Retirement fund specialist Alexander Forbes is challenging the conventional wisdom that you have to save an amount equal to about 12 percent of your annual income for 40 years from the day you start work until the day your retire in order to receive a pension equal to about 75 percent of your final pay cheque.
Instead, it is advocating a more lifecycle-orientated approach where you can start off with a lower amount when young and then increase this later in order to smooth consumption. At a younger age the extra amount can be used to ensure better protection of your dependants and your human capital and to take care of some of your other needs.
In adopting such an approach, various factors need to be balanced, and people's behavioural tendencies should also be taken into account to ensure a good outcome.
In its latest Member Watch research, Alexander Forbes shows that, for every R100 in pension income you receive in retirement, R65 will be from money you saved (with the investment returns) before the age of 45.
This illustrates that if you do delay saving until later in life you will need to save a considerably higher proportion of your income and/or you will need to continue working for longer to ensure you will be financially secure in retirement (see graphs 1 and 2; links at the end of this article).
There is a point when delaying becomes unrealistic and people succumb to behavioural problems. Hence, one needs to balance these factors to ensure a more smoothed approach over time.
Alexander Forbes’s challenge to conventional wisdom is not new, but it is the first time a major company has taken up the cause. Challenging the way we think about retirement planning was first suggested by actuarial academic Anthony Asher in a paper titled “Pension benefit design: flexibility and the integration of insurance over the life cycle”, presented at the 2007 convention of the Actuarial Society of South Africa.
In his paper, Asher said: "It also does not follow that significant savings contributions are necessary early in the life cycle … retirement saving should often not begin until the age of 40 or later. Before that time, people are better off repaying the mortgages on their homes. The costs of borrowing on a home loan and simultaneously investing through a retirement fund are significant."
Asher suggested that the popular view that you should save lots and early on for retirement is misguided. His research showed that couples will usually be liquidity-constrained when they are trying both to buy a house and raise children.
If they want to enjoy an even level of consumption spending over their lifetime, they should save less when there are real pressures on their income.
“A provocative alternative would be to suggest 60 as the age after which compulsion (of contributions) would apply. Most people are capable of working until 70, and the annuity rate for a man of 70 is about 10 times the annual annuity (pension) at age 60. This means that a single male could live on 50 percent of his income for the rest of his life if he contributed 50 percent for 10 years from age 60,” Asher said.
John Anderson, the national head of consulting strategy at Alexander Forbes, says Asher's statements, although probably considered quite radical, illustrate the point that, given sufficient flexibility (and compulsion in most cases), it “is possible to save sufficient retirement income protection starting later in life if one has accumulated assets or paid off debt”.
Anderson says it is not that people would not be saving; they would be protecting their dependants and their human capital and building up their asset base by, among other things, paying off a home loan.
The question the Alexander Forbes research asks is whether forcing people to save for retirement early on at the expense of protecting their human capital and creating an alternative asset base (such as property) is the right thing to do.
The alternative would be flexibility in contribution rates (as a percentage of your pensionable income) in a defined contribution retirement environment, allowing members the option of smoothing their consumption. Or put another way, you maintain the same standard of living, switching the money you used to pay for bringing up children and paying off your home loan to saving for retirement.
But Anderson says the counter-argument is that those maximising take-home pay simply adjust their lifestyles upwards (now and into the future) and would not switch to saving for retirement when they have built other assets and their children are no longer financially dependent.
This is compounded by the argument that occupational retirement fund members “do not have the financial literacy to make such decisions”. But Anderson says this can be partially resolved by the intervention of both the government and employers.
He says the government should lift the proposed new cap on retirement savings from the suggested 22.5 percent of taxable income in the older age brackets to encourage older people to save more for retirement and for those who started saving later in life to catch up; and employers could structure pay increases so that a growing portion of the increases in later years go towards retirement savings rather than improving the living standards of the employee approaching retirement.
Anderson says the conventional wisdom of having your retirement savings rate as a fixed percentage of your income over your employed lifetime must be questioned.
“Possibly a flexible approach, perhaps defaulted, could achieve better outcomes, adapt to lifecycle needs and dynamically adjust contributions based on needs over time with appropriate automatic mechanisms that help members do the right things.”
He says variable contribution rates can be supported for more than one reason.
Anderson says there is a growing problem that is highlighted by other Alexander Forbes research, namely salary increases given by employers to retirement fund members and the impact on their net replacement ratio (NRR) at retirement.
The average retirement fund return (median return of the Alexander Forbes Large Manager Watch Best Investment View) was 15.2 percent a year for the five years ending December 2010. Inflation over the period was about 6.9 percent a year. This equates to even an average retirement fund, compliant with prudential investment regulations, earning a real return of over eight percent a year.
Retirement fund trustees and members might think this return is not bad, because higher returns grow your money faster than lower returns.
But Anderson says there is a reality check. “You don’t live off the salary you earned two years ago; you live off what you earned this year. In most cases, your lifestyle adjusts to the income you are earning, and your retirement savings need to finance your new or latest lifestyle for a long time or else your lifestyle in retirement will need to be far more conservative.
