Overview of the UniSuper DBD Report

By Brian Halstead

In traditional DB funds, the employer guaranteed and structured a lump sum or pension scheme that, at the end of a career of 40 years, gave a benefit based on member’s final average salary.  If the member had less than the 40 years service then the amount of the lump sum was reduced by an actuarial discount.

As this required assumptions into the future, an actuary calculated the level of average contribution required to fund this benefit.  If the fund performed better than the assumptions the employer could reduce its contribution or if it performed worse the employer increased its contribution.  Employees could also contribute and this increased the benefit.

In the case of UniSuper the defined benefit is NOT guaranteed but the contributions as a % of salary are. Thus if the fund performs better than the assumptions which supporting the contribution rate, funds build up and if it performs worse, then the actuary calculates for trustees how much the benefits must be reduced to ensure that the fixed contributions generate enough funds to meet the benefits.

A reduction of this type is what is happening from the 1 January 2015.

Currently, the annual contributions from employers and employees are paid in to the DBD and new members have the right to stay in the DBD or move these contributions to the accumulation fund within first 24 months of employment (previously 12 months). After that time the member is locked into the fund until he resigns or retires and takes a lump sum or a DBD pension. The pensions are paid for life and once taken cannot be commuted back to a lump sum.

Therefore there are critical decisions to be made in the first 24 months of employment, whether to stay in DBD and at retirement whether to take a lump sum (which can then be invested in an allocated pension) or non-commutable pension. As the benefits from the DBD are not guaranteed, both these evaluations must focus not only on the benefits promised but also on the alternative retirement benefits that could gained by transferring the guaranteed contributions to the accumulation fund. The risks associated with both must be evaluated.

Part 1   The structure of the fund and the benefits including pensions.

A) Structure for members with Lump Sum Benefits

Defined benefit funds have traditionally been designed to advantage members who have salary increases greater than the average and have long service.  DB funds therefore relatively disadvantage those who stay a short time with salary increases at or less than average.

The following calculations on the DBD are based on employer contribution of 14% of salary and the member 7% after tax with benefits based on the new salary averaging formula.

If member starts at 25 and stays to 65 (40 years) then the lump sum retirement benefit is 9.2 times final average salary of the member so the $ amount depends on salary growth.

Here is an example to illustrate the results over 40 years with inflation at 2.75% and

two salary growth rates, one at  inflation and one 2.5% above  with results shown  per $10,000 starting salary.

Salary at 25   $10,000 $10,000
Salary growth  pa   2.75%   5.25%
Final average salary   $28,055   $70,077
Net contributions   $128,543   $231,228
Lump sum $   $258,114   $644,709
as % of contributions   201%   279%
Annual rate of return   3.8%   6.2%

Net contributions are gross contributions by employers and employees less the tax on superannuation contributions and the cost of the insurance policy. The rate of return on net contributions includes inflation of 2.75%. It is the rate after taxes and fees that an accumulation fund would need to earn to achieve the same lump sum results over 40 years. It could be claimed that 5.25 % pa salary increase is high but it is used to illustrate the impact of salary increase on benefit.  A 3.5% salary increase would give gives an annual rate of return required of  4.5% .

Further examples are given in Attachment 1

These calculations were done using a model methodology and assumptions used by UniSuper.

If the same two members entering at age 25, leave the fund after 10 years, the benefit is 1.8 times final average salary, giving the following results per $10,000 starting salary.

Salary at 25   $10,000   $10,000
Salary growth  pa   2.75%   5.25%
Final average salary   $12,433   $15,097
Net contributions   $20,568   $22,998
Lump sum $   $22,379   $27,176
as % of contributions   109%   118%
Annual rate of return   1.9%   3.8%

This rate of return includes inflation of 2.75%, so over 10 years the lower salary growth member does not even maintain the real value of their net contributions. In 10 year case the higher salary increase is not extraordinary and 5% was used by UniSuper in their latest booklet example.

