UniSuper’s boss flags a potential correction ahead

James, you say that "Over a 10-year period, UniSuper’s Balanced and Sustainable Balanced options have handily beaten all the index funds, with annual returns of 7.92% and 7.79% respectively." I think we need a clarification here because there are at least 20 index funds that have exceeded the return of 7.92% pa. Compounded over ten years this is a total return of 114%. Iress shows that over the ten year period to 31 Dec 2024 IVV has returned 300%, SPY 299%, QUAL 287%, VTS 284%, IOO 274%, IJH 201% and so on. cheers

Yep. Unisuper appears to only ever report against their CPI+ benchmark rather than actual indices. So you won't know what alpha they actually added. Comparisons of their performance against pure index balanced funds in also not apples with apples given UniSuper $51B default balanced fund is label only. It's equivalent to Growth for other providers given their higher allocation to Aus and International shares.

Stephen,

Good question. In their update, UniSuper compared their 'top-quartile' returns over ten years to 'index funds' but didn't specify which index funds they were referring too.

Will endeavour to get the answer.

James

Agree John always has a way of simplifying things. Whilst watching tariff is an obvious and direct economic event, Trumps international posturing over Ukraine, NATO, Gaza, WHO and USAID etc also has significant potential to create systemic fractures. Trump 1.O DOHA Agreement is just one obvious example to show just how bad things get when you don't think through the details and work against your supposed allies.

P.S. I'm sure Vanguard could readily correct the statement UniSuper easy beat all Indicies over 10 year period

P.P.S There is 90% chance in any given year there will be at least 10% drawdown. In fact there is greater probability of more than 20% draw down in any given 5 year period

Thanks James

A very good article and thought provoking comments from John Pearce.

Just one point that is often glossed over shown in the following comment on performance -

“Over a 10-year period, UniSuper’s Balanced and Sustainable Balanced options have handily beaten all the index funds, with annual returns of 7.92% and 7.79% respectively.”

If an investor achieves a compound annual return of 6% then over ten years looking back the average return is 8%.

The power of compounding often overstates the actual annual return in the presentation of long term returns.

As Buffett has often stated - the power of compounding is an investors best friend.

That's mathematical nonsense John.
Start with $100. Grow by 6% in year 1 that's now $106. Grow that by 6% and it's $112.36 after 2 years. Keep that going for 10 years and it's worth $179.08.
Now calculate the historical return of that: ((179.08/100) ^ (1/10))-1) = 0.06 That's 6% per annum, not 8%.

Compounding is a real thing, but it doesn't inflate returns the way you're claiming.

The comment that the average is 8% is correct - it's the simple annual average return being referred to. It's just being used as an incorrect reference (ie Unisuper achieved compound annual returns of 7.92, not the implied 6% in the comment).

I don’t see the reference to compound returns in super returns nor in the advertising of returns.

“Over ten years …..annual returns of 7.92% and 7.79% respectively.”

Annual returns are averaged looking back.

Further and let’s be very clear and honest. Compounded returns in pension funds - particularly in pension mode- cannot be quoted or calculated because of outflows ( ie pension payments).

Indeed each pension member will have different returns determined by the annual, monthly or quarterly pension payments made.

"I don’t see the reference to compound returns in super returns nor in the advertising of returns.":

Random different super vendor:
https://rest.com.au/investments/understanding-investments/investment-returns

"3 standalone years
Year 1 return 3.0%
Year 2 return 12.0%
Year 3 return 1.0%
Annualised over 3 years
3-year annualised return 5.2% per annum
Cumulative over 3 years
3-year cumulative return 16.5%"

Annualised over 3 years:
Average per y;
= AVERAGE(3%, 12%, 1%)
= 5.3333333%

Cumulative over 3 years;
= PRODUCT(1 + 3%, 1 + 12%, 1 + 1%) - 1
= 16.5136%

Compound annualised over 3 years;
= PRODUCT(1 + 3%, 1 + 12%, 1 + 1%) ^ (1 / 3) - 1
= 5.2266006%

Rest example "3-year annualised return 5.2% per annum" is compound not average annual rate.

I would expect the 'rest' of the super vendors to also use compound not average, because average exaggerates.

Looks to me that we have the basis of a future article on or how returns ate calculated and reported.

If contributions and pensions are constantly flowing in and out, franking credits claimed, fees and insurance premiums deducted , fines paid- then the calculation of returns ( including the value of unlisted assets ) looks like a work of art and / or financial construction.

"Compounded returns in pension funds - particularly in pension mode- cannot be quoted or calculated because of outflows ( ie pension payments)."

Funds in accumulation mode also have outflows.

The compound rate of return can be calculated for both accumulation and disbursement ('pension') funds and individual accounts where outflows have occurred as the 'Internal Rate of Return'.
Example spreadsheet functions: 'IRR(...)' and 'XIRR(...)'.

Proportionally large outflows over a short time period can distort the rate, but consistent outflows less so.

Dudley, your quoted formulas require precise and accurate inputs to derive precise outputs for the real calculation of returns for each individual super fund member.

You haven’t shown as to how those precise inputs - which occur daily if not hourly - are able to be collated for every member receives their individually calculated return.

It is financial engineering that serves a purpose, but it is important to maintain confidence in the system by not openly disclosing the extent to which calculations are based on assumptions and not facts.

