This article was first published in the first-quarter 2013 edition of Personal Finance magazine.
A major reason people buy an investment-linked living annuity (illa) is the assumption that if they buy a guaranteed annuity and die soon after retirement, “the life assurance company will take my money”.
But Momentum actuaries Mayur Lodhia and Johann Swanepoel say the result of investing in an illa is often that the very people to whom illa pensioners wish to leave money – usually their children – face the prospect of having to support them in the later years of their retirement, because their parents have run out of retirement savings.
They say many pensioners “may fail to realise that an investment in a guaranteed annuity involves the implicit purchase of life insurance”.
An illa can be broken down into two underlying components, one that provides an income while the pensioner is alive and another that pays a life assurance benefit (the residual capital) at death. The sum assured is equal to the outstanding capital, which is expected to decrease over time.
There is a common misunderstanding, Lodhia and Swanepoel say, that a life assurance company profits from the early death of pensioners who receive a guaranteed annuity. In fact, pensioners who survive longer than the average age benefit from those who die sooner than the average age. This is known as mortality pooling.
“The assurer only stands to profit if the entire annuity pool (all the pensioners) die sooner than expected, but the assurers also stand to lose millions of rands for each month the pool survives longer than expected.”
Lodhia and Swanepoel argue that a pensioner who dies before the average life expectancy in retirement (that is, 15 years, to age 80 in their research model) can expect to secure a better return from an illa, because the pensioner plus his or her beneficiaries will recoup the entire initial illa investment, together with the investment returns (less costs), through a combination of income drawdowns during the pensioner’s life and the capital payment at the time of his or her death.
But an illa becomes problematic for pensioners who survive beyond the average life expectancy of 80 in the mortality assumptions in their research model. The illa drawdowns beyond age 80 result in the erosion of capital. This in turn means that the illa is not able to provide either an ongoing real (after-inflation) income stream, or a meaningful death benefit, past a certain age. “This could quite possibly leave the member facing destitution,” Lodhia and Swanepoel say.
Compared with an illa pensioner, a guaranteed annuity fund member needs to survive a certain number of years to recoup his or her investment. In other words, in terms of the research model, if you retire at age 65 with a guaranteed annuity, you will recoup 50 percent of the capital and returns after 6.5 years, and 100 percent after 15 years. Live longer than the 15 years and you get back more than you have paid in from the continued guaranteed income flow – and you do not run out of money.
The researchers say the bequest motive should be addressed by purchasing separate life assurance cover, rather than used as a reason to invest in an illa to provide a retirement income and a benefit for heirs. This will separate the bequest motive from the income you need for retirement.
“In practice, this is seldom done, as members of retirement age find the explicit cost of life insurance too expensive. However, by purchasing an illa, members are implicitly paying the same life insurance premium. The cost of the illa ‘insurance’ is funded by a reduction in expected future income.”
Using the above examples, Lodhia and Swanepoel estimate that the cost of the illa “insurance” is about 24 percent of retirement savings.
“The purchase of an illa means that a pensioner is effectively choosing not to insure the risk of outliving capital,” Lodhia and Swanepoel say. This failure to assure yourself against the longevity risk may be somewhat at odds with levels of risk aversion you display in other areas of life. For example, most people choose to insure against theft or damage to homes and motor vehicles. In these areas, the probability of theft or damage is usually much lower than 50 percent, and the financial impact of an incident is usually less than the financial ruin that results from outliving your retirement capital.
Lodhia and Swanepoel say a likely explanation for the failure of many people in moderate good health to assure against the risk of outliving their capital is that retirement fund members, and arguably some service/product providers and financial advisers, do not fully understand the nature of an illa at the time of purchase, or members overestimate their ability to out-perform a guaranteed annuity and disregard the longevity risk.
With a guaranteed annuity, both longevity and investment risk are transferred to the life assurance company, which promises to pay an inflation-linked income for life, irrespective of the actual lifespan of the pensioner or the investment returns. The life assurer prices the guaranteed product (your pension) on its – usually conservative – estimate of the mortality of all the pensioners in the annuity pool and on prevailing interest rates, Lodhia and Swanepoel say.
“The guaranteed annuity pensioner is not exposed to any investment risk, as the insurer is liable for the agreed income stream irrespective of underlying investment performance,” they say.
Note: You can buy a guaranteed annuity that has a guarantee to pay the pension to your beneficiaries for a particular period. For example, an annuity guaranteed for 10 years and then for life will pay your pension for 10 years, whether or not you are alive during those 10 years. If you survive the 10 years, the pension will continue to be paid until your death, but nothing will be paid to beneficiaries on your death. The longer the period you want for the guaranteed payment of your pension, the more expensive the annuity will be – that is, you will receive a lower pension than you would if there had been no guarantee on the payment period.
PROPOSAL TO LINK DRAWDOWN RATE TO YOUR AGE
Age-related income drawdowns from investment-linked living annuities (illas) and the retirement investment trusts (rits) proposed by National Treasury could be on the cards.
Scaled annual drawdown limits based on age and investment returns are receiving increased support from the financial services industry and government as concern grows over the shift to illas and the often high withdrawal rates.
Currently, the annual withdrawal rates are restricted to between 2.5 percent and 17.5 percent. Initially, the limits were between five percent and 20 percent.
In the United Kingdom, products similar to illas and rits have scaled drawdown limits based on both a pensioner’s age and the investment returns, which in turn are based on the gilt index yield (top risk-rated bonds). For example, a 65-year-old pensioner in an investment market-linked product that provides a return of six percent a year may withdraw a maximum pension of eight percent of his or her residual retirement savings. If the rate of return over the year is eight percent, the maximum pension for the 65-year-old rises to 9.5 percent.
A pensioner aged 80 is permitted to withdraw a maximum of 12.9 percent when the returns for the past year were six percent, or 14.4 percent if the returns were eight percent. The maximum drawdown rates are reached at age 85.
The reason for restricting the initial drawdown rates is to reduce the chances of the pensioner not having sufficient income later in life to avoid impoverishment. An age-related increase in the drawdown rate is allowed because of the need to counter inflation, and, as the pensioner grows older, the threat that he or she will outlive his or her capital abates.
National Treasury and the South African Revenue Service are also considering proposals that the minimum illa drawdown rate of 2.5 percent be reduced to zero. A nil minimum drawdown rate will mean that pensioners with alternative sources of income, particularly members of occupational retirement funds who retire on fixed dates, will not be forced to withdraw, from an illa, an income they may not need.
In 2008, the body that represented the life assurance industry, the Life Offices’ Association, in reaction to growing concerns about the high withdrawal rates that pensioners were selecting, recommended scaled age- and gender-related drawdown rates for illas.
The maximum indicative rates started at 5.5 percent for men aged 55, rising to 6.2 percent at age 60, and then in five-year age brackets to age 85, where the drawdown rate was 17.5 percent. For women, who are expected to live longer than men on average, the rates were set slightly lower.
In March last year, financial services industry organisation, the Association for Savings & Investment SA (Asisa), lowered the indicative rates after research by the Actuarial Society of South Africa showed that a rate above five percent in the early years of retirement is dangerous. The new indicative table (see link below), which is based on drawdown rates and investment returns, shows how long it will take before a pensioner is likely to reach the maximum drawdown rate of 17.5 percent and thereafter experience a reduction in his or her income in real (after-inflation) terms.
In their presentation on pension choices to the Actuarial Society last year, actuaries Mayur Lodhia and Johann Swanepoel said the new way of illustrating the effect of drawdown rates is confusing for pensioners, and they recommended that Asisa revert to the former indicative tables.