Publications
Partners
What do you need to know about this country? Forty-five million people, an emerging economy, some would say middle-income, others not, certainly very sophisticated in services, but with pockets of devastating poverty as well. In fact, South Africa has one of the highest levels of socioeconomic inequality in the world.
It provides well for the very poor through a system of social grants. Private sector provision for the middle- to upper-income isn't too bad but there are a number of problems to sort out... markets sometimes need encouragement to work effectively.
SA's biggest problem is in the middle: above the unemployed, below the comparatively wealthy. Low- to middle-income workers don't save effectively for retirement. Anecdotal evidence suggests that anyone with an interrupted career is unlikely to provide well for those zero-income years of old age.
Roadmap
Examining this assignment you'll recognise two fundamental objectives: saving and redistribution. Identifying them and measuring them are rather different problems and many other goals should receive proper attention alongside those. And you'll identify three design trade-offs in the way in which you set up the system that affects the success with which you meet your goals: the locus of risk, the extent of the funding and the issue of management.
Saving
Why do governments want their citizens to save? Every one of us, unless unfortunate enough to die prematurely, goes through three stages of life: a growing phase, a working phase and a weakening phase. The increase in the average lifespan doesn't change the reality of those phases. Those who don't make adequate provision during their working life for the time that follows that must find resources elsewhere.
Governments wouldn't like to be the sole source of those resources, so it pays policymakers to encourage citizens to save. That's why the vast majority of governments literally pay their citizens to save - most frequently through tax incentives, occasionally by adding directly to the savings made by their people.
Most economists agree that encouraging household saving is the right thing for the policymaker to do, because strong saving brings economic benefit.
Another advantage of the government incentive is that it draws the savings of citizens into a supposedly safe environment.
As co-sponsor - which is exactly what government is - the policymaker has the right and obligation to safeguard those savings, imposing conditions on private sector entities that enhance the security and investment returns on deposited amounts.
It's in the interest of a government for citizens to save for retirement and most countries back that with finance. The notable exception to that is New Zealand, which provides a flat-rate pension to all citizens and doesn't provide any tax incentives to save. However, on the back of concerns that household savings rates are low it's introducing a saving arrangement with a once-off government subsidy.
Redistribution
For both moral and practical reasons every government looks after the poor. Policymakers do that in a number of ways. All governments play Robin Hood to a greater or lesser extent, taking more in taxes from the wealthy than the poor, systematically redistributing resources across the socioeconomic spectrum in the process.
Redistribution is also a key goal in old age policy. Some can't save for retirement and government has a form of obligation to provide for those people. Some will not.
While most take the view that redistribution is correct, differences on its appropriate extent are sharp, often dividing individuals and groups on ideological grounds. That's unnecessary and unfortunate, particularly in the absence of proper measurement. We must strive for objectivity, recognising the complexity of the issues and working harder to identify and measure the key parameters that demonstrate the extent of redistribution in the system.
Total national system cost, tax revenue foregone, projected individual retirement saving and replacement ratios are all difficult to determine. But in the absence of clear measures how can we formulate clear policy?
So redistribution and saving are the key goals, supported by the respective benchmarks - an objective without a yardstick isn't helpful - of minimum income in retirement and some measure of projected retirement income relative to working income. There are other goals, which should not be omitted in the question for simplicity.
Design must be a means to an end and the end must be framed with the goals in mind, supported by proper benchmarking.
Locus of risk
The DB/DC issue is hardly a debate in SA any longer. That's a little unfortunate, because the denial of a sensible design option is restrictive, particularly on an issue that so profoundly impacts the risk borne by our citizens. Hybrid arrangements that allow risk-sharing between employers and fund members have been effectively ruled out because the surplus legislation classifies them with defined benefit funds.
The reduction of corporate defined benefit funds and Government social security in favour of defined contribution alternatives is ubiquitous. That transformation is evident in widespread restructuring of corporate arrangements. It shows itself in the dramatic modification of national systems throughout Spanish-speaking Latin America - Brazil being a notable exception - and in the former Russia-aligned states of Eastern Europe and central Asia.
