The Costs to Australia of Renewable Energy

Submission to the Senate on the proposed 20 per cent energy requirement

KEY POINTS

The proposed renewable requirement would markedly raise the price of electricity:

• imposing a direct cost on the economy of $1.8 billion annually;
• increasing consumers' costs and reducing industry competitiveness;
• causing far more jobs to be lost than the subsidised jobs created.

INTRODUCTION AND SUMMARY

The Commonwealth Government's proposal to require 20 per cent of electricity to be derived from renewable sources offers Australia no advantage to while crippling the competitiveness of its energy supply.  It raises the average cost of generated electricity by 10 per cent, imposing a deadweight loss on the economy of $1.8 billion per year.

If the proposed Emission Trading Scheme (ETS) is also introduced, the 20 per cent renewable requirement will not even bring a net reduction in carbon emissions.  It would merely change the composition of that reduction, making it more expensive.

Ten years ago, with the National Electricity Market Australia had developed the world's lowest cost electricity supply industry.  This stemmed from deregulation and privatization combined with Australia's natural advantages in energy sources.  The outcome was of immense value to the nation's economic competitiveness.  We are progressively dismantling this asset.  Instead of facilitating the use of low cost energy, government regulations are increasingly subsidising high cost energy sources.

The proposed ETS tax would compound these adverse effects.  But, at least revenues from the ETS are to be directed back to people and institutions the government deems to be most needy or meritorious.  By contrast, the mandatory renewable requirements impose a hidden tax on the consumer, the revenues from which are given to the suppliers of intrinsically uncompetitive renewable energy sources.

Evidence from overseas shows that any jobs created from a forced requirement to use mandatory renewables come at a cost in excess of $200,000 each.  Moreover, Spain's experience indicates for each subsidised job created, there are 2.2 jobs lost.

Mandatory renewables are promoted by a classic combination of "Baptists and bootleggers".  The bootleggers, those with commercial interests in wind farms and other high cost renewables, like the protectionist lobby before them, point to visible developments in facilities built on the back of the economic distortions they seek.

But the damage from imposing high cost electricity is much greater than was the case with external tariff protection because of electricity's ubiquitous presence throughout the economy.  Moreover, unlike manufacturing facilities previously made possible by restraining import competition, wind farms and other renewable technologies can make no credible claims of being nurseries for infant industries that will mature into productive activities.

AUSTRALIA'S PROPOSED 20 PER CENT
RENEWABLE TARGET IN CONTEXT

TYPES OF AUSTRALIAN EMISSION REDUCTION MEASURES

The renewable target is being addressed at the same time as other measures which focus upon emission reductions.  Chief among these is the cap-and-trade scheme incorporating a carbon tax, and a range of budgetary expenditures by the Commonwealth and State Governments.  In all, there are five classesof programs designed to reduce Australian emission levels:

• direct funding from the budget for incentives to reduce emissions;
• the regulatory price based cap-and-trade proposal;
• requirements for specific amounts of renewable energy to be incorporated in the electricity that retailers supply to the consumer;
• funding promotional measures designed to encourage consumers voluntarily to reduce their emission levels;  and
• requiring businesses to incur expenditures through energy audits with a view to their discovering new means of saving energy or justifying reasons why such savings are not being made.

The last two classes are not further addressed in this paper.

GOVERNMENT BUDGETARY EXPENDITURE

The Commonwealth Government is spending considerable public funds on measures to foster lower energy use.  As there is no shortage of energy per se that normal price developments do not signal, these sums can be designated as focussed upon the objective of reducing greenhouse gas emissions.

During the current financial year some $2.6 billion is identified in the Department of the Environment (DoE) budget.  Among these are the "Clean Energy Initiative" and the "Climate Change Action Fund".

The Department of Climate Change (DCC) elects not to include some of the DoE measures as directed at climate change and therefore publishes a lower costing (just over $1 billion).  Excluded from the DCC summation is important expenditures on home insulation and low emission assistance for renters which have over $1.4 billion budgeted in the current year and 2010/11.  These formed part of the fiscal package designed to avert the effects of the global financial crisis on Australia.

Aside from their emission abating functions, a rationale for these expenditures is "market failure" to produce the goods and services that consumers generally want.  While it is possible that all classes of the $2.6 billion of expenditures undertaken by government on our individual behalves represent our best interests, such spending choices have seldom proven to be optimal in the past.  Governments have long histories of promoting particular industry sectors through tariffs and subsidies but such "winner picking" is now generally considered to have been misplaced.

In the case of energy industries, the widespread belief 30 years ago that there was a looming energy supply crisis brought government subsidies and regulations that were designed to pressure decision takers into making energy savings they would not have willingly made.  Though there are benefits in energy savings, these are rarely costless and usually have a negative worth when they involve capital expenditures with a zero or very low payback.

THE EMISSION TRADING SYSTEM (E.T.S.)

The ETS incorporated within the Carbon Pollution Reduction Scheme is the only one of listed classes of abatement policy that is an "economic instrument" in the accepted definition of the term.  Economic instruments are almost universally considered to be more efficient than deterministic command-and-control measures which mandate specific solutions.  Regulations that specify particular technologies or types of technologies fail to make use of the power of markets to discover the lowest cost ways of meeting the goal.

Economic instruments use taxes or quantitative goals to allow the objectives to be met most cheaply.

Taxes and quantitative limits both bring additional costs.  These are passed on in prices and cause consumers to reduce their demand for the product incorporating the targeted constituent.  This effect on gross demand is complemented by further abatement resulting from supply switching;  with the ETS, that means from the fuels that are high in carbon emissions relative to the energy they produce, to those that are lower in carbon emissions.

The supply switching effect is transmitted differently by a tax as opposed to a quantitative limit.  With a tax, supply switching is achieved by the charge's greater direct cost on the supplies that use the taxed component most intensively.

With a tradeable rights approach, owners of the outputs that use the taxed component most intensively have an incentive to switch their production profiles or sell the emission rights to others who have a more favourable input profile.  This indirectly increases the cost of the products that use the targeted input most intensively, causing it to be substituted by products where the targeted input is less intensively present.  In both cases the switching between product classes reduces the average rate of tax per unit of product.

The switching process is readily seen with electricity generated by gas rather than coal.  The former has only half of the latter's rate of carbon dioxide emissions per unit of electricity.  As a result, the impost on electricity generated by gas is only half of that on coal based electricity and gas will improve its competitiveness because it faces a lesser impost and will increase its market share, thereby reducing aggregate emissions for a specific level of electricity generated.

The degree to which the price increase brings a demand reduction rather than a switch in types of supply depends on a number of factors.  Importantly, these include the relative ease by which carbon intensive energy supplies can be substituted and the responsiveness of consumers to higher prices.

Economic instruments therefore pursue the goals by one of two routes.  The first involves setting a rate of tax per unit of emission and allowing the quantity to be determined by the market.  The second sets a quantity and allows the market to set the price.  Normally, whether a tax or a quantity is the parameter that is set, the government will have in mind a likely outcome that is acceptable in terms of the reciprocal factor.

Like other countries, Australia has opted to introduce an Emissions Trading System.  Emission rights are to be sold with some given away to particular users who are judged likely to suffer excessive costs from the higher impost they would be required to bear.  As evidenced by the Queensland Premier's call on 15 July for special treatment for the State's "gassy" mines, should the scheme become operational, there are likely to be considerable refinements to it.

