| |
The Baby Bonus: A Dubious Policy Initiative
Ross Guest
Click here for PDF version
The Baby Bonus is an expensive way to alleviate a fertility decline that need not concern us, argues Ross Guest
In 2004, Treasurer Peter Costello urged Australian couples to have ‘one (baby) for your husband and one for your wife and one for the country’. To encourage this, the Budget that year put in place the ‘Baby Bonus’, a lump sum payment to the parents on the birth of each child. It has risen from $3,000 on commencement on 1 July 2004 to its current level of $4,000 and is set to rise to $5,000 on 1 July 2008. With 259,800 registered births in Australia in 2005, that amounted to eligible payments of around $780 million.
Australia, like most OECD countries, has an array of family policies that are ostensibly designed to achieve social welfare objectives, such as alleviating the cost to parents of raising children, but which may also have a pronatalist effect. Examples include Family Tax Benefits, the Child Care Benefit, Child Care Tax Rebate, legislated parental leave and, more broadly, publicly funded services for children such as public school education. Each helps removes financial obstacles to having children.
The link between the government payment and childbirth is more direct for the Baby Bonus than in the case of the other policies. It is not tied to any economic activity such as earning income, or expenditure on child care or education—it is simply conditional on the birth of a child. The Baby Bonus was also accompanied by government rhetoric, such as the Treasurer’s ‘one for the country remark’, about the need to increase the fertility rate. Costello made speeches strongly supporting a higher fertility rate and has argued that the Baby Bonus will help achieve that objective.[1] For both these reasons—the direct link between the payment and childbirth, and the government’s rhetoric accompanying the policy—a pronatalist intent can be attributed to the Baby Bonus which distinguishes it from other family support policies. Whether the Baby Bonus has policy merit as a way of increasing the number of births in Australia depends on the answers to three questions:
Do we need a pronatalist policy in Australia?
Will the Baby Bonus raise fertility?
Is the Baby Bonus a good pronatalist policy?
Is a pronatalist policy necessary?
A pronatalist policy cannot be justified simply on the grounds that the fertility rate has fallen, not even that it has fallen below replacement level. It is worth pointing out that there is no foreseeable chance of Australia’s population actually falling, unlike Italy, Russia and Japan for example. Below replacement fertility does not imply a falling population level in Australia due mainly to our immigration intake. Indeed even under the scenario with the lowest fertility rate (1.5 compared with the current rate of 1.8) and lowest immigration level (80,000 compared with 110,000 currently) projected by the ABS, Australia’s population would not fall by 2051 or even by 2100.[2] On current trends Australia’s population is projected to increase from 25 million to 33.4 million in 2051 and to 43.5 million by 2100.
Given that there is no foreseeable risk of Australia’s population dying out we must find other reasons for pronatalist policies.The argument for any kind of government interference in the private decisions of individuals—and one cannot get more private than the decision to have a child—has to be made on the grounds of market failure. That is, it has to be shown that the private decisions lead to sub-optimal outcomes from society’s point of view. In the context of pronatalist policies then, a natural first question is: what is the optimum population size for Australia? Then we might ask: are the unfettered choices of parents likely to put us on a path to the optimum population?
Unfortunately searching for the optimum population is a difficult exercise. Unlike The Hitchhiker’s Guide to the Galaxy where the answer to the ultimate question was revealed as 42, there is no uncontroversial answer to a country’s optimum population size. There are too many unknowns and unsettled ethical issues. The Jones Report on ‘Australia’s Population Carrying Capacity’ puts it this way: ‘the search for a magic figure [for optimum population] or a Rubicon between safety and danger is chimerical’.[3] In a similar vein, Peter McDonald and Rebecca Kippen argue that ‘it is not sensible to specify particular discrete population targets’ because of uncertainty about future fertility, mortality rates, environmental issues and other circumstances.[4] They argue that in any case our choices of a population target would be very limited due to ‘demographic realities’, being the realities of past fertility rates and achievable future rates, and of a limited range of immigration levels that would be both desirable and achievable.
