As a guardian, if not a parent, of the Association of Superannuation Funds of Australia Retirement Standard, I can get a bit offended when people say unkind or inaccurate things about it.
This happens from time to time. The ASFA Retirement Standard is a victim of its own success, in that its almost ubiquitous use as a goal for retirement incomes is seen as confronting by those who want their own child (product) to be regarded as the best and brightest in the class. While it is the right of every product provider or distributor to assert the merits of their own product, this does not require criticising other approaches.
The attacks have come from both ends of the scale. For funds and advisers dealing with high-net-worth individuals, the ASFA comfortable standard of living’s level of expenditure might cover only the restaurant bills and overseas travel of their clients. At the lower end of the income and wealth scale, the argument has been made that the ASFA comfortable level is unobtainable and that, a bit like pictures of models in magazines, can lead to less financially attractive individuals developing a poor self-image of themselves and, consequently, depression.
ASFA has never claimed that the comfortable budgets are appropriate for everyone. Amongst other things, we publish the poor cousin of the comfortable budgets – what we call the modest level. However, the modest budgets have much to be modest about. These budgets are set at a level not much more than the age pension or the estimated incomes needed to avoid poverty in Australia.
In its submissions to the government on what the objective for retirement incomes, ASFA has suggested a multi-pronged approach. For low-income earners, something around the level of the age pension might be appropriate. For the bulk of the market, ASFA’s comfortable level is an appropriate benchmark. And for high-income earners, retirement expenditure based on a percentage of pre-retirement income, with that percentage declining with increasing wealth, might be appropriate.
Wade Matterson and Jeff Gebler, from Milliman, in a recent article in Professional Planner, have both criticised the ASFA standard and put forward a proposal for a retirement expenditure goal linked to current patterns of household expenditure. Such an approach is arguably both a step back and makes use of limited and largely irrelevant data.
If you want to know how current retirees live, you do not have to trawl through the household expenditure survey data the ABS publishes. All most of us have to do is look at the lifestyle of our parents or grandparents, depending on your age.
Most current retirees have a lifestyle that is not much above the poverty level, with Australia having one of the highest relative poverty levels for the aged in the world. That is not something we should aspire to for future generations of retirees.
Surveys of the population show that the bulk of those now employed aspire to a standard of living in retirement commensurate with the level ASFA calls comfortable. Big data is about better understanding the circumstance of your customers and other members of their household and then applying that knowledge. That is what financial planners customarily do.
Some have criticised the ASFA for using single figures for retirement savings needed to support different levels of income. Actuaries, in particular, like stochastic modelling of outcomes, as it makes use of their great strengths at putting together spreadsheets and historical data. However, scattergrams of results are not good communication tools, as they look more like tests for colour blindness.
Also, the future is not a Monte Carlo simulation of the past. We are entering an era of economic outcomes and investment outcomes with low interest rates and low rates of inflation quite unlike what’s come before. In addition, the asset test for the age pension, while it has many less-than-desirable attributes, means that for most people the government is taking on investment risk. In good years, the government takes the benefit of improved investment returns. In low-return years, it has to pay out more age pension. This interaction is rarely modelled in stochastic exercises, because it is too hard and detracts from the primary argument of the proponents of such approaches.
The ASFA Retirement Standard can be improved and a review of it is being undertaken to make marginal adjustments. However, other competing approaches to setting retirement goals have many and major limitations.
Ross Clare is director, research and resource centre, for the Association of Superannuation Funds of Australia.