As with most ETS schemes, that of Australia incorporates a fall back price at which the government will make unlimited permits available.  That price for Australia is proposed to be $10 per tonne of CO₂-e in the first year and an indexed $40 per tonne of CO₂-e in subsequent years ($57 in after-tax terms).

MANDATORY REQUIREMENTS ON ELECTRICITY USAGE

ADVICE OFFERED ON MANDATORY REQUIREMENTS

The Productivity Commission (PC) in its submission to Garnaut ⁽¹⁾ expressed its strong reservations about the merits of mandatory incorporation of specific technologies in its usual understated and measured manner.  It did so in referring to the MRET scheme for promoting renewables.  The PC's conclusions were

"with an effective ETS in place, the MRET would:

• not achieve any additional abatement but impose additional costs
• most likely lead to higher electricity prices
• provide a signal that lobbying for government support for certain technologies and industries over others could be successful."

The PC examined many claims where a measure other than a price mechanism like the ETS might be supplemented by other regulatory arrangements and concluded that none of these were persuasive.  For example it might be argued that there is an asymmetry of information between the market and those more knowledgeable of future trends in government.  However, such claims have seldom in the past been born out and considerable costs have been incurred by taxpayers or through regulatory measures in supporting particular industries or technologies that turned out to be blind alleys or inappropriate for Australia.

As the PC argued, an MRET scheme in conjunction with renewable targets would have no effect on the aggregate level of emissions since these are set by the ETS scheme.  At best, it could lead to a zero net cost but only if the approach to emission reductions were to follow the renewable route completely on the basis that other reduction approaches, like substituting gas for coal as a generation fuel, were higher cost.

Other rationales for an MRET scheme have included the greater degree of self sufficiency it offers and that it supports an infant industry.

The self sufficiency case is seldom of merit in an inter-dependent world and, in any event, is irrelevant to Australia which is a net energy exporter and certainly has no shortage of local low cost coal and gas supplies suitable for electricity generation.

Support for an infant industry is often the refuge of the politico-economic scoundrel.  In Victoria the state scheme was promoted as the harbinger and catalyst of a whole raft of new industries particularly involving the manufacture of high technology blades.  The outcome is discussed below.

CURRENT AUSTRALIAN SCHEMES

The Mandatory Renewable Energy Target (MRET) From relatively modest beginnings there has been a considerable escalation in subsidies to renewable energy.  When Prime Minister John Howard announced the proposal to introduce a scheme in 1997 he said it was for an additional 2 per cent of electricity that was to come from "renewable or specified waste energy".  Although an additional 2 per cent by 2010, on the basis of the consumption that year of some 250,000 GWH would have been 5,000 MWh, the MRET target was reinterpreted and almost doubled to be 9,500 GWH.

A review of the scheme in 2004 (the Tambling Review) recommended it be increased to 20,000 GWh by 2020.  In June 2004 the Commonwealth announced that it did not believe expanding the target was economically justified and that it did not intend to increase the requirement.

NSW NGAC

The NSW Government introduced its NSW Greenhouse Gas Abatement Certificates (NGAC) scheme in 2003 which set a benchmark for emission standards of 7.27 tonnes of CO₂ per head.  On top of the Commonwealth MRET scheme, based on business-as-usual estimates of market growth and carbon intensity of electricity generation, an additional 19% of electricity in NSW must be derived from low carbon emitting sources.

Queensland GEC

Queensland has its 13 per cent GEC scheme under which 13 per cent of the non-exempt load must be derived from gas based electricity generation.  The exempt load is 9,000 GWh hence, after also deducting for loss factors in transmission and distribution, the requirement covered 75 per cent of the 2005 load.

Victoria's VRET

Victorian Premier Bracks in November 2005 argued that there was a, "lack of national leadership" by the Federal Government in not increasing the MRET scheme from the 9500 GWh target set for 2010 and said this, "is costing Victoria – economically and environmentally -- and cannot be allowed to continue." Mr Bracks argued, "Victoria's aim to facilitate the development of up to 1000 megawatts of wind energy by 2006 represents $2 billion worth of capital investment.  Then there are the jobs and the other economic spinoffs that accompany such a significant outlay".  The VRET scheme requires an additional 3,274 GWh of eligible renewable electricity by 2016.  It was expected to create "up to 2,000 new jobs, most of them in regional Victoria".  An indexed $43 penalty is imposed on shortfalls in a regulated entity's obligations.

At the time of the initial discussions of mandatory requirements for renewables, many among the wind farm lobby were confident that eventually wind would prove competitive with fossil fuels even without a subsidy.  That view was even shared by many climate sceptics for example in 1995 in The Skeptical Environmentalist, Bjorn Lomborg noted how windmills' productivity had improved, and "In the long run they will undoubtedly become competitive and even cheaper (than fossil fuel plants)". ⁽²⁾  No reputable authority would claim that today.  Indeed, anticipating a continued lack of competitiveness in the economics of renewable energy, the 20 per cent renewable proposal increases the after tax fall-back price from the MRET's $57 to $93 per MWh.

The Victorian scheme's hopes for new cutting edge businesses manufacturing blades and other components have been highly disappointing.  Thus the Pacific Hydro wind farms were to be the springboard for a Vestas blade factory in Portland, and were expected to generate 400 direct and indirect positions according to the sponsor, by mid 2006.  In spite of extensive State and Federal subsidies the project folded.

Victorian Industry Minister Theo Theophanous blamed the Commonwealth for failing to invest in renewable energy, "The reason why Vestas has been unable to continue its operations in Portland is very squarely and directly as a result of the Federal Government's refusal to extend its renewable energy scheme to allow the industry to expand," he said on the ABC. ⁽³⁾

This was rejected by then Commonwealth Minister Ian Macfarlane who argued, "Our Government has provided through the MRET scheme some $5.5 to $6 billion in terms of assistance to the renewable energy sector".  Mr Macfarlane added, "In particular with Vestas, [we] have provided them with tax relief over the last two years of around $10 million to ensure they are able to be competitive with the turbines they do produce." ⁽⁴⁾

Victorian Energy Minister Theophanous also announced the Victorian Wind Energy Network central to which was to be a blade manufacturing facility by Bolwell Corporation at Ararat.  This too folded in spite of extensive government support.

THE SPANISH EXPERIMENT WITH RENEWABLES

The Spanish experience with solar and wind power subsidies has been more far-reaching in relation to the market's size than that seen anywhere else. ⁽⁵⁾  Spain embarked upon a vigorous subsidy program which was intensified following the election of a new Socialist Government under Prime Minister Zapatero in 2004.  By 2008, electricity production was as shown in Table 1.

Table 1 Spain electricity generation share

Energy Source	Per cent
Combustible	64
Nuclear	20
Hydro	7
Wind and solar	9

Source OECD;  http://www.swivel.com/data_columns/show/10362480?value=Spain

Wind and other exotic renewables had shown a rapid increase in market share and claimed to be at 11 per cent in the first half of 2009.

However, as demonstrated by the work of Gabriel Calzada Alvarez and his researchers, the outcome has contributed to an economic disaster in Spain.  At 18.1 per cent Spain's unemployment rate is currently the highest in the OECD area, whereas previously it was close to the average.  That regrettable outcome is not solely related to the Spanish Government's aggressive pursuit of wind power investment but this was most certainly a significant factor.