Arguably, Australia is already on an appropriate demographic path. This idea is supported by Glenn Withers citing economic analysis showing that Australia’s current population growth rate of 1.25% is optimal for Australia’s income per capita.[5] A prominent example of such analysis is by John Neville who finds that the optimum rate of growth is equal to 1.36% per annum but that anywhere in the range of 1.1% to 1.6% per annum would be close enough to the optimum—that is, it would generate close to the optimum growth of income per capita.[6] This leads us into the field of population economics which we must look at further, if only briefly, if we are to properly evaluate the merits of a policy intended to boost the fertility rate.
The economic analysis of optimum population starts with the principle that optimum population size must balance two opposing forces: those that yield advantages to size and those that generate disadvantages to size. Advantages derive from economies of scale including those arising from public goods (like national defence and transport infrastructure) where the total cost of the ‘good’ is not affected by an increase in population up to a point thereby allowing a higher population to lower the cost per capita.[7] A larger market can also stimulate innovation by making it more profitable. A higher population also implies a greater probability of knowledge breakthroughs which have positive snowballing effects on productivity (a process of so-called endogenous growth).
Disadvantages to population size derive from the exhaustible supply of natural resources (land for example) which give rise to congestion and environmental costs and diminishing marginal productivity of labour. A bigger workforce can also imply cheaper labour which can discourage innovation in labour-saving technology. There is mounting evidence that productivity and fertility are negatively related, and perhaps in both directions.[8] One channel is where parents with more children spend less on education per child. This suggests that population growth driven by high fertility may lower productivity.
Balancing these advantages and disadvantages of bigger populations in order to find the net effect is an unresolved question both in theory and empirically, a point made in the seminal article by David Cutler, James Poterba, Louise Sheiner and Lawrence Summers.[9] But for a middle sized country like Australia which faces neither extreme congestion nor extreme under-population, a reasonable conclusion is that foreseeable variations in population size are not going to have a significant effect on productivity at least.
The analysis of optimum population must consider not just population size (or its growth), but also dependency. Analysis must consider the consumption needs of dependent young and dependent old people, and the fact that the proportions of young and old in the population are changing. There are two points.
First, a transition to a higher fertility rate would necessarily reduce consumption per capita for a number of decades. The reason is simply that the youth dependency rate rises while it takes some decades for this to be offset by a fall in the old age dependency rate. When the old age dependency rate eventually falls, consumption per capita recovers.
Second, once the population has become stable at a higher fertility rate after many decades, the total dependency rate (youth plus old age) is likely to be similar to that for a lower fertility stable population.[10] Hence the difference in living standards wouldn’t be much different either.[11] So we suffer a loss for several decades, followed by a recovery, then once population becomes stable we are more or less back to where we were in terms of total dependency. A qualification is that this ignores the issues of economies and diseconomies of population size referred to above—it simply refers to the dependency effects on living standards.
A further consideration is that a transition to a higher fertility path has implications for intergenerational equity due to productivity growth over time. The ongoing effects of productivity growth through technological progress ensures that living standards will continue to grow irrespective of our fertility rate, as the Productivity Commission and others have pointed out.[12] According to the Productivity Commission Australia’s living standards are expected to be nearly double their current level with current demographic trends, notwithstanding population ageing. This means that a transition to higher fertility has implications for intergenerational equity. We saw above that a boost to fertility would impose a cost on people alive now and in the near future, which will reverse itself in several decades, boosting the living standards of people who would in any case have been twice as well off as people today. This seems overly generous to future generations at the expense of current generations.
What can we say in summary about the need for a policy to boost the fertility rate? The evidence is that there is no need for one because we are on, or close to, an appropriate demographic path already. Also, a higher population growth rate will have little effect on total dependency in the long run, and along the transition path it will penalise current generations and reward future generations who will be richer anyway. Issues about the net effect of economies versus diseconomies of larger populations are relevant but unresolved, and hence can be ignored for practical policy purposes, especially for a country like Australia.
Will the Baby Bonus raise fertility?