These rates of return, which are after tax and fees, can be compared with other funds.  ASIC web site Moneysmart has estimated returns for accumulation funds at 3.15% for the cash option, 4.9% for the conservative option and 5.8% for the moderate option all after tax. The UniSuper actuary, in his January 2013 report, gives the best estimate assumed return after tax at 7% and funding assumed return at 5.5%. An estimate of fees would need to be subtracted from these rates.

As can be easily seen, the returns to younger members who stay a short time and with lower salary increases are much worse than comparative returns above. Even with the higher salary increase, younger members over 10 years are obtaining a rate of return between the Moneysmart cash and conservative options.

The member whose salary only increases by the rate of inflation over 40 years only receives a rate of return of 3.8% which falls between the cash and conservative options and well below the UniSuper assumptions.

These returns over 10 years are well below the UniSuper best estimate return and the funding return assumptions. Therefore to require those funding return assumptions  the members entering later and with longer service members must be advantaged with higher returns. This  can be seen in Attachment 1

The following example compares the lump sum from defined benefit and the accumulation fund return, using the actuary’s rates, less fees estimated at 0.8% per annum.

Salary at 25   $10,000 $10,000
Salary growth  pa   2.75%   5.25%
10 Years
Lump sum Defined Benefit   $22,379   $27,176
Accumulation Fund
Funding return 4.7%   $25,043   $27,066
Best estimate return 6.2%   $26,789   $29,715
40 Years    
Lump sum Defined benefit   $258,114   $644,709
Accumulation Fund
Funding return 4.7%   $306,349   $479,090
Best estimate return 6.2%   $424,244   $634,490

Only at the higher salary growth rates does the defined benefit give a better return than that estimated for an accumulation fund .  At the salary increase equal to inflation DBD return is more that 20 % lower than the accumulation fund

While these lower lump sums and rates may have been acceptable if the fund was guaranteed and members did not have to consider any risks, without a guarantee members must also consider the risk of further benefit reductions as outlined in Part 2.

Another of the interesting features in the fund structure is that for members older at entry, accrual of benefits is at a faster rate over 10 years.

An example shows that for entry at age 45, after 10 years the benefit is 2.1 times final average salary. The results per $10,000 starting salary are below

Salary at 45   $10,000   $10,000
Salary growth  pa   2.75%   5.25%
Final average salary   $12,433   $15,097
Net contributions   $20,568   $22,998
Lump sum $   $26,109    $31,705
as % of contributions   127%    138%
Annual rate of return    5.1%    7.0%

Thus a member who enters at age 45 gets 16.6 % greater lump sum after 10 years than a member who enters at 25 although they have made the same contributions. That equates to twice the return.

Detailed results of returns with various ages of entry, service and salary increases and using old and new formulae are shown on Attachment 1.

It is clear that the age of member at entry, how long member intends to stay and anticipated rate of salary increase, are critical factors in determining if a member would be better off in the accumulation fund and disadvantaged by staying in the DBD.

The advantage for some members was reported to trustees by the actuary in 2012 report as follows;- “Eligible new employees have 24 months to elect between the DBD and Accumulation 2. This introduces a risk that only members who will be advantaged by being in the DBD (e.g. those who expect high salary increases) may remain members.”

These advantages and disadvantages of the structure have not been adequately explained to the members in the PDS or any brochures and members should be given the opportunity for a limited period to exit to accumulation. 

B) Changes being made

For benefits accrued after 1 January 2015 the average final salary will be calculated over five years, not three, and not adjusted for inflation.  For a member joining after this time the lump sum benefits are reduced 5.2% if salary increases at the rate of inflation and 7.3% for members with salary growth 2.5% above inflation.  This reduction is the same no matter what the length of service.

This is shown in the examples below.