Installed renewable power was targeted at 20,155 MW in 2010 – threefold Victoria's brown coal generation capacity and similar to Australia's entire black coal generation capacity.  At the end of 2008 over 15,000 MW of wind and 4,000 MW of solar photovoltaic capacity had been installed.

A second objective was to have 20 per cent of electricity consumption from renewable sources.  Including hydro-electricity, renewable energy was close to that figure in 2008.

The issue that Alvarez addresses is the cost and consequences of that seemingly successful performance.

In the case of wind power there is a complex remuneration system (which actually encourages the installation of under-sized units).  In 2008 the average cost of wind exceeded the pool price by 59 per cent.  In the case of solar the increase was far greater at sixfold.

Alvarez models the effect of these cost increases on jobs throughout the economy.  He concludes that though the stimulus to renewables has resulted in increased jobs, especially in construction, the increased cost of electricity has resulted in 2.2 jobs being lost for every job created.  He also estimates the cost of each "green" job was €571,000 and the cost of each wind industry job €1 million.

In terms of industry, Alvarez points to several high profile relocations of firms from Spain in response to the higher costs of energy.  These include Hoeschst, Sidenor, Ferroatlantica and the stainless steel company Acerinox (which relocated its production to the USA and South Africa).

Spain's experience, though involving higher costs than others, has not been unique.  Indeed, according to the Rhine-Westphalia Institute for Economic Research, "Each of the 35,000 solar jobs in Germany, for instance, is subsidized to the tune of €130,000." ⁽⁶⁾

EFFICIENCY OF DIFFERENT ELECTRICITY SOURCES

COSTS BASED ON AUSTRALIAN EXPERIENCE AND ESTIMATES

In Australia, South Australia is home to most of the wind generation installed and under consideration.  This is due to the state having the most advantageous wind conditions and because its pool prices tend to be a little higher than those of most other states (as a result of its reduced availability of coal).

Some 326 MW of wind generation is installed or planned around Lake Bonney in the state's south west and much of the rest is planned near existing transmission lines north of Adelaide.  Capacity factors average around 27 per cent.

South Australia's Electricity Supply Industry Planning Council (ESIPC), now incorporated within the Australian Energy Market Operator (AEMO), estimated that the state has 880 MW of wind generation projects "advanced" with 295 MW considered close to commitment. ⁽⁷⁾  That approaches one third of the state's capacity and though, because of wind's unreliability, it would never supply that share of electricity, in 2008/9 it contributed 14 per cent of supply.

Work conducted for ESIPC by KPMG estimated future supplies as follows.

KPMG's forecasts show SA's wind capacity rising from 740 MW in 2008-09 to 2,162 MW in 2012-13, without any allowance being made for an extension of the RET target to 45,000 GWh by 2030.

ESIPC also demonstrates one of the outcomes of Australia's National Electricity Market's "energy only" approach to pricing.  The pool price received closely approximates the worth of the energy.  Because of the unreliability of wind energy, it tends not to be available at the highest priced occasions – sometimes because it is at full capacity already, often because high price events coincide with no or excessive wind.  As a result, wind tends to attract a lower average price than more flexible forms of generation.  In the first half of this year wind's average price was $48.56 per MWh, some 53 per cent below the $74.26 per MWh received by other South Australian generators.

Table 2 below illustrates the different prices over recent years.

Table 2:  South Australian pool prices for wind and other generated electricity

Year	Volume Weighted Price for Wind Generators		Volume Weighted Price for Other SA Generators
	Full Year ($/MWh)	Summer ($/MWh)	Full Year ($/MWh)	Summer ($/MWh)
2004-05	NA	NA	39.25	32.62
2005-06	32.57	39.59	43.91	67.50
2006-07	49.69	51.55	58.71	67.21
2007-08	63.31	63.94	102.01	149.92
2008-09*	48.56	91.80	74.26	165.28

Source ESIPC

In addition to this, the priority in scheduling of most wind generation tends to distort the costs of other generators by forcing them to back off at periods when they would otherwise be operating, thus increasing the average costs of supply for these units.

The increased wind generation availability also has a cost in terms of networks, which need to be upgraded to carry the increased wind based electricity output that is available at times that are propitious for its generation.  This essentially means designing a network to carry less average power than is the case with a network carrying output solely from conventional fossil fuelled power stations.  Those costs are likely to be carried by consumers in general under the "regulatory test" applied for new transmission investment.

Based on data prepared by Carbon Market Economics for AEMO, the costs of wind are currently around $93 per MWh. ⁽⁸⁾

AEMO also has considerable detail prepared annually by ACIL Tasman on the costs of other generators.  Excluding carbon costs, in real 2009/10 dollars, these vary from around $47 per MWh for conventional black and brown coal generators to a little over $50 per MWh for gas, $70-80 for black and brown coal incorporating carbon capture and storage, ⁽⁹⁾ and around $100 for nuclear.  These costs for the coal based generators are inflated by the high world demand for new power stations on the back of very rapid growth in China and appear to have declined somewhat in 2009.

The carbon capture and storage costs (incorporated in the "IGCC black CCS_SWNSW" row below) are speculative and are likely to be in excess of those incorporated into the modelling.

Table 3 illustrates comparative costs.

Table 3:  Comparative generation costs

	2010	2012	2014	2016	2018	2020
CCGT (WC)_SWQ	55.63	50.41	50.55	50.66	52.14	52.57
SC BLACK (AC)_CQ	52.78	50.71	49.35	48.78	48.35	47.94
SC BLACK (AC)_NCEN	51.85	49.86	48.62	48.13	47.78	47.43
SC BROWN (WC)_LV	50.22	48.33	47.33	47.04	46.88	46.72
IGCC BLACK CCS_SWNSW	97.24	89.65	84.12	79.96	76.09	72.87
USC CCS BROWN (WC)_LV	82.94	77.35	74.59	73.16	71.80	70.87
Nuclear_NSA	101.41	100.62	99.85	99.09	98.33	97.58

Source:  AEMO/ACIL Tasman

Wind costs with the proposed $65 per MWh government penalty (an after tax penalty of $93) are below fossil fuel generated electricity costs and assure the use of wind to the degree stipulated.  This would remain true even though wind generators' inflexibilities mean they earn less on the spot market than controllable generators.

Unlike nuclear, where future efficiency improvements are likely, wind is now a mature technology, which has shown asymptotically lower costs which are now close to the their maximum.

A popular model, the Vestas V80, uses a rotor 80 meters in diameter and has a generator rating at 1.8 MW.  Set at a wind design point rating of 7 metres per second, on a site that averages this wind speed, it delivers a capacity factor of just under 30 per cent.  This approaches the practical maximum achievable which several authorities suggest is around 35 per cent and indicates that there are few technological improvements to be had with the design of wind turbines that will lower their costs. ⁽¹⁰⁾

THE COSTS OF THE 20 PER CENT RENEWABLE PROPOSAL

Using wind as the least cost renewable option, additional costs fall within three components:

• The premium required for wind over the least cost energy supply of $46 per MWh ($93 less $47 for coal)
• The costs of the additional back-up for wind supply.  Wind's reliability according to AEMO is only 7 per cent.  In order to ensure the system is maintained, fast start (gas) back-up plant is required.  The most market-oriented means of establishing the cost of this is to use the $300 base cap traded through the over-the-counter market, which gives traders the certainty that they can avoid losses in the event of their supply not performing.  The price of the $300 cap varies between $11 and $15 per MWh, with the averageat $12. ⁽¹¹⁾
• Additional costs stemming from the increased transmission likely to be required as a result of the lower average flow of electricity from wind sites.  In marginal cost terms this is likely to be low since the wind farms locate in areas where transmission is amply available.  Where new lines need to be built or existing lines reinforced, the costs do become significant.  However for the purpose of this analysis these costs are not included.