Now, if everyone agreed with this laissez-faire view we wouldn’t have had the Baby Bonus to boost fertility. So let us assume that the above analysis and/or conclusion is flawed and that we do need a pronatalist policy. This leads us to the second question: Will it work?
International empirical evidence suggests that public policies generally do boost fertility, although part of the measured effect may be a timing effect where mothers bring forward childbirth in their lifecycle rather than increasing the number of births over their lifetimes (their so-called completed fertility). The Rand Corporation outlines the European evidence[13] and a larger survey of European public policies has also been completed by Anne Gautheir.[14] There is quite strong evidence that Canada’s Baby Bonus, which is very similar to Australia’s in amount and structure, increased fertility—Kevin Milligan found that it increased fertility by 25% for families entitled to the full benefit.[15] There is also evidence for the US that tax benefit policies between 1948 and 1997 were associated with an increase in fertility.[16]
For Australia, the birth rate has edged has edged up to 1.81 in 2005 from 1.75 in 2003 before the revised Baby Bonus was introduced.[17] However we can’t read much into this because the birth rate does jump around a bit—it was 1.82 in 1995 and dropped to 1.73 in 2001. There is some evidence of an announcement effect of the Baby Bonus in Australia—Joshua Gans and Andrew Leigh estimate that over 1,000 births were delayed, by rescheduling of inducement and cesarean procedures, to ensure that the birth occurred after 1 July 2004 in order that the parents were entitled to the Baby Bonus.[18] Although this is evidence that parents respond to incentives in making family planning decisions, it is not evidence that completed fertility rates will respond to the Baby Bonus.
In summary, we can say that based on the international evidence the Baby Bonus will probably boost fertility but we can’t say by how much.
Expensive babies
However, from a policy evaluation perspective it is not sufficient to ask whether the Baby Bonus will increase the fertility rate. Like any public spending programme the Baby Bonus must be ultimately financed by taxation which imposes a deadweight loss on society by distorting private choices, in particular the work/leisure choice. Hence public spending programmes should be delivered as efficiently as possible, meaning with minimum deadweight loss for a given outcome. This is where the research on the Canadian Baby Bonus by Kevin Milligan is relevant.[19] Milligan concludes that the cost of the Canadian Baby Bonus was high. He econometrically estimates the number of extra children born as a result of the child benefit and divides this into the total child payments paid out. He finds that each extra birth ‘cost’ $15,000. The amount is high because a lot of parents—about 88 percent—would have had children without the cash payment. The Baby Bonus that those parents received was therefore ‘wasted’.
The policy lesson to draw from the Canadian study is that even if the response of a pronatalist policy is strong the cost can be high. In Australia’s case, it was pointed out in the introduction that the number of births in 2005 multiplied by $3,000 implies a cost to the public purse of around $780 million. This expenditure should be compared with alternative policies designed to achieve the same outcomes, and ultimately with all alternative expenditure programmes on a cost-benefit basis.
It would be possible to better target the Baby Bonus in order to minimise the ‘waste’ of expenditure that is paid to parents who would have had a baby anyway. One option is to pay parents only for their second and third children, as applied in the initial version of Singapore’s Baby Bonus introduced in 2001; or perhaps only their third child on the basis that most people who have one child have a second.[20] In Australia, this might be seen by some as an excessive degree of social engineering. But is it really any different from many other tax and transfer policies which have conditions and rules attached to them in order to alter our behaviour in some way? For example, a parent with a child aged over six is only eligible for the parenting allowance if they work (as well as meeting other conditions). Yet spouses caring for children are only eligible for the Family Tax Benefit B if they are not working. The potential conflict between these two policies is another issue.[21]
With respect to ‘waste’, concerns have been raised about how the Baby Bonus is being spent, especially by very young parents, with anecdotal evidence about splurges on plasma TVs and similar items. In response, from January 2007 mothers under 18 years will receive the Baby Bonus in fortnightly instalments over a six month period, presumably in order to discourage the spending on lumpy luxury items. Economists tend to view with some equivocation rules aimed at directing the spending by recipients of transfer payments. On the one hand, rules that constrain people’s choices may lower their ‘utility’. This is the case for transfers to be paid as cash lump sums if practical, thereby providing maximum flexibility in allowing households to manage their budgets. Pensions and unemployment benefits are examples of unfettered cash payments, albeit paid fortnightly.