John again I have to correct you. Super fund returns are calculated using fund value calculations that allow for inflows and outflows to generate the outcome for a constant initial investment that's reinvested. That is, it's the return of the underlying investment portfolio.
Of course for every individual who makes contributions over time - and makes withdrawals - that number won't fully reflect their personal results perfectly. But as a measure of the super fund's performance from point to point its accurate and valid.

Linda,

"how those precise inputs - which occur daily if not hourly - are able to be collated for every member receives their individually calculated return":

Can not post files or images here so description must suffice.

Spreadsheet columns: Date, Invest, Withdraw, Return, Net
on each row, Net = sum(Date, Invest, Withdraw, Return)
First row in column Invest must be an amount invested - must be negative (out of hand into fund)
Withdraw and Return amounts must be positive.
Last row Withdraw must be fund / account balance - positive.
Calculate =XIRR(Net, Date)
Google Excel XIRR

As you say must be precise with presentation of inputs.

SMSF Trustee

If you are correctly describing yourself and you are a SMSF Trustee, for one or two members, then I am sure that you could probably work out your returns - but they would still be an estimate.

As for Industry and Retail Super Funds - which you seem keen to support - with their one million odd members - pension and accumulation - the returns are constructed from a range of estimates, with assumptions ,with reserving, with the sharing of franking credits, and with unlisted valuations

Do you have specific ( industry) knowledge to the contrary and what is your source?

"on each row, Net = sum(Date, Invest, Withdraw, Return)":
err;
on each row, Net = sum(Invest, Withdraw, Return)

Dudley

Just 4 inputs ( for example) which individually will affect the calculation of daily returns.

Describe how the formula works

1. Unlisted asset valuations.
2. Franking credits ( finalised at end of financial year).
3. Accumulation tax (15%) and nil tax for pension funds - finalised at end of financial year; and
4. Currency movements - hedged and unhedged - changes all the time


Cheers

Linda, "Just 4 inputs ( for example) which individually will affect the calculation of daily returns.":

The account owner 'takes what is given or takes their money'.
To get their IRR based on daily balances:
At first date, value = -AccountBalanceAtThatDate.
Between dates, value = (+AccountBalanceAtThatDate - AccountBalanceAtDayBefore).
At last date, value = +AccountBalanceAtThatDate.
= XIRR(Values, Dates)

The considerate fund manager will try to apportion non-daily changes on fund returns, such as franking credits and revaluations, to each account daily.
They may know the franking rate as it may be declared by company at each dividend declaration.
Not perfect but does not change the IRR much for account owners over the longer term who make few or small investments or withdrawals.

Dudley

“The considerate fund manager will try to apportion non-daily changes on fund returns, such as franking credits and revaluations, to each account daily.”

I rest my case

Linda, "I rest my case":

It'sa nota so bad.

10% at end of year:
Date Value
1/01/2021 -1
1/01/2022 1.1
XIRR 10.00%

10% over a year:
Date Value [ daily =(1 + 10%) ^ (1/365) - 1 ]
1/01/2021 -1
2/01/2021 0.000261158
3/01/2021 0.000261158
...
29/12/2021 0.000261158
30/12/2021 0.000261158
31/12/2021 1.000261158
XIRR 10.00%

John A, you asked me: Do you have specific ( industry) knowledge to the contrary and what is your source?

Yes, I do. I worked for 40 years in the industry, including with a major funds manager. I tracked fund returns daily and spent a lot of time understanding how the fund accountants did the calculations of unit prices, fund values and fund returns.

The point you are missing in this whole discussion is that when John Pearce or anyone from a super fund says, 'our fund returned X% over Y years' they are not saying that every individual investor who came into the fund, made extra contributions or withdrew funds for whatever reason, made X% over Y years. They are saying that the fund in which people came into and went out of made that return.

You are criticising them for providing a common point of comparison, both across their own different fund styles as well as with other fund managers, on the underlying portfolio. It's a measure of their management skill. This is a perfectly valid and important piece of information. But it's never put forward as an individual's personal return. That's why I've reacted to your criticism - because it's a straw man argument.

As for my SMSF returns, yes they are calculated by my administration provider based upon my own actions within the fund. Let's say I had 100% on day 1 in an asset that earned 10% over 5 years, then 20% for 5 years to get around 15% pa over 10 years. (That's a very very rough approximation.) But if I sold that asset after 5 years and invested in something that made 10% pa over the next 5 years, then my fund return is going to be 10% because it will allow for the change in asset and returns on the assets held. I didn't earn the 20% pa on the first asset because I sold out of it before it's return accelerated.

But that doesn't mean that the first asset didn't earn 15% pa over 10 years, just because I personally didn't own it for the back half of the period. Which is what you are in effect trying to claim is the falsehood in super fund returns not taking individuals into account. Of course a UniSuper investor who pulled money out ahead of the better years or put money in ahead of the worse years within the last 10 won't have earned the 8% that the fund earned. But so what? That doesn't mean that UniSuper didn't earn 8% pa, just that the individual wasn't fully invested in that fund for that period of time.

The industry has standards that all funds must adhere to. Asset consultants and others check performance numbers all the time. Your criticism is simply not valid.

An excellent article. All makes good sense backed by reliable stats.