The premium costs per MWh for wind are therefore estimated at $58 ($58,000 per GWh).  In order to reach 45,000 GWh, this premium would need to be applied to the non-commercial (mainly hydro) part of renewable energy supply stipulated in the Bill.  The commercial component is about 14,000 GWh, hence the cost would apply to 31,000 GWh and amount to $1,800 million per annum in 2020, ⁽¹²⁾ with this rising in subsequent years in line with the 20 per cent target.  The cost of the renewable option is not greatly dissimilar from substituting nuclear for coal of $1,580 million, though nuclear might show considerable cost reductions, perhaps amounting to 40 per cent. ⁽¹³⁾

In addition to these costs there would be consequential magnification effects.

Offsetting this, in the context of an ETS, the expanded MRET would entail some reductions in the cost of the ETS (100 per cent in the unlikely event that the renewables were the lowest cost option).  The PC reported that the cost of specifying renewables was double that of a more market-oriented approach, findings the PC noted were similar to those by consultancy projects conducted by Access Economics and CRA.

CONCLUDING COMMENTS

There can be few other cases in the history of modern economies where governments have taken action deliberately to increase the costs of production in their economies on the scale being contemplated with the Renewable Energy proposal.  In the past, those societies that have rejected lower costs have been condemned to economic decline and in many cases a loss of sovereignty.  Historically, such was the fate of the Ottoman Empire and a looming repetition of this in the case of China was observed by key nineteenth century Japanese politicians.  This led to the Meiji Restoration with the recognition that Japan needed to aggressively accept the least costly production methods of the western world or become a backward, foreign controlled state.

The issue of renewable certificates under the Carbon Pollution Reduction Scheme Bill (2009) introduces a variable tax on generators, with the rate dependent on their emissions per unit of energy supplied.  An indexed emission cost of $40 per unit -- $57 in after tax terms -- (following $10 per unit in the first year) would increase the costs for a brown coal generator by 150 per cent and more than double those of a black coal generator.

Specifying a particular form of technology rather than leaving the market to select the lowest cost option will invariably result in additional costs.  The proposed 20 per cent renewable target for electricity would impose a considerable burden on the Australian consumer and on the nation's industry.  As a stand-alone policy, a 20 per cent renewable target means substituting existing sources of electricity by others that are far more expensive.  Windmills are the least cost means of supplying renewable energy in the absence of the availability of large scale hydro.  These entail a premium that is roughly double the cost of commercially available coal based electricity.

In those respects, a 20 per cent renewable requirement would mean the equivalent of a tax on non-renewable electricity supplies to raise the sums necessary to allow wind to be viable.  That amounts to a tax of $1,800 million on the consumer with the sums raised hypothecated back to the builders and owners of windmills.  In terms of the impost on energy consumers, as renewables are over twice the costs of conventional sources and apply to a fifth of total supply, the tax is equivalent to something in excess of 10 per cent of the total generation costs.

The 20 per cent renewable proposal in the Renewable Energy Bills involves intentionally increasing consumers' expenditures and deliberately reducing the competitiveness of productive activities by raising their input costs.  In doing so, because the ETS proposal is to control total Australian emissions, the 20 per cent proposal will contribute nothing to the goal of emission reductions that it ostensibly pursues.  It would simply re-arrange the aggregate nature of emission reductions, while imposing additional costs.

---

ENDNOTES

1.  What Role for Policies to Supplement an Emissions Trading Scheme?  May 2008

2.  The Skeptical Environmentalist, by Bjørn Lomborg, Cambridge University Press, p. 134.

3.  http://www.news.com.au/heraldsun/story/0,21985,22287835-662,00.html

4.  http://www.abc.net.au/news/newsitems/200608/s1723948.htm

5.  Denmark has a considerable amount of wind installed but is fully electrically integrated with the rest of Europe

6.  http://online.wsj.com/article/SB122937766062908297.html

7.  http://www.esipc.sa.gov.au/webdata/resources/files/Public_Forum_presentation_Final.pdf

8.  See http://www.aemo.com.au/planning/419-0032.pdf

9.  The stated costs for this are highly speculative and at the present juncture optimistic.

10.  Under Betz Law, a wind turbine can never convert more than 59 per cent of the wind flowing past it into energy.  This is further reduced by mechanical losses and the fact that the turbine only reaches its 59 per cent maximum at a specific "design point".  The Betz limit is not directly factored into the turbine rating.

11.  Personal communications with electricity traders, though a price can be estimated from the d-cypha site.

12.  This cost may be reduced by renewables featuring less uncompetitive sources like bagasse (which have limited availability) and, if the practice is allowed to continue, by Snowy and Hydro Tasmania manipulating the scheme by oscillating their production from year to year so that every second year they produce above baseline.  It would be increased to the degree that solar panels (with a cost over fourfold that of wind) fulfil part of the 20 per cent quota.

13.  The Economic Future of Nuclear Power, a study conducted at the University of Chicago, August 2004

A Real Education Revolution: Options for voucher funding reform

EXECUTIVE SUMMARY

The federal government's "education revolution" perpetuates the waste, inefficiency and perverse incentives that come with funding micro-management.

• The current school stimulus package is a prime example of this, with a range of ill-suited, cost-padded infrastructure works being laid out across the country.
• If Australia wants a real education revolution, it should ensure that government funding follows the student and not the preferences of education ministers or bureaucrats.
• One way to unlock the funding chains would be to introduce a school voucher, empowering parents to make educational choices and tackling systemic educational disadvantage.
• This paper explores a wide variety of voucher options.  It is estimated that vouchers could cost federal and state governments anywhere from an extra $700 million to $10.3 billion each year.
• The fiscal cost of seven national voucher schemes is outlined below.

Voucher scheme	Estimated additional fiscal cost ($ billions)
Universal voucher ($12,000 for each student)	6.9
Differential voucher (students with disabilities) ($12,000 for students without disabilities;  $15,000/$25,000/$40,000 for students with disabilities depending on disability severity)	8.0
Differential voucher (indigenous students) ($12,000 for non-indigenous students;  $15,000 for indigenous students plus $7,000 relocation allowance for rural/remote area indigenous student)	7.2
Differential voucher (low-income students) ($12,000 for students from medium to high income families earning more than $1,000 per week;  $15,000 for students from low income families earning less than $1,000 per week)	10.3
Targeted voucher (students with disabilities) (current funding for students without disabilities; $15,000/$25,000/$40,000 for students with disabilities depending on disability severity)	1.5
Targeted voucher (indigenous students) (current funding for non-indigenous students;  $15,000 for indigenous students plus $7,000 relocation allowance for rural/remote area indigenous student)	0.7
Targeted voucher (low-income students) (current funding for students from medium to high income families earning more than $1,000 per week;  $15,000 for students from low income families earning less than $1,000 per week)	5.2

• Far from being radical, vouchers are a mainstream policy pursued in 30 countries around the world from the United States right through to developing countries such as Columbia.
• There is sufficient empirical evidence that vouchers encourage more choices in schooling, improve academic outcomes and are highly popular with parents.
• One of the greatest advantages of vouchers is that they allow children from disadvantaged backgrounds to access high-performance schools.
• Many vouchers around the world are specifically earmarked for students from low-income families or minority backgrounds, or students with disabilities, that deserve the best education available.