On the other hand, private expenditure on certain items can be sub-optimal from society’s point of view without public policy intervention. This is the argument for tied transfers such as tax expenditures (for example, the Child Care Tax Rebate) which are designed to encourage private expenditure. Tax expenditures are a form of government co-contribution. The Singapore Baby Bonus has a government co-contribution in addition to the unfettered cash payment. The co-contribution is a dollar for dollar matching of parents’ saving into a special account for the child’s education, the Children Development Account. This is an option for the Australian Government if it is concerned about the way the Baby Bonus is being spent. Having said that, it conflicts with the other principle mentioned which is to allow maximum parental choice through unfettered cash payments.
More generally, attempts to target the Baby Bonus to reduce ‘waste’ as Kevin Milligan puts it would make it more complex and add to the already complex array of family policies.[22] There is a view that the Baby Bonus, child care payments and other family payments have already become far too complex and in some cases conflicting, leading to calls for the whole set of policies to be replaced by a single universal cash payment per child.[23]
In summary, is the Baby Bonus a good pronatalist policy? The negatives are that it wastes expenditure by paying parents who would have had children anyway, and a more debatable negative is that it allows parents to spend the money on whatever they want which may include plasma TVs and so on that may not benefit the child. The positives are that it is simple and transparent, and it allows maximum parental choice with regard to expenditure. My view is that Singapore got it about right with its initial model, in two ways: by paying only for the second and third child, and by having two components—a lump sum cash payment and a co-contribution toward education expenses. Not paying for the first child partly alleviates the ‘waste’ problem, and the economic case for encouraging large families by paying a Baby Bonus for fourth and subsequent children is weak due not least to concerns about the average investment by parents in education per child when there are large numbers of children.
Conclusion
This article has argued that, on the balance of the evidence, Australia is on or close to an appropriate demographic path, and so there is no need for a pronatalist policy. However, other countries, especially in Europe but also Singapore, have explicit pronatalist policies; and in Australia there is a view that we should also have one. The Australian Baby Bonus is, however, an expensive way of lifting the birthrate. Though it may slightly increase the number of children Australian women have, for the most part it will reward parents who would have had children anyway. If it was restricted to second or third children, it could have the same effect at the margins at much less cost to taxpayers.
Endnotes
[1] See for example, Peter Costello, Address to the media upon the release of the 2004 Federal Budget, Canberra, 11 May 2004; and the Launch of the 2006 Census of Population and Housing, National Press Club, Barton, Canberra, 24 July 2006.
[2] ABS (Australian Bureau of Statistics), Population Projections Australia, 2004 to 2101, ABS Cat No 3222.0 (Canberra: ABS, reissued 14 June 2006).
[3] Australian Parliament, The Jones Report—Australia’s Population ‘Carrying Capacity’: One Nation--Two Ecologies, report by the House of Representatives Standing Committee on Long Term Strategies, (Canberra: Australian Government Publishing Service, 1994), p 129.
[4] Peter McDonald and Rebecca Kippen, Population Futures for Australia: the Policy Alternatives, Parliament of Australia, Research Paper No 5 (Canberra: APH, 12 October 1999), available at http://www.aph.gov.au/library/pubs/rp/1999-2000/2000rp05.htm.
[5] Glenn Withers, ‘Steady As She Grows’, The Australian, 25 March 2002.
[6] I am grateful to Glenn Withers for referring me to John Neville’s 1990 paper: see John Neville, The Effect of Immigration on Australian Living Standards, Bureau of Immigration Research (South Carlton: Commonwealth of Australia, 1990).
[7] I include impure public goods here. The key requirement in order to provide advantages from population size is that the good be non-rival in consumption up to a point. Transport infrastructure is an example though it is not a pure public good (like national defence) because it is excludable.
