees when they retire are left to the retail market. Here they must bear the risks of retirement on their own. These include “the risks of poor or commercially-biased financial advice and high charges”.

Examined are living annuities:

Essentially investment accounts provided by life insurance companies, the policies must pay an income of between 2,5% and 17,5% of the account value to the investor annually. They’re complex products.

Individuals who buy them must make and continually review decisions that involve difficult trade-offs e.g. how much income to drawn down, in what underlying assets to invest and which provider to choose.

Charges appear to be “very high”. The median level, excluding guarantee charges and performance fees, appears to be more than 2% of individual balances per year, and may be much higher. Annual charges of 2% represent 40% of the income than an individual is drawing and will consume 20% of the policy’s value over its life.

As for conventional annuities:

Only about 20% of retirees choose them. Purchases appear to be driven strongly by short-term considerations and sales incentives.

In particular, it adds, the commission earned by brokers for selling a living annuity may be up to 10 times larger over the product’s life than the commission for selling a conventional annuity.

Large market players rate individual purchasers of conventional annuities only by age and sex: “As a result, poorer individuals, or those who are ill, may elect not to purchase conventional annuities because they represent relatively poor value. For those who do purchase them, conventional annuities appear to be reasonably priced. Value for money may be lower for those buying cpi-linked products.”

It points out that holders of living annuities are subject to a host of charges, most being paid from their fund balances: costs of financial advice and brokerage fees; platform fees to the provider; asset management and performance fees to the asset manager; audit fees, trustee fees, brokerage, vat and securities transfer taxes. These reduce the returns on the underlying funds and “exert a substantial effect” on the income that individuals can obtain from their living annuities.

Under consideration by National Treasury are certain policy options that can reduce the costs of living annuities and the amount of financial advice required:

  • Create a new tax-free vehicle based on collective-investment schemes. It will not permit investment choice, although individuals could choose between different underlying investments and switch from one vehicle to another;
  • Restrictions on permitted drawdowns will remain. They must incorporate all charges and may be age-related;
  • Commissions for intermediaries will be more strictly regulated;
  • To increase assistance to retirees, being considered is that all funds select a default retirement-income product (conventional annuities can qualify) to which members are automatically enrolled but may opt out if they wish. All default options should include a minimum degree of longevity protection.

Under some of these options, says the paper, insurance companies may struggle to bear significantly more long-term risk at a reasonable price: “Government is willing to explore ways in which (these) risks borne by the private sector can be shared by the public sector”. A possibility is reinsurance through selling longevity-linked bonds.


On preservation, portability and governance

Treading carefully into certain sensitive areas.

Unchallengeable is the proposition that a lack of preservation – employees cashing in their pensions when they switch jobs – is the single greatest enemy of retirement-savings accumulation. For one thing, it defeats the glorious miracle of compounding. For another, it throws out the whole purpose of funding for the long term.

Preservation of benefits until retirement age should be a no-brainer. In fact, it isn’t. This is most forcefully evidenced where dismissed mineworkers at Marikana have resisted reinstatement because they want immediate cash, from their pensions, to pay off the loan sharks who’ve entrapped them.

To make preservation mandatory is far from easy, given its troubled history (TT Cover Story Sept-Nov ‘12). National Treasury is treading gently. Such are the exceptions and allowances that the version proposed is Mandatory Lite.

Importantly, accrued or vested rights will be protected. In other words, current fund rules will apply even after the legislation takes effect. So those who’re now members of funds which allow them to cash in on leaving a job – the common practice – are unaffected.

Once the legislation is implemented, new fund members who’re retrenched will still be able to withdraw “sufficient funds”. Options being considered are:

  • Allowing full access to funds when leaving employment, but levying a tax that would be a disincentive to withdrawal;
  • Permitting partial access to a cash lump-sum before retirement, but requiring preservation for the remainder;
  • Monthly withdrawals for individuals unable to find new employment;
  • Full preservation of existing assets’ growth or of new contributions.

Retirement funds, its says, will be required to “nudge” members to save for the long term by creating “appropriate defaults biased towards saving”. For example, funds would have preservation sections. When members leave their jobs, their savings would be defaulted into such a section unless they indicate otherwise. (This is clever because the lethargy pervasive amongst members inclines them towards default options.) And to promote great portability – the ability of an employee to transfer savings from one fund to another – all funds will be required to accept accumulated savings of new employees.

On the governance of pension funds, it’s noted that “many trustees may lack the competence and necessary skills” to act in the best interests of members. Accordingly, compulsory trustee training is being considered and the elevation of Financial Services Board circular PF 130 (on good governance of funds) to a directive status is proposed. It would then be legally enforceable.

Further to improve governance and management, there’s the prospect of the principal officer’s role being professionalised: “This will possibly entail having the principal officer play an executive role on the (fund’s) board, with responsibility for the day-to-day running of the pension fund. Accountability, however, will always remain with the board of trustees.”

National Treasury’s reform trio . . . Ismail Momoniat, Olano Makhubela and David McCarthy
Photo: Finanacial Mail


On non-retirement savings

Need to stimulate individual household savings and the national savings rate.

It will be off a base that’s pitifully low: so low that, at the individual level, far too many people invite old-age penury mitigated only by reliance on the state; at the national level, economic growth is constrained by inadequate capacity for investment.

Part of the problem, to incentivise discretionary non-retirement saving, is that SA’s tax-free interest-income thresholds are not sufficiently visible and restrict individuals’ investment options to those that bear interest.

