This article was first published in the first-quarter 2013 edition of Personal Finance magazine.
Your first objective, when you purchase a pension with your retirement savings, should be a constant real income – which means an income level that rises in line with inflation, according to Momentum actuaries Mayur Lodhia and Johann Swanepoel.
In a research paper presented at the Actuarial Society of South Africa’s convention late last year, Lodhia and Swanepoel compared investment-linked living annuities (illas) and traditional life assurance guaranteed annuities as vehicles for achieving a sustainable income in retirement.
In a nutshell, their findings show that a guaranteed pension is, in most cases, the better choice to provide the required level of income.
“The dangers of illas have started to emerge,” they say, “with increasing instances of retirees outliving their retirement capital.”
Lodhia and Swanepoel also turn on its head the conventional wisdom that, in the current environment of low interest rates, it is best to delay buying a guaranteed annuity in the hope that interest rates will improve and that buying at an older age – and thus reducing the period in retirement – will result in a better guaranteed pension.
However, the actuaries acknowledge that many retirees select illas because they allow them to choose the underlying investments, in the belief or hope that choosing the right investments will enable them to make up for a shortfall in their retirement savings and give them returns that outstrip the rate of inflation and the rate at which they draw down an income from their savings.
The problem is that illa pensioners often mistakenly assume they can afford to draw down a high income initially (higher than they would receive from a guaranteed annuity) and that strong future investment returns will compensate for the shortfall created by the high drawdown.
But research undertaken by various people and published in previous issues of Personal Finance magazine shows that an initial annual drawdown rate above five percent is likely to result in impoverishment, particularly if the pensioner lives beyond the age of 80.
The Lodhia/Swanepoel research compares the capacity of living and guaranteed annuities to provide a minimum after-inflation income for life. It also quantifies the implicit cost of life assurance that is embedded in a living annuity and the associated impact on retirement income.
The researchers say the benefits of illas are evident if a pensioner dies soon after retirement, in which case the remaining capital is preserved for his or her dependants.
“Illas present better value for impaired lives (unhealthy pensioners) at retirement,” they say.
“However, the high take-up of illas implies that sales have not been limited to impaired lives and that many pensioners in average to above-average health depend on illas to provide a sustainable income for life.
“In addition, the low savings rate in South Africa implies that most retirees are fully dependent on their retirement savings for a post-retirement income – that is, they have little other provision to rely on,” the researchers say.
The main thrust of Lodhia and Swanepoel’s research was to consider how illas and guaranteed annuities compare in terms of:
* A straightforward choice between the two at retirement at age 65 (see “Seeking out-performance can pose a risk”, below); and
* The option of investing in an illa at retirement at age 65 and then switching to a guaranteed annuity at age 75 (see “Delaying the switch until your 70s ‘does not pay off’”, below).
In both cases, the researchers found it was better to opt for a guaranteed annuity immediately at retirement. They based their calculations on a man with no dependants and with R1 million in retirement savings at the time of retirement. The research model assumed a nominal (before-inflation) interest rate (investment return) of eight percent and inflation of 5.5 percent, providing a real investment return of 2.5 percent. The model ignored costs, but Lodhia and Swanepoel say it would have favoured guaranteed annuities if costs had been included.
The guaranteed annuity they selected provided annual payments in arrears that increased at the rate of inflation, but it excluded joint-life options and guarantee periods.
The assumed level of the initial income for both the illa and the guaranteed annuity was the guaranteed annuity rate at age 65, which in turn depended on the prevailing interest rate and a life assurer’s assumptions about how long, on average, the pensioner could expect to live in retirement.
The analysis highlights the impact of the annual illa drawdown cap of 17.5 percent and what the researchers say is the often overlooked benefit of mortality pooling embedded in a guaranteed annuity. The cap is an attempt to preserve the capital and ensure that the income lasts for longer, Lodhia and Swanepoel say.
They say “in practice, the cap does tend to preserve capital, but at the expense of the illa’s ability to provide an adequate retirement income for life. Once pensioners reach the cap, they will experience a reduction in retirement income.”
The exception is the unlikely situation in which the capital accrues investment returns greater than the drawdown limit of 17.5 percent.
Lodhia and Swanepoel say their research model shows income from the illa increasing as the drawdown percentages are adjusted to keep pace with inflation, and increasing compared with the income provided by the guaranteed annuity.
But as the percentage of the income drawdown increases, the capital is eroded, so the required percentage drawdown is larger in later years. At about age 75, the income requirement as a percentage of the capital balance in the illa is likely to exceed the drawdown limit of 17.5 percent.
The pensioner cannot draw down more than 17.5 percent, so he or she will experience an annual reduction in the income provided by the illa, whereas a guaranteed annuity would have provided inflation-linked increases. And the illa income decreases in nominal terms, “which makes matters worse in real terms, and the illa pensioner will not be able to maintain his standard of living”, Lodhia and Swanepoel warn.
“By contrast, a guaranteed annuity is not subject to a limit and is therefore able to provide an income that consistently increases with inflation.”
For the same upfront investment and the same net investment returns, a guaranteed annuity can provide a much more attractive income stream than an illa, because there is no drawdown limit on a guaranteed annuity.
A guaranteed annuity is designed to return all the capital and investment returns, less expenses, by the expected age of death (the average life expectancy of all the pensioners in the risk pool at the time the annuity is purchased). In effect, a guaranteed annuity allows a pensioner to draw down up to 100 percent of the actuarial reserve and more than 100 percent of the initial capital investment in later years, without the risk of running out of money.
“Guaranteed annuity pensioners who live longer than expected effectively recoup more than 100 percent of their initial investment, plus returns, less costs, while the maximum recouped from an illa investment is 100 percent, plus returns, less costs,” Lodhia and Swanepoel say.