INTRODUCTION

Since November 2007, the federal government has proceeded to implement its much-touted "education revolution" platform.

Critical policies in this area have included laptop access for all students from years 9 to 12, trade training centres in schools, more investment in teacher quality and a national curriculum.  In the 2008-09 budget, the government announced at least $6.5 billion in new measures. ⁽¹⁾

Earlier this year, the government also announced a three-year $14.7 billion "Building the Education Revolution" stimulus package.  Under the program, any of Australia's 9,540 schools are eligible for funds to build infrastructure on campuses.

The reality is that the government's policies largely amount to a continuation of funding methods firmly entrenched at both the commonwealth and state levels.  The so-called "revolution" in education amounts to little more than maintaining, and in some cases consolidating, the already heavy degree of centralised control over education.

This control gives the appearance that the government is "doing something" in education, yet at the significantly increased risk of waste, inefficiency and perverse incentives often associated with micro-managed government programs.  The current schools stimulus package is a prime example of this, with a range of ill-suited, cost-padded infrastructure works being laid out across the country. ⁽²⁾

The criticisms of the schools stimulus have come thick and fast, but they could be readily extended to other centralised school funding models used in Australia.  Experts have variously panned the initiative as a "fake education revolution" ⁽³⁾ and a "squandered opportunity," ⁽⁴⁾ with the spending having more to do with "political opportunism and spin than raising standards, improving outcomes and delivering a more effective and responsive education system." ⁽⁵⁾

Even principals and school councils, who would normally accept government funding, come what may, have criticised various elements of the government's stimulus spending program. ⁽⁶⁾

To engender a real revolution in the education system, there is a need to break down the walls of centralised school funding.  Parents need to be empowered in their choice of schools, and schools and teachers need incentives to become more responsive to the diverse needs of their students.

One of the most enduring suggestions to break down government control in education is to ensure that its funding follows the student, and not the education ministers or bureaucrats.  This method of funding is commonly known as portable school funding entitlements, or a "voucher" system.

In essence, a voucher system provides each parent with government funds to spend at the school of their choice.  By separating government financing of education from the operation of schools, vouchers can help break down monopoly control over school services delivery and promote competition between government and non-government schools.

As has been increasingly understood by both sides of the political spectrum, vouchers also represent a powerful tool to tackle educational disadvantage.  This is because parents of children with special education needs, or from low-income families, are financially empowered to take their children out of failing schools and into high-quality educational institutions.

This paper examines key aspects of potential voucher reforms for the Australian school system.  The next section will discuss some of the basic mechanics of voucher funding schemes.

Some potential voucher options, and their budgetary cost to governments, will be canvassed.  Other issues will then be explored, including the impact of federalism and distributional issues.  An examination of the international literature on the effectiveness of school vouchers will also be undertaken.

As will be noted below, for less than the total cost of the government's education revolution spending to date, Australian governments can embark on real funding reform -- via a voucher system -- that unlocks constraints on choice and meaningfully tackles educational disadvantage in our communities.

WHAT ARE VOUCHER FUNDING SCHEMES?

The basic attributes of a school voucher funding model were explained by Nobel Prize winning economist Milton Friedman in a classic 1955 article on the role of government in education:

"Government, preferably local governmental units, would give each child, through his parents, a specified sum to be used solely in paying for his general education;  the parents would be free to spend this sum at a school of their own choice, provided it met certain minimum standards laid down by the appropriate governmental unit.  Such schools would be conducted under a variety of auspices:  by private enterprises operated for profit, non profit institutions established by private endowment, religious bodies, and some even by governmental units." ⁽⁷⁾

In practice, different voucher mechanisms are used in an effort to promote parental schooling choice and break educational disadvantage:

• Universal voucher -- all parents receive a voucher worth the same amount of money towards the education of their child
• Differential voucher -- all parents of school children receive a voucher, for example adjusted either for the income level of the family or the costs of educating children with special educational needs (e.g. indigenous students, students with disabilities, students with a language background other than English, or geographically isolated students)
• Targeted voucher -- only certain households or children receive a voucher.  For example, vouchers could be confined to low income households, poorly performing students or students with disabilities.

These vouchers could either be provided directly to parents to spend on their child's school education, as Friedman originally suggested, or the voucher entitlement could be sent directly to the school of a parent's choice upon enrolment of the student.

More than 30 countries have adopted voucher-type schemes to fund the school education of young people.  This reality demonstrates that the voucher system of funding is feasible to implement and has become mainstream education policy.

HOW MUCH WOULD DIFFERENT
NATIONAL VOUCHER SCHEMES COST?

An important consideration for any proposed policy change is its expected impact on the government budget.

This section provides costings for a variety of hypothetical voucher schemes funded in a collaborative fashion by the commonwealth and state governments.  The information presented is based on 2006-07 school demographic data, unless otherwise specified, and relates to the recurrent costs of educating children in Australian schools.

For the purpose of this section, an assumption is also made that the voucher schemes do not induce a shift in enrolments by school sector.  As will be noted later in this paper, however, the experience of school vouchers in other countries suggests that they do encourage significant movements in school enrolments.

UNIVERSAL VOUCHERS

One voucher option would be to fund each Australian school student on an equal per capita basis, ending any sense of funding discrimination between students.

Table 1 provides indicative costings for a scheme whereby the commonwealth and states provide a $12,000 universal voucher for each student.

Table 1:  Indicative costs of universal voucher scheme, $ billions

	Govt schools	Non-govt schools	Total
Current funding (a)	26.9	7.3	34.2
Universal voucher ($12k) (b)	27.3	13.8	41.1
Change (b) - (a)	0.4	6.5	6.9

Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.  Sub-totals may not sum to totals due to rounding.

Source:  SCRGSP, 2009, Report on Government Services 2009,
Productivity Commission;  Author's calculations.

It is estimated that this voucher package would cost governments an additional $6.9 billion per annum.

DIFFERENTIAL VOUCHERS

Another strand of school vouchers are differentiated schemes that segment and vary the amounts of funding parceled out to groups of students.  In this paper, it is assumed that all mainstream students would receive at least $12,000 per annum and, in different cases, students with disabilities, indigenous students and students from low-income families would receive extra funds.

Students with disabilities

Governments currently provide additional resources to students with disabilities, with students assessed as having more severe disabilities generally receiving more funds than their less disabled counterparts.

Even so, there are significant variations between the states in terms of how much funding is directed to each disabled student. ⁽⁸⁾  It has also been observed that there exists a large government funding shortfall for students with disabilities in non-government schools. ⁽⁹⁾

One way to ensure consistency in how students with disabilities are funded is to create a national differential voucher where these students receive additional fund loadings on a standardised basis.

Table 2 provides indicative costings for such a voucher.  In this scenario, $12,000 is given to each student without a disability with extra funding allocated to students with disabilities according to three hypothetical categories of disability severity (level 1 $15,000, level 2 $25,000, level 3 $40,000). ⁽¹⁰⁾

Table 2:  Indicative costs of differential students with disabilities voucher scheme, $ billions

	Govt schools	Non-govt schools	Total
Current funding (a)	26.9	7.3	34.2
Differential voucher ($12k non-disabled)   Differential voucher ($15k level 1 SWD)   Differential voucher ($25k level 2 SWD)   Differential voucher ($40k level 3 SWD) Total differential voucher (b)	25.8 0.9 1.4 0.1 28.2	13.4 0.2 0.4 0.0 14.0	39.2 1.2 1.8 0.1 42.3
Change (b) - (a)	1.3	6.7	8.0

Students with disabilities = SWD.  Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.  Sub-totals may not sum to totals due to rounding.

