For economic and social researchers, data can come in handy to either reinforce your own views or to cast doubt on the views of those that disagree with you.

In the last couple of months, I have been fortunate enough to have access to some mighty fine spreadsheets.

One of my favourites is the ATO two per cent sample file of personal income tax and superannuation data items. Two per cent does not sound very much but there are a lot of people lodging a tax return. A spreadsheet of 285,000 rows and lots of columns is an impressive thing, though it does tend to make the notebook computer grind a bit when you open it up and do sorting and calculations and the like.

The ATO data confirms what most sensible people already knew. Superannuation is the main financial asset (apart from the family home) for the great majority of Australians, particularly younger Australians. Australians may have a love of property, but investment properties are owned by a relatively small proportion of Australians, as are direct shareholdings in companies. The reality is that without compulsory superannuation, the great bulk of Australians would have little or no savings, retirement or otherwise.

Conducting a controlled experiment in the medical or physical sciences is hard enough and can raise some ethical issues, but in the social sciences it is even harder. You do not have the luxury of carving out one section of the population on a random basis and applying compulsion to superannuation contributions and having another equivalent section left to save on a voluntary basis.

However, we do have clear historical evidence for Australia and evidence for what happens in other countries. In regard to boosting retirement savings, and savings more generally, compulsion trumps opt-out arrangements which trump purely voluntary arrangements for saving.

Of course, for the wealthier who tend to save without compulsion there is a substitution effect with compulsory superannuation leading to lower voluntary savings in superannuation or other forms of savings. However, at the lower income end of the community this substitution effect is going to be very low, perhaps 10 per cent or under, on average for low income workers.

Leaving retirement savings decisions to individuals, as some conservative economic commentators advocate, is fine for the wealthy but would condemn the bulk of working Australians to a lifestyle just above the poverty line in retirement.

We are also starting to get statistics and data about individuals who have applied for early release of superannuation because they have become unemployed or have had their paid working hours reduced substantially.

Perhaps it is just as well I do not have the full spreadsheet on such individuals as the number of early release applications is rapidly approaching two million in total, with the weekly volume of applications in mid-May still around the 200,000 a week mark. This latter figure is down though on the first week rush of around 700,000 applications.

The second tranche for this measure is not too far away, with the first of those applications able to be made from 1 July, and funds making the first payments a few days later.

The volume of applications second time round is likely to be down a bit on the volume for the first round. This will be through force of circumstances for some. Temporary residents get only one shot at early release. For Australians more generally, data available to ASFA indicate that between 5 and 10 per cent of applicants in the first round pretty much exhausted their superannuation balance, with another 25 per cent of applicants having under $6,000 left.

On a more positive front, growth in employment as various industries and activities re-open as restrictions are lifted should mean fewer people will need or want early release of their superannuation.

There also have been some surveys on what people do with the early release payments. The ABS when surveying the population get the more worthy intentions from individuals, such as paying off debt or adding to other form of savings. However, the reality has been somewhat different, as shown by analysis of bank account transactions of individuals receiving early release payments.

This should not really come as a shock. If you give someone who has been doing it hard an amount approaching $10,000 (for a substantial proportion the largest amount of cash they have ever received in their life) then it is not too surprising that a substantial proportion goes on things like online gambling, alcohol or other truly discretionary expenditure. However, the bulk of the money has been spent on sensible things. Expenditure like paying down credit card debt and paying the rent has been what many recipients have been doing. The pattern of expenditure might also change going forward, with the amount of online betting going down as Australians are able to return to poker machine palaces and parlours.

The early release arrangements have not been without their flaws. However, the scheme has demonstrated the fundamental importance of superannuation for the savings of working Australians. What Australia needs going forward is a move to a 12 per cent SG as soon as possible, particularly given that the recent early releases have substantially diminished the retirement savings of nearly 2 million Australians. Younger Australians can catch up on their retirement savings, but they will need a 12 per cent SG to do so.