“Every time you receive a salary increase in excess of the return earned by your retirement fund, it means that your accumulated savings have not kept pace with your living standard change and hence your expected living standard at retirement will need to reduce unless you make some other plan to close the shortfall.”
In 2005, Alexander Forbes looked at the impact of investment returns over the previous five-year returns on the average actual NRRs of retirees from funds administered by Alexander Forbes.
The company repeated the exercise in 2010.
The comparison showed that the average NRR achieved by 2010 retirees across salary bands compared to the retiree class of 2005, despite good real returns over the period, were much the same (that is, an average NRR of 28 percent!).
“Returns contributed positively but were offset by increases in the cost of annuities (because of low long-term interest rates and projected longevity of retirees) and salary progression.”
Anderson concedes that the reductions in NRR would also be affected by factors such as whether the 2005 group of retirees preserved more or less over their working lifetime when compared with the 2010 group, whether they made better or worse investment decisions, and whether contribution rates had changed.
But the research shows that the message on the need to preserve retirement savings is not getting through. The 2005 research showed that because retirement fund members were withdrawing and spending retirement savings, the potential NRRs of 75 percent were being reduced to between 15 and 24 percent. Only about 10 percent of the people who have lost or left their jobs over the past 18 months have preserved their retirement savings, while only three percent of non-member spouse divorcees who received their share of retirement savings have preserved the money for retirement.
To illustrate the problem of the effect of above-inflation salary increases, Alexander Forbes took three examples based on its retirement member database. The examples are found in “What raises do to your NRR”, below.
So there is a double whammy, namely:
* The cost of retirement (pensions) is going up due to longevity and lower investment yields; and
* Salary increases are mainly being used to fund a higher lifestyle. If you are receiving increases of five percent above inflation or more, you need to save a greater percentage of the increase for your lifestyle to be maintained into retirement. The problem is that contributions currently remain fixed as a percentage of salary, which means the gap will not be closed.
Anderson says the situation is exacerbated by the low rate of preservation of retirement savings when members lose or change jobs and retirement funds.
The Alexander Forbes model used in calculating the wage/savings gap over the period from 2000 to 2010 assumes that individuals preserved all their retirement savings over the 10-year period. Anderson says that, unfortunately, Alexander Forbes research has shown that the majority of people in the age range studied do not preserve their savings when moving jobs, “so in reality many individuals have had their NRR reduced even more than the above because they have not preserved over the 10-year period”.
And that is not the end of the bad news for retirement fund members - even those who have sought to preserve their savings and to make up for any current shortfalls in their NRRs by saving more. Anderson says the really bad news is that you can expect:
* Far lower real investment returns over the next few years compared with the last 10 years.
* Steadily increasing longevity into the future due to healthy lifestyles and innovations in medicine. He refers to some recent research in the United States which found that the first person who will live to 150 is alive today, and that person is probably already in their fifties.
“If the projected NRRs have not kept pace with investment returns over the last really good 10 years, then it is clear that it will be very difficult for them to keep pace in future, given the slow economic recovery, and technological and other developments such as increasing longevity.”
Anderson says in places such as France and the United Kingdom a “perfect storm” of disappointing investment returns over the past 10 years and improving longevity has resulted in retirement funds being hit hard, with consequent moves towards lower pensions and/or the extension of the retirement age.
The result has been widespread civil unrest, which has frequently turned violent.
In these countries, the investment shortfalls and improved longevity have resulted in significant deficits in what are defined benefit funds, where the funds take the risk that members will receive a pre-determined pension.
In South Africa, the predicted forthcoming perfect storm will hit individuals directly, as in most cases South Africans are members of occupational define contribution funds, which provide no guarantees on the level of pension.
Anderson says the uniform limits across all generations on the tax deductibility of contributions needs to be seriously questioned.
Currently you are permitted to deduct from your taxable income contributions of up to 7.5 percent of your pensionable income (basic pay without allowances) to an occupational retirement fund and 15 percent of any earnings (other than your pensionable income) in contributions to a retirement annuity fund. Your employer can make contributions up to 20 percent of your taxable income to an occupational retirement fund, on which you are not taxed.
This could be made worse for people seeking to make good on retirement savings by the newly proposed tax regime for retirement fund contributions.
The proposal is that by next year you will be restricted to a maximum retirement fund contribution deduction of 22.5 percent from your pensionable income from all sources (you and your employer) and a rand cap of R200 000 will be placed on tax-deductible contributions.
Anderson believes, for example, that somebody starting to save at age 40, who has to save more than 22.5 percent of salary to achieve a financially secure NRR, should be allowed to do so, particularly if the retirement fund member did not make use of the tax deductibility of contributions before.
Anderson says such a person should be encouraged with tax incentives to save more to limit the impact on society down the road.
It also allows benefit designs more suited to the changes in individual lifecycles to be better accommodated and hence achieve a smoothing of consumption over time.
Anderson says that already where employers offer “cafeteria plans” (where employees decide how much they want to contribute to a range of employee benefits based on a total cost to company), younger members who opt out of the default contribution rate typically select a lower contribution rate, and older members appear to select a higher rate (see graph 3; link at the end of this article).