National Treasury proposes that the current thresholds be increased by replacing it with a broader tax-incentivised vehicle that comprises two types of accounts:

  • Interest-bearing accounts which may invest in bank deposits, retail saving bonds or interest-bearing collective investment schemes (CISs) such as money-market funds;
  • Equity accounts which may invest in CISs that hold JSE-listed equities or directly own property.

Within these tax-preferred vehicles, earnings and capital growth will be exempted from income tax and contributions made from after-tax income will be capped. For both components combined, the proposed limits per individual are R30 000 annually and R500 000 over a liftetime (periodically adjusted for inflation). There may also be “transition mechanisms” such as allowing taxpayers to invest more of their lifetime limits, the older they get, beyond the annual caps.

The idea is clearly to encourage households – really, households with lower to middle taxable incomes – to save through the tax system. Present thresholds, notes the paper, cost the fiscus over R3bn a year. The proposed thresholds will presumably cost it considerably more, provided they work, but at a cost well worth it in terms of the advantages to the national economy of a deepened savings pool for domestic investment that in turn has the potential to increase tax revenues.

Why do people not save? The paper cites behavioural-economics research that “people often make decisions which do not appear to be in their best interests”. They procrastinate, putting off things such as saving for retirement. They stick with the default option, even if it isn’t the best. They avoid decisions that are too complex. They’re readily confused and prone to accepting misleading advice.

Behavioural economics also suggests:

The sole focus on the impact of tax on the rate of return is no longer appropriate in considering the design of policies to encourage savings;

The need is rather product design and savings incentives that can overcome problems of self-control in individuals’ savings decisions and limited financial-planning skills;

Essentially, savings are a self-control problem driven by the strong bias in preference towards present as opposed to future consumption.

These are practical realities of which National Treasury is keenly aware. As a means to address them, the proposed tax-favoured vehicles for non-retirement savings “will have to be implemented along with a range of other measures”. These include “more transparency in financial-product operation, measures to reduce the costs of savings products, as well as educational campaigns to improve financial literacy and the savings culture”.

As measures to promote preservation of retirement savings are introduced, the paper points out, these new accounts will provide an alternative for short to medium-term saving to meet consumption needs. A consultation period on the design options and phasing-in will precede the finalisation of these proposals.


On improved tax incentives

Ways to establish a simplified and uniform model for retirement savings.

At present there are separate dispensations for the tax treatment of pension, provident and retirement-annuity funds, both in the contributions to them and the benefits from them.

As a business expense against tax, employer taxpayers can deduct contributions to pension or provident funds of 10%-20% of the ‘approved remuneration’ of employees. In the case of tax-exempt entities, there are effectively no limits on the size of deduction. These contributions aren’t part of employees’ taxable incomes.

For their part, employee taxpayers may not claim a deduction on their own contributions to provident funds. But on pension funds they may claim a deduction on contributions to a 7,5% maximum of ‘retirement-funding employment’ income. With a retirement-annuity fund, a maximum 15% deduction of ‘non-retirement-funding employment’ income may be claimed against contributions.

This tax regime, where the different concepts are defined, is highly complex. It requires administrators to monitor the original dispensation under which contributions were made, the paper notes, and restricts movement between different fund types.

Two further problems are identified:

  • The tax emption has no nominal monetary cap in the case of higher-income employees. This allows them to make tax-exempt contributions “way in excess of the amount required to maintain a reasonable standard of living in retirement”;
  • Tax-exempt employers are able to assist employees in postponing tax by making large contributions to pension or provident funds rather than paying by cash income.

Expanding on proposals in the 2011 and 2012 Budgets, the paper refers to three distinct tax calculations although in practice the same value is attributed to ‘approved remuneration’ and ‘retirement-funding employment’ income. Rules of retirement funds typically define ‘pensionable salary’ for purposes of contributions made by employer and employee, as well as (where applicable) the value of the benefit payable for fund-provided risk benefits.

Normally, where a retirement fund has only one contributing employer, the fund rules may define the actual ‘pensionable salary’ (income from ‘retirement-funding employment) as fixed remuneration (salaries and wages) only. This would exclude such variables as commissions, bonuses and overtime. With umbrella funds, the rules normally allow the contributing employer to determine the components and value of the ‘pensionable salary’.

Respective contributions to a fund of the employer and employee are based on the employee’s salary, but in most cases not on the full salary. So many individuals can, where they contribute to a retirement-annuity fund, additionally claim a tax deduction because a portion of their salary is not taken into account when calculating the employee’s and employer’s contributions.

A more effective tax regime should be broad-based, easy to understand and administer, restrict abuse and be equitable so that incentives don’t advantage some taxpayers over others, and be measurable in terms of the incentives having the desired impacts.

To these ends, one proposal is that there should be no adjustments in calculating ‘remuneration’ for the employment-income leg: “By allowing individuals to use their ‘remuneration’ as a base, the regime is simplified for employees as well as for the employers that have to process the deductions. The benefit that will result from an exact calculation of employee income does not justify the administrative cost that such a calculation would necessitate.”

With any tax regime at any time, success relies on simplicity for understanding and administration as much as it does on perceptions of fairness and equity. These are the clear goals National Treasury has set out to achieve. In the process, there’s reason to hope, they’ll help to kick up SA’s household savings from their dangerously dismal levels.