Source:  SCRGSP, 2009, Report on Government Services 2009,
Productivity Commission;  Author's calculations.

It is estimated that this voucher package would cost governments an additional $8 billion per annum.

Indigenous students

There has been widespread public concern about the persistent (average) academic achievement gap between indigenous and non-indigenous school students across Australia.

According to a compendium to the latest Productivity Commission Report on Government Services, the proportion of indigenous students that did not achieve national literacy and numeracy testing benchmarks was substantially higher than the proportion of all students. ⁽¹¹⁾

Although this achievement discrepancy is not a standard phenomenon across the country, ⁽¹²⁾ a number of proposals has been canvassed in recent years to help close the average school performance gap.  These include scholarships for indigenous students to attend high-achievement non-government schools, ⁽¹³⁾ greater flexibility for schools servicing primarily indigenous students, ⁽¹⁴⁾ improved curriculum standards and better school attendance rates.

A policy proposal that has been largely overlooked so far is a voucher that enables indigenous students to move from poorly performing to high-performance schools.  This would represent a demand-side reform helping to lift school participation and academic performance, complementing the largely supply-side solutions described above.

Table 3 provides indicative costings for a voucher whereby $12,000 is given to each non-indigenous student with extra funding allocated to indigenous students ($15,000).  Given the lack of diversified school supply in remote and very remote areas, the voucher scheme will be complemented by a one-off allowance ($7,000) for each eligible indigenous student to relocate to a metropolitan school if his or her family so chooses.

Table 3:  Indicative costs of differential indigenous students voucher scheme, $ billions

	Govt schools	Non-govt schools	Total
Current funding (a)	26.9	7.3	34.2
Differential voucher ($12k non-indigenous)   Differential voucher ($15k indigenous) Total differential voucher (b)	25.7 1.9 27.6	13.5 0.3 13.8	39.2 2.2 41.5
Change (b) - (a)	0.7	6.5	7.2

Voucher amount includes $7,000 relocation allowance for indigenous students residing in remote and very remote locations, calculated by weighting total number of indigenous students by the proportion of total students attending schools by remote/very remote geo-locations.  Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.  Sub-totals may not sum to totals due to rounding.

Source:  SCRGSP, 2009, Report on Government Services 2009,
Productivity Commission;  Author's calculations.

It is estimated that this voucher package would cost governments an additional $7.2 billion per annum.

Students from low-income families

While Australian schools are recognised for providing a world-class education compared to many other OECD economies, concerns have been raised about student underperformance in our education system.

According to a federal government study, Australia has too long a "tail" of underperformance linked to disadvantage. ⁽¹⁵⁾  This outcome is concentrated amongst students from low socioeconomic status (SES) communities, with the difference between students from the lowest SES quartile to those in the highest equating to more than two years of schooling in both reading literacy and mathematics.

To address this, governments have embarked on a range of measures including an emphasis on improving teacher quality, strategies for improved student engagement and retention and improving school transparency and accountability.

An appropriately designed differential voucher scheme should also be on the policy agenda, because it can give students from low SES backgrounds a better chance in life by accessing quality education in high-performance schools.

Taking the income levels of students' families as a proxy for SES, Table 4 provides indicative costings for a voucher whereby $12,000 is given to each student from a middle- to upper-income family and extra funding ($15,000) allocated to students from low-income families.

Table 4:  Indicative costs of differential lowincome families voucher scheme, $ billions

	Govt schools	Non-govt schools	Total
Current funding (a)	26.9	7.3	34.2
Differential voucher ($12k middle to upper income)   Differential voucher ($15k low income) Total differential voucher (b)	16.3 13.6 29.9	10.5 4.0 14.6	26.9 17.7 44.5
Change (b) - (a)	3.0	7.3	10.3

Voucher allocated to students from families earning less than $1,000 per week.  Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.  Sub-totals may not sum to totals due to rounding.

Source:  Barbara Preston, 2007, "The social make-up of schools";  SCRGSP, 2009,
Report on Government Services 2009, Productivity Commission;  Author's calculations.

It is estimated that this voucher package would cost governments an additional $10.3 billion per annum.

TARGETED VOUCHERS

The final type of voucher reform considered in this paper are those exclusively targeted at specific elements of the school population.  The intention of such vouchers is to provide students with additional education needs a financial opportunity to access the best schooling options available.  Remaining students are assumed to remain on the funding programs that currently exist at the commonwealth and state levels.

Students with disabilities

Table 5 provides indicative costings for a voucher allocated to students with disabilities, that vary in accordance to three hypothetical categories of disability severity (level 1 $15,000, level 2 $25,000, level 3 $40,000).

Table 5:  Indicative costs of targeted students with disabilities voucher scheme, $ billions

	Govt schools	Non-govt schools	Total
Current funding (a)	26.9	7.3	34.2
Funding (non-disabled)   Targeted voucher ($15k level 1 SWD)   Targeted voucher ($25k level 2 SWD)   Targeted voucher ($40k level 3 SWD) Total targeted voucher (b)	25.5 0.9 1.4 0.1 27.9	7.2 0.2 0.3 0.0 7.8	32.8 1.1 1.7 0.1 35.7
Change (b) - (a)	1.0	0.5	1.5

Students with disabilities = SWD.  Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.  Sub-totals may not sum to totals due to rounding.

Source:  SCRGSP, 2009, Report on Government Services 2009,
Productivity Commission;  Author's calculations.

It is estimated that this voucher package would cost governments an additional $1.5 billion per annum.

Indigenous students

Table 6 provides indicative costings for a $15,000 voucher allocated to indigenous students.

Table 6:  Indicative costings of targeted indigenous students voucher scheme, $ billions

	Govt schools	Non-govt schools	Total
Current funding (a)	26.9	7.3	34.2
Funding (non-indigenous)   Targeted voucher ($15k indigenous) Total targeted voucher (b)	25.4 1.9 27.4	7.3 0.3 7.6	32.7 2.2 34.9
Change (b) - (a)	0.4	0.3	0.7

Voucher amount includes $7,000 relocation allowance for indigenous students residing in remote and very remote locations, calculated by weighting total number of indigenous students by the proportion of total students attending schools by remote/very remote geo-locations.  Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.  Sub-totals may not sum to totals due to rounding.

Source:  SCRGSP, 2009, Report on Government Services 2009,
Productivity Commission;  Author's calculations.

It is estimated that this voucher package would cost governments an additional $0.7 billion per annum.

Students from low-income families

Table 7 provides indicative costings for a $15,000 voucher allocated to students from low-income families.

Table 7:  Indicative costings of targeted low income families voucher scheme, $ billions

	Govt schools	Non-govt schools	Total
Current funding (a)	26.9	7.3	34.2
Funding (middle to upper income)   Targeted voucher (lower income) Total targeted voucher (b)	16.2 13.6 29.8	5.7 4.0 9.7	21.8 17.7 39.5
Change (b) - (a)	2.8	2.4	5.2

Voucher allocated to students from families earning less than $1,000 per week.  Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.  Sub-totals may not sum to totals due to rounding.

