I like a big spreadsheet, and one with over 277,000 rows is particularly special, albeit tricky to navigate.

Over the last few weeks there has been a veritable cornucopia of data released by the Australian Taxation Office (ATO), including a range of summary tables about the varied characteristics of taxpayers, such as taxable incomes, how much tax they pay and the distribution of superannuation balances.

The ATO also released individual data for two per cent of overall personal income tax returns, appropriately made anonymous by tweaking some data items such as address to a very broad geographic area and averaging some items across groups of taxpayers, especially where a big income or big superannuation balance is involved. There also has been the release of comprehensive data about SMSFs.

It is a very useful resource for better understanding the characteristics of the Australian population and for modelling the impact of various possible policy changes. Fortunately, we seem to have a moratorium on any further detrimental changes to the taxation treatment of superannuation.

I always like to see how my personal characteristics compare to the averages and medians for taxpayers as a whole. Comparisons of various things by postcode also are useful for those who want boasting rights relative to their neighbours.

I quite like it that Pymble has a higher average taxable income than residents of nearby Turramurra, but unfortunately both missed out being included in the national top ten postcodes for average income.

However, some care is needed in making use of averages. Properly used they can be a useful diagnostic tool, facilitating the identification of areas of relative success (or failure). This then invites the development of approaches to bolster success and reduce failures.

Simplistic use of averages is another matter. Moving from Turramurra to Pymble will not necessarily lead to an increase in your income (but of course you would have the opportunity to meet some very nice people already living there).

So, it goes with some sectoral comparisons with regard to superannuation returns.

SMSFs at the clubhouse

Bragging rights can be very important for some people, with the success of your SMSF part of the chat at many golf clubhouses, Rotary Club meetings and suburban barbecues.

After the release of the latest SMSF statistics at least one organisation made the claim that SMSFs had performed very well, given a 10.2 per cent average return in 2016-17 compared to a 9.1 per cent average return for APRA regulated funds.

A deeper dive into the data tells a more complicated tale, and the comparison becomes less stark. Much of the claimed outperformance of the SMSF funds was actually driven by the average returns achieved by SMSFs with more than $2 million, with many of those funds in the tax-free pension stage. For funds with assets between $1 million and $2 million, the average return was 8.4 per cent, dropping to an average return of 7.0 per cent for SMSFs with assets between $500,000 and $1 million. The conclusion to be drawn is that having more than $2 million in your super is a very helpful thing.

The clear message from all of this is that the right question to ask is whether you are in the right fund and investment options given your circumstances, rather than one type of fund being inherently better or worse than another. There is no sort of contagion at work where the large balance and good returns achieved by somebody else will necessarily flow to you.

It also is interesting to note that limited recourse borrowing has grown substantially in recent years, with around 12 per cent of SMSFs in accumulation mode in 2016-17 making use of such arrangements. Often assets supported by such borrowings make up a large proportion of the assets of the SMSFs concerned. Subdued growth in property prices might be a drag on the future performance of a substantial number of SMSFs.

The ATO data reinforce what many researchers have understood for a long time. Averages can be interesting but what is important are the factors driving those averages. Good analysis is needed along with well thought through policies to produce the good superannuation outcomes that Australians need and deserve