SEEKING OUT-PERFORMANCE CAN POSE A RISK
Pensioners are easily persuaded to opt for an investment-linked living annuity (illa) at retirement by the argument that a well-constructed, balanced investment portfolio, with a diversified equity component and managed by a well-known portfolio manager, should out-perform the income provided by a guaranteed annuity in the long run.
One reason this argument is so persuasive is that, typically, the underlying assets of an inflation-linked guaranteed annuity are invested in a low-risk, cash flow-matched, inflation-linked strategy, research by actuaries Mayur Lodhia and Johann Swanepoel shows.
They say the argument is flawed, because it overlooks the “returns” provided by the mortality pooling of pensioners within a guaranteed annuity portfolio.
Although it is possible for an illa to out-perform, this is not without risk, they say, and many pensioners cannot afford to gamble with their retirement savings.
Lodhia and Swanepoel cast doubt, too, on the belief that pensioners can receive better returns, even in the short term, by opting for an illa with the intention of switching to a guaranteed annuity later. This has its dangers, they say, because there is no “sweet spot” at which it is optimal to switch from an illa to a guaranteed annuity. Instead, their research shows “it is best to annuitise at age 65”.
Lodhia and Swanepoel’s calculations show, for example, that a retiree who defers buying a guaranteed annuity for 10 years (to age 75) and starts at an income level from an illa that would be provided by a guaranteed annuity would need to generate a total return on their capital of 11 percent in each year of the 10 to break even with the guaranteed annuity when the pensioner switches at age 75. “This translates into a return of inflation plus 5.5 percent for every year of deferment without risk. This is a very high return target to achieve consistently and impossible to achieve without exposing an investment portfolio to high risk.”
Lodhia and Swanepoel concede that, over the very long term, equities have provided an average annual return of inflation plus seven percent – with associated high levels of volatility. But they say that, arguably, the global financial crisis has caused return expectations to be revised downward.
A proper comparison, they say, should also be based on the investments that underlie an illa achieving risk-free out-performance. This is because the guaranteed annuity achieves its results without exposing a pensioner to any investment risk. The risk is borne by the life assurer.
“In reality, this [risk-free out-performance] is not achievable; in a no-arbitrage world, risk-taking is a fundamental requirement for investment out-performance,” they say.
Swanepoel and Lodhia say that, to do better with an illa than with a guaranteed annuity, a pensioner who survives for 10 years after retirement will have to take on almost the same level of risk as an investment portfolio that is 80-percent invested in equities and 20 percent in interest-earning investments. This exceeds the 75-percent limit on equities contained in the prudential investment requirements for retirement funds and which is used on a voluntary basis for illa investments.
National Treasury is proposing that stricter, lower-risk prudential investment regulations be applied to illa investments than those applied to retirement funds, to reduce volatility risk.
Swanepoel and Lodhia say the high investment risks that will have to be taken to match or out-perform a guaranteed annuity “carry an associated emotional burden. Retirees would need to take bets that place their financial well-being at risk, at an age where they are arguably less equipped to make such calls.”
And retirees will go through all this stress and strain only to end up in the same position in which they would have been had they bought a guaranteed annuity at retirement, the researchers say.
DELAYING THE SWITCH UNTIL YOUR 70s ‘DOES NOT PAY OFF’
It is a high-risk strategy to use your tax-incentivised retirement savings to purchase an investment-linked living annuity (illa) with a view to switching to a guaranteed annuity in your 70s, in the hope that interest rates and, along with them, guaranteed pension rates will rise.
The amount you receive as a guaranteed annuity is determined mainly by:
* Interest rates. Life assurance companies invest a large proportion of your money in long-term bonds (money lent mainly to governments, utilities – such as Eskom – and corporations). If interest rates are low, guaranteed annuity rates will also be proportionally lower than they would be when long-term interest rates are high.
* Age. The older you are, the greater the income a life company is prepared to pay you as a pension, because, on average, you are expected to live, and to receive a pension, for a shorter period than in the case of someone who is younger.
* Gender. Men, on average, have a shorter lifespan than women, so men receive larger guaranteed pensions than do women.
In the current low-interest rate environment, many retirees are investing their retirement savings in illas with the intention of switching to a guaranteed annuity in future, hoping that interest rates will have improved and that being older will work to their advantage. But Momentum actuaries Mayur Lodhia and Johann Swanepoel have found that many things will have to go your way for this strategy to work.
Based on various assumptions, you need to achieve the following to break even on a switch from an illa to a guaranteed annuity 10 years after retirement at age 65:
* Earn an investment return of inflation plus 5.5 percent for each of the 10 years without taking on any additional risk; or
* Take an illa pension at 65 that will provide an income that is 32-percent lower than the guaranteed annuity; or
* Start with capital of R1.2 million for an illa, rather than the R1 million used to buy the guaranteed annuity, to provide the same level of income; or
* See an increase in interest rates from 2.5 percentage points to 13.2 percentage points over the 10 years); or
* A combination of all of the above.
Deferring the switch for five years would require interest rates to rise by inflation plus two percentage points, while an increase of 87 percent would be required to break even if the switch were deferred for 14 years.
“A pensioner who defers purchasing a guaranteed annuity by more than 14 years cannot rely on an increase in interest rates to break even with what they would be receiving from a guaranteed annuity,” Lodhia and Swanepoel say.
They say an illa pensioner could possibly expect to break even by deferring purchasing a guaranteed annuity for two to three years after retirement, but this is only if interest rates rise in line with market expectations. A pensioner who opts for an illa initially must bear the risk of adverse movements in interest rates.
“Recent trends overseas indicate that not only could the low interest rate environment persist, but real yields could, in fact, decrease to levels previously thought unimaginable,” Lodhia and Swanepoel warn.