Annual rate of return   5.1%   7.0%

Salary at 25   $10,000   $10,000
Salary growth  pa   2.75%   5.25%
After 10 years
Averaging Formula   Lump sum    Return   Lump Sum   Return
New   $22,379   1.9%   $27,176   3.8%
Old   $23,609   3.0%   $29,318   5.4%
After 40 Years
Averaging Formula    Lump sum    Return   Lump sum   Return
New    $258,114    3.8%   $644,709   6.2%
Old    $272.604   4.0%   $695,529   6.5%

More details of the returns with old and new formulae is given in Attachment 1.

The reduction is adjusted by the proportion of service prior to January 2015, therefore, the impact on those with long service and close to retirement is very small and impact on those currently being paid pensions is nil.

As this is a significant reduction especially for the younger members with less service, they should be given the opportunity to exit to accumulation if they so desire.

C) Structure for members entitled to take pensions

Pensions are determined by a % of average final salary dependent on years of service.

Division A Pensioners after 40 years receive annual indexed pension of 64% of average annual salary and Division B 63.2%.  The pension is smaller for shorter service and easily calculated from information provided by UniSuper.

As the fund has no guarantee, with no more contributions from employers for the pensioner and only the lump sum members left in the fund, the question is how much does the fund have to earn on that lump sum to ensure pensions can be paid?

Making the simple assumption that pension is taken at 65 and all survive 20 years to 85, then the rate of return on the lump sum required to fund an inflation-indexed pension after costs, is 5.9% with inflation estimated at 2.75%.  If member survives 15 years, return required is 2.9% and 25 years 7.4%. Other benefits of pension include reverting to surviving spouse and these are clearly set out in the PDS.

Pensions in place are not being impacted by the change being made on 1 January 2015. New pensions will be reduced by the small pro rata impact of lower average final salary on years of service after 1 January 2015. Clear examples are given on the UniSuper website


The funds for pension benefits are not segregated from other funds and there are no reserves for market downturns.  The decision on whether to take the lump sum or the pension is based on understanding your potential lifespan and if the fund is likely to be able to make the necessary returns above over your lifespan.  There may well be other benefits flowing from taking an annual pension that are very important.

It is necessary for those currently thinking of taking a pension and those already receiving a pension to understand the risks to the fund performance in Part 2.


General warning

The material in this paper is of the nature of general comment only and does not represent professional advice. It is not intended to provide specific guidance for particular circumstances and it should not be relied on as the basis for any decision to take action or not take action on any matter that it covers. Members especially older members close to retirement who need to preserve their benefits for retirement and should not be taking significant risks with those funds, should obtain appropriate professional advice before making any such decision. To the maximum extent permitted by law, the author disclaims all responsibility and liability to any person, arising directly or indirectly from any person taking or not taking action based on the information in this paper.


    • Nathan

    • December 17, 2014

    • 4:58 am

    • Reply

    Hi Brian, this is a very interesting read, particularly as a UniSuper member! I’m curious as to how you calculated the total for the accumulation fund in your example above?

      • Brian Halstead

      • January 6, 2015

      • 11:02 pm

      • Reply

      Hi Nathan,

      Thanks for your question.

      In the example in Part 1 comparing DBD and accumulation the accumulation balance was calculated starting with nil balance, a starting salary and then each year increasing the salary by the rate of salary increase. The balance at the end of the year is calculated by adding the earnings (at specified rate) on the previous years balance to that balance and adding the new contributions for the year (14% pre tax of salary for employer and 7% post tax for employee less the contributions tax (15% on pre tax contributions) and less 0.9% of salary to cover the cost of insurance. The insurance cost looks high but it is the rate UniSuper uses in calculating the rate of return of the DBD. The earnings rate shown is after tax and fees and rates used in the example are 4.7% and 6.2% pa.

      Kind Regards

    • PatriciaFob

    • January 23, 2015

    • 4:23 pm

    • Reply

    Greetings! Very useful advice in this particular post! It’s the little changes which will make the largest changes. Thanks for sharing!

    • Brian Aldon

    • September 10, 2015

    • 7:38 am

    • Reply

    Very neat blog article.Thanks Again. Really Great.

Leave a Reply