Robbing Peter to pay Paul not the answer

By Hielke Buddelmeyer, Guyonne Kalb and Stephen Sedgwick

The Weekend Australian,
4-5 October 2008, p. 22

THE tax and transfer system has become more generous in recent years, especially for households with children. However, the system is still failing in important respects, especially because it still imposes high effective rates of tax on many secondary earners, mostly women and in particular mothers.

The Government has rightly commissioned the Ken Henry review to examine the combined effect of the tax and transfer systems. Any revised arrangements need to satisfy the principle of simplicity and support higher workforce participation. It seems unlikely, however, that a simpler system can be revenue neutral.

Twenty-five per cent of commonwealth revenue is recycled as transfer payments. More than half of Australian households receive some kind of income support payment, many of whom also pay at least some income tax. Indeed, sole parents or couples with at least two young dependants have to earn well more than $50,000 before income tax exceeds their income support payments. The phenomenon by which households pay tax and receive benefits is called churn.

Neither tax collection nor the administration of income support is costless. Efficiency, administration and compliance costs are incurred by individuals and the community. These most likely rise as the complexity of the system increases. Some studies have put the compliance costs associated with the main commonwealth taxes at 1.4 per cent to 2 per cent of gross domestic product. The Henry review acknowledges community concern about such costs.

Not all churning is bad. The question for the Henry review is whether the benefits justify the costs. Income support payments are intended to enable recipients to sustain a reasonable living standard, having regard to changing community norms, without seriously undermining incentives for self-provision, including through work. Australia's strong reliance on means-tested payments has helped to contain budget costs, focusing support on those most in need. Phasing out payments as other income rises increases the tax rate, which can reduce the incentive to earn extra income.

Churn has increased during the past decade because the generosity of payments has increased and they are withdrawn more slowly to lower effective marginal tax rates. This is especially true of family payments and child-related payments.

Treasury's Intergenerational Report estimates that real gross domestic product a head is likely to grow about 0.5 per cent a year more slowly during the next 40 years (meaning fewer dollars to sustain living standards for the growing population of retirees and other non-workers). Relatively fewer traditional working-age individuals are expected among the total population and relatively fewer of them are expected to join or re-enter the workforce. Despite increased female participation, workforce participation is thus projected to fall significantly.

Overall, participation in Australia is comparable with most developed Western countries. Nevertheless, Productivity Commission analysis shows that we lag New Zealand and Canada in the case of prime-aged men; Canada, the US and Britain for women of child-bearing age; and New Zealand, the US, Britain and Canada for people nearing retirement. A revised tax and transfer system should provide stronger incentives for these groups to work or work more hours.

Although social norms and perceptions of what is best for young children matter, the size of the labour force is important as well. The evidence is that the group likeliest to respond to economic incentives is mothers.

The Melbourne Institute Tax and Transfer Simulator shows that those most sensitive to changes in effective marginal tax rates are single parents who have by far the highest elasticities, though the group is small in number. They are followed by married women (at all levels of household income) and low-income earners, especially single women. This is true whether the decision is to increase hours of work or to enter the workforce.

Work choices of men are much less sensitive to changes in effective tax rates. Although single men approaching retirement can be enticed back into the workforce, the elasticity is small and not evident for older men. It will be interesting to see whether tax-free superannuation payouts for the over-60s lead to different behaviour in this group; self-funded retirees now have an increased and customised tax-free threshold.

So what has been the overall effect on effective tax rates of the changes of recent years? The evidence is mixed: changes to the tax and transfer system between fiscal 2004 and fiscal 2009 have decreased effective average tax rates across some income ranges but increased it across others.

The resulting effect on labour supply is complex and depends on the characteristics of each household. The actual effect on hours supplied or workforce participation reflects the net effect of two competing forces: lower tax rates improve the return to working, which encourages greater labour supply, but also support greater leisure (reduced labour supply) by raising income. In addition, family payments have increased (raising income) and now are withdrawn over a wider income range, increasing the effective marginal tax rate across this range. Both effects reduce labour supply.

Inevitably some, especially couple households with children, face higher effective tax rates. For example, a secondary earner with one child under five who increases their income from $15,000 to $20,000 would lose an additional 18c or more in the dollar than previously (and almost twice that amount in some instances) if their partner's income was in the range of $20,000 to $60,000 (slightly above average male earnings). The effective rate faced by a secondary earner in a low-income couple family with children can be surprisingly high.

Consider a primary earner earning up to average male earnings. At virtually every income level a secondary earner would lose 40 per cent to 50 per cent to taxes and reduced benefits if they joined the paid workforce; more in some cases. The picture is less uniform for a secondary earner trying to decide whether to work additional hours. But effective marginal tax rates in excess of 50 per cent (and up to 74 per cent) are common.

The complexity of the calculations and the need to account for childcare costs (not considered in the above analysis) mean it is often not straightforward for a potential secondary earner to predict the net benefit to the household from joining the workforce or working additional hours. This complexity may itself induce conservative behaviour; that is, work fewer hours than would maximise family utility.

The simpler and more transparent the interaction between the tax and transfer systems, the easier it will be for workers to make informed judgments. The introduction of the low income tax offset, for example, was a cheap but confusing way to introduce a higher tax-free threshold for those at the bottom of the income distribution; it now does not completely phase out until well into the middle-income ranges.

The fastest way to reduce complexity and improve work incentives for secondary earners in low-income households is to raise the tax-free threshold or extend the income point at which family payments and benefits begin to reduce. However, these are expensive options that may extend churn further into the middle-income ranges and in practice often require some clawback by raising tax from higher-income earners, which may have perverse work incentives. It seems unlikely that a simpler system can be introduced without considerable losers or decreased revenue.

Stephen Sedgwick is director of the Melbourne Institute of Applied Economic and Social Research at the University of Melbourne.