A kludge is a collection of parts that clumsily but effectively provides a temporary solution to a fault or problem, especially in computer systems.

This week, John Daley, Brendan Coates and Danielle Wood, of the Grattan Institute, wrote: "Australia's superannuation system has become a textbook example of 'kludgeocracy'. Complexity originally driven by the search for fair outcomes has ultimately provided large benefits to vested interests with the time and resources to push for technical changes that serve their interests."

Their paper, Super Tax Targeting, is utterly damning.

Unfortunately, Australia's superannuation scheme is such a kludge that it is hard to explain what a great big rip-off it is for 80 per cent of people. And what a huge boon it is for the top 20 per cent of income earners, who take more than half of the $25 billion of annual superannuation tax concessions.

Meanwhile, the bottom 15 per cent whose marginal income tax is zero have their compulsory superannuation contributions taxed at 15 per cent.

Australia's superannuation system has no legislated aim. We have to assume it is to encourage and even force saving for retirement and to reduce dependency on the age pension, thus saving the public purse. But any intelligent Martian looking at this kludge could only conclude that it is in fact an elaborate tax dodge for the wealthy.

The Grattan Institute paper shows the scheme does not save the public purse nor does it encourage much saving. This is because such a large portion of the concessions go to people who would never qualify for the pension anyway, and those high-income people would still save the same amount irrespective of tax concessions.

The concessions will cost $40 billion by 2017-18 compared to the total cost of the age pension last financial year of $41 billion.

True, high-income earners may put their savings in other tax-effective vehicles, such as negative gearing, but that is an argument for addressing those tax vehicles, not an argument for maintaining the present super system.

The people benefiting most from the present system do not need any encouragement to save. Further, there is no moral argument for them to get government help.

At present, high-income earners can put up to $35,000 a year into super and it is taxed at 15 per cent instead of the top marginal of 45 per cent. They can put after-tax money in, too, and its earnings are taxed at just 15 per cent instead of 45 per cent if invested outside superannuation.

And when they retire they can take it all out tax-free.

Remember treasurer Peter Costello on Budget night 2006 announcing superannuation withdrawals by retiring people over 60 would be tax-free and the measure would "sweep away the current raft of complexity faced by retirees, increase retirement incomes, give greater flexibility as to how and when superannuation can be drawn down, and improve incentives for older Australians to stay in the workforce".

Never mind the cost and the fairness.

Now let's go back to the kludge.

Note the Grattan Institute paper did not condemn kludges per se. Rather it objected to the ability of well-resourced people to exploit kludges by pushing for technical changes that advantage them.

It cuts both ways. A good kludge full of fairness-yielding complexity can equally be exploited by well-heeled pressure groups or political parties by urging its replacement with a "simple" system.

Costello swept away the complexity in favour of appealing simplicity, but it was unfair simplicity.

Similarly, Labor's Mick Young swept away the complicated requirement to number every square for a valid Senate vote and delivered us the utterly unfair system that gives Senate seats to people with practically no voter support.

We need to be wary of both kludges and calls for simplicity, and look for underlying fairness.

Simplicity can appeal to fools who are then duped. Kludges can muddy the waters so only the most diligent can see the unfairness while most voters cannot be bothered.

Simplicity usually does not deliver fairness in tax or electoral systems, or, indeed, other area of government policy.

Kludges, on the other can. But it requires trust by voters and intelligence and morality on the part of legislators.

The fundamentals of the Superannuation Guarantee system are sound and can be made to be fairer. Levying up to 12 per cent of income over a working life to provide for retirement is perhaps Paul Keating's greatest legacy.

Tweaking that system to favour mostly Coalition-voting self-funded retirees is perhaps Costello's worst.

Time to fix it.

The Grattan Institute recommends cutting the $35,000 deductible super input to $11,000. With some other tweaks this would save $3.9 billion a year – and rising. It says earnings in retirement should be taxed at 15 per cent, not zero, saving $2.7 billion a year – and rising.

That is a good start with three caveats – which a kludge could fix.

First, many women who leave the workforce to have children often hit their 50s and 60s with quite low superannuation balances and sometimes high incomes. They should be allowed to catch up. The Grattan paper acknowledges their plight but in effect says, "Too bad, a price worth paying for the overall good."

I think women have been discriminated against on the income front for far too long to justify that.

Secondly, sometimes superannuation accounts go backwards, as in the 2008 financial crisis. Those losses should be deductible against future earnings.

Thirdly, inflation – even at present low rates – can erode retirement savings nest eggs. Income from labour, on the other hand, gets adjusted for inflation through regular wage and salary increases. So any tax on fund earnings should be done only on real earnings – that is the net earnings after enough of the nominal earnings have been put back into the capital fund to retain its value after inflation.

The last one is a real kludge – quite complex, but ultimately fair.

Ultimately, complexity is not the issue, but fairness and practicability. And with modern computers increasingly complex systems can be made workable. That, indeed, was the origin of the word "kludge".

crispinhull.com.au