household debt ballooned from 50 per cent of disposable income in the early
1990s to 150 per cent. Nevertheless, most econometric studies conclude that
superannuation has increased household saving. An oft-cited study estimates
the economy-wide saving offset at just 38 cents in the dollar (Connolly and
Kohler 2004). In 2011 the Treasury estimated national saving was 1.5 per cent
of GDP higher than it would have been (Gruen and Soding 2011).
But boosting saving does not necessarily justify superannuation. In an open
economy like Australia’s, with access to global capital markets, domestic
savings need not govern domestic investment. It may be argued that concerns
about high current account deficits and falling savings rates were always
misplaced (see Kirchner 2012). High current account deficits, a supposed
result of‘too little’domestic saving, could in fact welcome signs of strong
foreign demand to invest locally.
And lifting saving is not necessarily fair either. The fact that the legal
liability of the‘guaranteed’superannuation contributions falls on employers
obscures the incidence of compulsory superannuation. With an elastic supply
of capital, the economics of the incidence is clear: the burden falls on
employees, just as with a payroll tax paid by employers. Any extra saving
therefore arises largely as a result of lower take-home pay for low- to middle-
income earners, who cannot easily reduce other saving.
Notwithstanding the likely aggregate boost to overall saving, a popular
belief in a‘savings gap’in retirement underpins calls to lift the SG rate
further. In 2013 the supposed shortfall for the current working-age popula-
tion was $727 billion or $67,000 per person, according to one widely quoted
periodic study sponsored by the Financial Services Council. Yet such ana-
lyses typically ignore assets beyond superannuation, such as housing, equity
in business, otherfinancial assets, and expected bequests (Rothman 2011). It
also neglects the fact that several categories of expenses are typically absent
in retirement: including looking after children, commuting to work, and
mortgage payments.
In fact, the combination of the age pension, the SG, and other assets are
currently producing replacement rates between 50 per cent and 65 per cent,
with the lower percentages applicable to higher income earners for whom the
means-tested age pension is a smaller share of future income. These rates are
already around in the middle range of those found in OECD countries, where,
in 2011, typical replacement rates varied from 110 per cent in Greece to 30 per
cent in Ireland (OECD 2014). But they are projected to rise to 80 per cent or
more over the next twenty-five years.
‘None of these replacement rates assume any contribution to spending in
retirement from reverse mortgages or part time paid work’, notes Rothman
(2011, p. 21). It is revealing that potential income from the largest component
of retirees’wealth—their principal residence—is entirely excluded even from
We Must All Be Capitalists Now