Source:  Barbara Preston, 2007, "The social make-up of schools";  SCRGSP, 2009,
Report on Government Services 2009, Productivity Commission;  Author's calculations.

It is estimated that this voucher package would cost governments an additional $5.2 billion per annum.

ASSESSING THE OPTIONS

It is clear that an Australian school voucher scheme will require additional taxpayer resources to implement successfully.  According to the case studies outlined above, a voucher could cost anywhere between an additional $0.7 billion and $10.3 billion per annum for commonwealth and state governments.

A broader perspective of voucher fiscal costs can be gleaned from a sensitivity analysis where the voucher funding rate, for each of the cases studied, is progressively increased (Figure 1).

Figure 1:  Voucher funding rates compared to base case

Base case is defined as each of the voucher schemes examined in this paper.  For differential vouchers, funding rates for non-mainstream students are adjusted only.  Increases in funding rates for students with disabilities vouchers are averaged out across the three levels of disability severity.  Indigenous voucher schemes include $7,000 relocation allowance.  Figures are based on publicly available data, and should be regarded as indicative estimates of cost magnitudes only.  Excludes capital funding.

Source:  Barbara Preston, 2007, "The social make-up of schools";  SCRGSP, 2009,
Report on Government Services 2009, Productivity Commission;  Author's calculations.

While a $12,000 universal voucher is less expensive than the differential voucher options -- that provide funding of $12,000 for each school student as a minimum amount -- the universal funding option becomes prohibitively expensive as its base rate increases.  Indeed, across most of the funding rate scales the universal voucher is the most expensive reform option.

By far the most cost-effective voucher options would be the targeted schemes for students with disabilities or indigenous students, reflecting the relatively small numbers of these students in the overall student population.

This analysis shows that a national voucher scheme would impose additional costs upon governments that already spend a collective $39.3 billion on primary and secondary education. ⁽¹⁶⁾  Whereas school vouchers might be rejected on a "smaller government" policy criterion alone there are other considerations to bear in mind, not least the expected beneficial outcomes arising from vouchers. ⁽¹⁷⁾

ARE THERE ANY OTHER ISSUES FOR
SCHOOL VOUCHERS IN AUSTRALIA?

Some other voucher-related policy issues to consider include the status of commonwealth-state funding relationships, distributional issues, and school fees and parental contributions in a voucherised funding system.

HOW WILL THE COMMONWEALTH-STATE FINANCIAL DIMENSION AFFECT VOUCHERS?

The state and territory governments are constitutionally responsible for providing education to children of school age.  They own and manage government schools, and provide substantial funding to this sector.  They also regulate, and provide funds to, non-government schools.

In practice, the state government school funding system is heavily skewed towards their own government schools.  In 2005-06, the states injected about $23 billion in taxpayers' funding into government schools (Figure 2).

By contrast, states provided only $2 billion to non-government schools.  In other words, the states' funding contribution to government schools was almost 13 times that provided to Catholic and independent schools combined.

Figure 2:  Recurrent funding for school education, 2005-06

Source:  Independent Schools Council of Australia, "Snapshot 2008".

Although unilateral commonwealth action in the past has been decisive in promoting choice and quality -- for example, the SES model encouraged the growth of low-fee independent schools -- the prevailing distribution of funding by the states distorts parental choices in favour of "free" (direct to individuals) government schools.  In effect, a "choice penalty" is applied to those parents who prefer to send their children to Catholic or independent schools.

If the full benefits of vouchers are to be realised, it is important that a voucher governs both the commonwealth and state government education funding streams.  Indeed, this is the very reason why this paper has examined the financial implications of national schemes.

Clearly, a negotiation process will need to be conducted by governments if a national voucher is to become a reality.  For example, one way to parse out a differential voucher model across the levels of government would be for:

• the states to fund a basic voucher amount (per capita entitlement) for all mainstream students, in both government and non-government schools;
• the commonwealth to fund additional voucher amounts for students with identified educational needs, either students with disabilities, indigenous students or students from low income families. ⁽¹⁸⁾

Apart from ameliorating the additional funding burden faced by any one level of government, there could be other advantages in this commonwealth-state voucher split.  A basic state voucher for mainstream students would mean that they could no longer discriminate between children in different schooling systems.  A voucher mechanism would also modernise the confusing and, at times, archaic funding formulas currently used by the states.

For its part, the commonwealth can more effectively target its funding toward needs students.  This would be consistent with its implied constitutional role of addressing equity concerns more generally.

Other suitable arrangements could be agreed by the commonwealth and states for universal or targeted vouchers respectively.

WHAT ABOUT DISTRIBUTIONAL ISSUES, OR THE "MORE MONEY FOR KING'S" ARGUMENT?

Despite the defeat of the Mark Latham-led federal opposition in 2004, and with it its notorious schools "hit list" policy, ⁽¹⁹⁾ political controversy remains about the distribution of school funding.

There remain occasional public bouts of antipathy within some segments of the community, including public sector teacher unions and other government school education lobbies, about high-fee independent schools receiving government funding for their enrolled students.

It is quite conceivable that any funding change which could accelerate the enrolment drift from government schools may renew debates about funding to the non-government school sector.  The aforementioned vested interests may be particularly vocal in their opposition to change, potentially weakening political support for voucher reform.

The political ramifications of such criticisms could be ameliorated if the voucher was provided directly to parents.  This arrangement would underline the fact that it is the exercise of choice by parents on behalf of their children, and not by governments or any other interest, that is paramount under a voucher system.

The amount of government revenues received by schools would become a by-product of parental choice.  Therefore, any adverse funding shift felt by government schools, or any other school for that matter, would be directly related to their inability to respond to the demands of ordinary Australians concerning the quality of education.

SHOULD SCHOOL FEES AND PARENTAL CONTRIBUTIONS STILL PLAY A ROLE?

In a number of countries, governments have stipulated conditions to ensure schools are not permitted to charge tuition fees on voucher-carrying students.  Similarly, parents are in some instances prevented from topping up government voucher funding with their own contributions.

Such conditions might save parents some of their disposable income (although this could be offset by higher taxes used to fund vouchers).  However, there are good arguments for fees and voluntary contributions to remain in a voucher funding environment.

A transparent fee structure, as currently applied by non-government schools, effectively serves as an important signal to consumers of the quality of educational services provided.  For example, a higher tuition fee charged by a school is generally a marker of better educational quality that many parents are willing to pay.

Any government ban on charging fees, as part of a voucher system, would effectively remove price signals from school education hampering informed school choices by parents.

Many Australian parents, who send their children to non-government schools and pay taxes to pay for other children to attend government schools, willingly make financial sacrifices reflecting their commitment to their child's future.  This behaviour puts in practice the old proverb that "you do not value anything you do not pay for".

It can be reasonably argued that the retention of a fee and contributions structure in Australian schooling will ensure that parents maintain a keen level of interest and engagement in student academic performance and school efficiency.

DO VOUCHERS WORK IN SCHOOL EDUCATION?

Milton Friedman's ingenious idea of 1955 still evokes often passionate debate, even in those countries that have embraced voucher-style reforms.

Not surprisingly educational researchers have played a large role in this, as they attempt to tease out some of the effects of voucher funding.  This debate is likely to continue for some time, nonetheless there is sufficient evidence to back claims that school vouchers are beneficial in a number of ways.

VOUCHERS PROMOTE CHOICE IN SCHOOLING

Vouchers have the potential to dramatically expand the suite of attainable schooling options for children.  By promoting school choice, vouchers give parents a greater sense of empowerment over the education system and makes schools more responsive to student learning needs.