Australia has introduced mandatory individual account saving; Sweden has added a small individual account component to the national system; China and India are introducing change; Nigeria has launched compulsory saving to personal accounts; Britain is giving it serious consideration. The list is almost endless.
This process of change has transferred enormous risks from institutions to individuals. While the recipients may benefit from that transfer they may also lose out.
It's puzzling that SA's policymakers appear to have decided that individuals should bear much of the risk of saving for their retirement, despite the fact that Government is a strong risk-taker in the defined benefit arrangement provided to public servants.
Funding
Not all old age plans are funded. Many social security arrangements are completely unfunded, while others have some asset backing and could be described as partially funded. Is the absence of funding a risk? Absolutely, because today's workers are paying the pensions of today's elderly and shifts in the ratio of workers to pensioners present significant financial problems. But that can be managed when the system is set up, through adjusting the retirement age to reflect changes to the balance, for example.
Many European policymakers wish they could turn the clock back to the different world into which those promises were delivered. If they could do it again they'd most certainly design pressure valves into the system, like a formula-driven flexible retirement age.
There are two problems with the illusory solution of funding. It can present a veil of security. If a significant proportion of the assets are invested in gilts then government is borrowing from today's workers and might as well not be pre-funding that portion of the fund anyway.
The second is that funding is expensive. Total expense ratios show private sector retail charges for equity management running at between 2,2% and 2,5%/year of assets. That's a high price for participants to pay for the security of funding, particularly if part of it's illusory.
Still on the issue of funding, dangerous assumptions are common. For example, that a DC arrangement must be funded. At a company level, perhaps, but at national level most definitely not. National DC systems occur in a number of countries, from Sweden to Mongolia via Italy and Russia, and they contain the risks without incurring the expense of pre-funding.
Challenge the misconceptions. An unfunded system is really a massive loan from workers to government, repaid on retirement. That's dangerous. But funding isn't all that it seems to be either.
Management
Most would agree that the public sector provides scale but poor motivation for efficiency, while the competitive dynamics in the private sector incentivise inefficiency but undermine scale. Most would also concede that competition doesn't consistently minimise price: plenty of examples present themselves in SA's financial services industry.
Despite that it's hard to find anybody able to apply their minds objectively to the issue of where pension savings should be managed.
Surely it should be possible to combine the best of both worlds, producing an outcome that's in the best interests of savers?
Fortunately, many lessons are available to us that demonstrate potential solutions to the conundrum. Scale can be obtained by collecting contributions centrally and permitting participants the option to redirect them to private sector entities.
That can be improved further by the introduction of "blind accounts" - under which private managers don't know whose money they're managing, removing much of the marketing incentive.
Further cost efficiency is theoretically obtainable through auctioning the right to provide investment management services to savers.
None of these are pie in the sky: they all exist, in a stable form, in other parts of the world and are being considered by policymakers in a variety of countries.
And that means they're on the desks of SA's public servants, carefully contemplating the way forward best for all South Africans.
Private sector players need to help, but they must get away from false perceptions if they're to do so effectively.
For example, arguing against centralised collection of compulsory contributions purely on the principle that this is private sector domain isn't going to cut it.
Demonstrate that it's not in the interests of SA's citizens or plan for its introduction in some form.
Appeal for objectivity
Back to the assignment: at its core straightforward - identify and prioritise the objectives and resource constraints; for each design option, estimate the extent to which they would be met; and implement the optimal system as efficiently as possible.
SA is deep in the discussion but sadly lacking concerning the fundamentals.
As far as I can tell we haven't properly identified the goals or established objective measurement mechanisms. Is the openness to considering options with objectivity out of reach or do we just need to try harder?
More serious thinking is needed.
Rob Rusconi
RUSCONI consults and researches in SA and overseas on policy in pensions, social security and contractual savings. He's an actuary with working experience in healthcare, insurance product development, pensions and investments in SA and Britain.