“Therefore, on average, older members contribute more towards retirement funding than younger members. This is likely to be a function of younger members needing cash-flow to meet their asset accumulation and consumption, and possibly older members making up for any shortfalls if they contributed insufficient amounts previously.”
Anderson says that the solution to the problem should not be one of tax incentives only. Employers can also start looking at annual salary reviews, not just from a remuneration perspective but also from a benefits perspective, and package the two together. For example, if in 2011 an employer agrees to increase total remuneration by 10 percent, instead of simply passing this through as a 10-percent increase it can instead be implemented, as an example, by:
* Limiting the actual salary increase to 7.5 percent;
* Increasing retirement contribution rates from 10 percent to 11.5 percent (that is, one percent of the salary increase is used to increase your retirement fund contribution rate); and
* Introducing a dread disease benefit to address the shortfall that employees have in this cover. In this case, the cost amounts to one percent.
So, in total, the remuneration increase of 10 percent is split into 7.5 percent for current, short-term standard-of-living needs and 2.5 percent for improving long-term financial wellness (retirement and dread disease).
Anderson says by doing so, a change in culture can be facilitated whereby individuals see salary adjustments for what they are, namely:
* An increase that helps them to improve current living standards/needs; and
* An increase that they need for the future, to help smooth consumption spending once their human capital (ability to earn) is depleted, as well as protecting human capital during “shock” events (such as dread disease in the above example).
However, Anderson says there is an equal responsibility on you, in your own interests, to actively manage your NRR, particularly in the years when you do not have responsibility for children, by ensuring you:
* Preserve your retirement savings for retirement; and
* Save a portion of any salary increase towards your retirement by making voluntary additional contributions to your occupational retirement fund and/or retirement annuity fund up to the maximum amount allowed for tax deduction purposes.
If you require further savings to improve your NRR, you will need to save outside of the tax-incentivised regime with after-tax money.
WHAT RAISES TO YOUR NRR
Here are three examples that demonstrate how above-inflation increases in your income will affect your net replacement ratio (NRR).
* On May 1, 2001, a 30-year-old retirement fund member is on track (under a fairly standard set of assumptions) to achieving an NRR of 75 percent by May 1, 2011. The member maintains the same level of contributions and is invested in a typical lifestage investment portfolio.
Based on typical lifestage-type investment strategies, the average return for the last 10 years was 14 percent, or a real (after-inflation) return of 8.08 percent.
However, Alexander Forbes found in its research, using the progression of a typical retirement fund member's salary on its database, that the salary progression (which includes merit, promotional and annual inflationary increases) averaged 11.06 percent a year (5.09 percent above inflation) over the 10 years.
The effect on the member is that by May 1, 2011 – when he was 40 and 20 years away from retirement - he saw his NRR decline from 75 percent to 63 percent despite the fact that real returns over the last 10 years have been significant.
This translates to an effective reduction in pension of R1 200 a month for someone earning R10 000 a month.
For someone earning R25 000 a month, the reduction in pension is R3 000 a month.
* On May 1, 2001, a 40-year-old retirement fund member was on track to achieving an NRR of 75 percent. On May 1, 2011, the fund member was expected to achieve an NRR of 69 percent. The return was 14.05 percent a year and the salary progression was 9.03 percent a year (compared with inflation of 5.97 percent a year). This results in a six percentage-point reduction in the NRR, which amounts to R600 a month less in pension for someone earning R10 000 a month. For someone earning R25 000 a month, the effective reduction in pension is R1 500 a month.
* On May 1, 2001, a 50-year-old retirement fund member was on track to achieving an NRR of 75 percent. On May 1, 2011, the fund member was expected to achieve an NRR of 67 percent. The investment return was 12.78 percent a year and salary progression 8.47 percent a year (compared to inflation of 5.97 percent a year). There is therefore an effective reduction of R800 a month in pension for someone who earns R10 000 a month. For someone earning R25 000 a month, the effective reduction in pension to an NRR of 67 percent is R2 000 a month.
DEFINITIONS
NET REPLACEMENT RATIO
Your net replacement ratio (NRR) is your initial pension at retirement calculated as a percentage of your final pay cheque. In calculating your NRR, actuaries take into account various factors, including:
* How much you have saved. The more you have saved, the higher the NRR.
* The expected returns on your savings. The higher the expected returns, the higher the NRR.
* How long pensioners live on average. The higher the average age, the lower the NRR, because your money has to last longer.
* Long-term interest rates. As a major proportion of pension savings are in interest-earning investments, the lower the rates, the lower the NRR, because your money will not earn as much as it would if interest rates were higher.
LIFESTAGE INVESTING
Lifestage investing reduces the risk of exposure in your investments mainly by adjusting the proportion of equities to interest-earning investments.
So for the first 20 years you may invest up to 75 percent of your retirement savings in equities (the maximum amount permitted in terms of the prudential investment guidelines detailed in regulation 28 of the Pension Funds Act), but by the time you reach retirement age the percentage of equities in your savings could be below 40 percent. There is no set rule, because retirement ages differ, as do the years of saving between individuals.
This article was first published in the third-quarter 2011 edition of Personal Finance magazine.