[8] Recently Ross Guest and Robyn Swift in Fertility, income inequality and labour productivity (Brisbane: Griffith University,2006) (available on request from authors), find, using cointegration analysis, a negative bi-directional relationship between fertility and productivity in the long run for both Australia and the United Kingdom. See also Oded Galor and Hyoungsoo Zang, ‘Fertility, Income Distribution, and Economic Growth: Theory and Cross-Country Evidence’, Japan and the World Economy, 9:2 (1997), pp 197–229; and Gary Becker and Robert Barro, ‘A Reformulation of the Economic Theory of Fertility’, The Quarterly Journal of Economics 103:412 (February 1988), pp 1-25 for one-directional effects.
[9] David Cutler, James Poterba, Louise Sheiner and Lawrence Summers ‘An Aging Society: Opportunity or Challenge?’, Brookings Papers on Economic Activity (Spring 1990), pp 1–74.
[10] A stable population is one where the age-specific fertility and mortality rates have been constant for long enough that the age shares of the population have settled down to constant levels, and the population growth rate has become constant.
[11] See David Weil, ‘Population Growth, Dependency and Consumption’, American Economic Review, Papers and Proceedings 89:2 (May 1999), pp 251–255, for an analysis of these effects.
[12] See, for example, Ross Guest and Ian McDonald, ‘Would a Decrease in Fertility be a Threat to Living Standards in Australia’, Australian Economic Review 35:1 (March 2002), pp 29–44. See also PC (Productivity Commission), Economic Implications of an Ageing Australia, (Canberra: PC, April, 2005).
[13] Rand Corporation, Population Implosion? Low Fertility and Policy Responses in the European Union, Rand Corporation Research Brief (2005), available at www.rand.org/pubs/research_briefs/2005/RAND_RB9126.pdf
[14] Anne Gauthier, Public Policies Affecting Fertility and Families in Europe: A Survey of the 15 Member States, Paper prepared for the European Observatory on Family Matters, Annual Seminar, 15–16 September 2000, Sevilla, Spain.
[15] Kevin Milligan, Subsidizing the Stork: New Evidence on Tax Incentives and Fertility, University of British Columbia Department of Economics Discussion Paper No 01-21 (2002); and Kevin Milligan, ‘Quebec’s Baby Bonus: Can Public Policy Raise Fertility?’, Backgrounder, C D Howe Institute, Toronto (24 January 2002).
[16] Cristobal Ridao-Cano and Robert McNown, ‘The Effect of Tax-Benefit Policies on Fertility and Female Labor Force Participation in the United States’, Journal of Policy Modeling 27:9 (December 2005), pp1083–1096.
[17] ABS (Australian Bureau of Statistics), Births Australia, 2005, ABS Cat No 3301.0 (Canberra: ABS, reissued 17 October 2006).
[18] Joshua Gans and Andrew Leigh, Born on the First of July: An (Un)natural Experiment in Birth Timing, Centre for Economic Policy Research, Research School of Social Sciences, Australian National University, Discussion Paper No 529 (2006).
[19] Milligan, Subsidizing the Stork, Discussion Paper No 01-21 (2002); and Milligan, ‘Quebec’s Baby Bonus’, Backgrounder.
[20] The initial version of the Singapore Baby Bonus was extended in August 2004 to the first and fourth child, but not to fifth or subsequent children. The Singapore policy also currently pays a higher amount for the second child than the first, and an even higher amount for the third and fourth child.
[21] This conflict has been identified by Peter McDonald as reported by Mike Steketee in ‘Population up as biological clocks chime to baby bonus’, The Australian, 20 October 2005.
[22] Milligan, Subsidizing the Stork, Discussion Paper No 01-21 (2002); and Milligan, ‘Quebec’s Baby Bonus’, Backgrounder.
[23] Peter McDonald, ‘Reforming Family Support Policy in Australia’, People and Place 11:2 (2003), pp 1–15.
|