In Sweden, the introduction of a universal voucher in 1992 has led to the number of private schools expanding from 70 to about 800 today. ⁽²⁰⁾  The share of students that attend private schools quadrupled in ten years. ⁽²¹⁾  As noted by UK social policy expert David Green, "the whole point of the Swedish voucher scheme was to stimulate the opening of new schools so that parents in poor localities had the power to choose." ⁽²²⁾

There are instances where the relaxation of voucher eligibility constraints has fuelled a spike in demand for this type of school funding.  The Milwaukee low-income voucher was subject to a strict cap of one per cent of the city's public school population prior to 1998, which was lifted in 1998 and again in 2006.

The lifting of the Milwaukee voucher cap helped to raise the numbers of students benefitting from the program (Figure 3), in turn increasing enrolments in private schools.

Figure 3:  Milwaukee Parental Choice Program recipients, 1990-91 to 2008-09

Data based on headcounts as at September of each year.

Source:  Wisconsin Department of Public Instruction (accessed 19 June 2009).

VOUCHERS IMPROVE EDUCATIONAL OUTCOMES

Voucher programs around the world have been shown to improve the academic performance of students.  This is because vouchers give parents the freedom to choose schools that work best for their children, and creates incentives for schools to improve their educational performance.

In a non-randomised study of the impact of the Cleveland voucher scheme, Metcalf found that recipients had significantly higher test scores than public school students in reading, writing and science. ⁽²³⁾

Peterson, Howell and Greene found that voucher students "had gains of 7.5 national percentile points in reading and 15.6 NPR in math.  These gains were achieved even through the students ... were among the most disadvantaged students in Cleveland." ⁽²⁴⁾

Similar test score gains by voucher recipients were found in Milwaukee. ⁽²⁵⁾  In addition, a study by Warren found that students participating in the voucher program enjoyed better school completion rates than those in public schools. ⁽²⁶⁾  It is estimated that if the graduation rates in public schools equaled that of voucher students from 2003 to 2007, the number of public school students graduating during that period would have been about 20 per cent higher.

An important finding in the literature is that vouchers also tend to improve the academic performance of students that remain in public schools.  In the words of American education economist, Caroline Hoxby, vouchers are a policy akin to a tide that raises all boats. ⁽²⁷⁾

In 2005, the US state of Ohio introduced the EdChoice Scholarship Program that offers vouchers to students in underperforming public schools across the state (excluding Cleveland, which has its own voucher).  A recent study by the Friedman Foundation for Educational Choice found that this voucher has improved the performance of failing schools, with reading and mathematics scores all improving due to voucher competition. ⁽²⁸⁾

Before being struck down by a court, Florida provided a voucher to students in failing schools. ⁽²⁹⁾  A number of studies have backed the proposition that this performance-based scheme improved outcomes in public schools.  For example, Greene found that student achievement improved at a faster rate in schools that received the lowest F rating. ⁽³⁰⁾

VOUCHERS TACKLE EDUCATIONAL DISADVANTAGE

Apart from facilitating greater choice in schooling, there has been growing recognition that vouchers present a powerful tool to redress various dimensions of educational disadvantage affecting young people.

Many of the American public voucher programs are specifically designed to target disadvantaged groups in the community.  For example, in 2004 the US federal education department established a voucher scheme for low-income students in the Washington DC area.  Over 94 per cent of users during its first year of operation were African-American students. ⁽³¹⁾

A study of the Milwaukee voucher by Witte found that it serviced disproportionate numbers of students from low-income, African-American, Latin American or single parent families:  "the MPCP ... [Milwaukee Parental Choice Program] ... was established and the statute written explicitly to provide an opportunity for relatively poor families to attend private schools.  The program clearly accomplished that goal." ⁽³²⁾

It is notable that vouchers have been implemented to help the disadvantaged in an increasing number of developing countries around the globe.  For example, a targeted voucher has been used in Colombia since 1992 to defray the cost of providing a school education for poor families.  Within two years of its introduction nearly 91,000 low-income students were serviced by the program. ⁽³³⁾

As outlined above, vouchers have been associated with an improvement in student academic performance.  Such results have not only been detected for students accessing vouchers, but also for disadvantaged students that do not move to another school.  This is because all schools must improve their educational services for enrolled students in a voucher funding environment, if they are to avert the threat of exit.

VOUCHERS COME WITH HIGH APPROVAL RATINGS BY PARENTS

It is clear that wherever vouchers have been implemented they have been met with an overwhelmingly positive response from parents.  This is not only illustrated by the increasing numbers of students receiving vouchers worldwide, but by surveys of parental satisfaction.

An evaluation of the Milwaukee voucher program indicated that 53 per cent of parents with children enrolled in voucher schools gave the school an "A" grade, compared to 34 per cent of parents in public schools. ⁽³⁴⁾

For the Washington DC voucher about 66 per cent of parents who were offered a voucher gave their child's school an "A" or "B" grading, compared with 63 per cent of parents who were not offered a voucher. ⁽³⁵⁾

Closer to home, New Zealand implemented a small, pilot voucher scheme for children from low-income families in the late 1990s.  Surveys showed that 97 per cent of parents were either "satisfied" or "very satisfied" with their child's progress at the school of choice.

Teachers and principals were also reported to be enthusiastic about the benefits of the scheme for students.

CONCLUSION

Access to a quality school education is widely recognised by researchers and policymakers as a vital ingredient for full economic and social participation by young people as they transition into adulthood.

Australia provides a standard of education that is highly regarded around the world.  In other words, it makes a good fist of getting towards the world-class system that parents, teachers and school administrators strive to achieve.

Nonetheless, there is scope for improvement.  Our school funding systems are a hodge-podge of complexity that effectively discriminate between various groups of students.  One only has to appreciate, for example, the funding plight of students with disabilities in non-government schools to know that the way Australia funds its schools needs to change.

Despite many funding streams being badged with the word "targeted" they merely throw money at problems without addressing their root causes, such as the lack of sufficient incentive for schools to vigorously compete against each other in providing high-quality, tailored education solutions for children.

Current funding models conveniently keep the education bureaucrats at work, but at the cost of greater waste, inefficiency and perverse incentives as currently witnessed by the problems with the federal government's "education revolution".

It is for all of these reasons that Australia should seriously investigate the adoption of voucher funding for the school system, taking control out of the hands of distant education officials and back into the hands of parents.

Enabling funds to "follow the student" to whichever school they end up is a simple, yet powerful, idea that has been shown internationally to promote choice, improve student outcomes, tackle educational disadvantage and improve parental satisfaction in schooling.  Voucher experiments around the world have clearly shown that they work, for parents, principals, teachers and students alike.

So, without further ado, let the real education revolution of vouchers begin!

---

ENDNOTES

1.  Hon Julia Gillard MP, 2008, Education Revolution 2008-09, Budget Ministerial Statement (accessed 18 June 2009).

2.  Some of the key issues have been raised by Gavin Lower and Natasha Bita, 2009, "Millions handed to axed schools in stimulus debacle", The Australian, 12 June;  Natasha Bita, "Hall costs go through the roof", The Australian, 13 June;  Natasha Bita, 2009, "$2m handed over as closure looms", The Australian, 15 June;  Justine Ferrari, 2009, "Kevin Rudd's local school gets $3m for double-ups", The Australian, 16 June;  Gavin Lower, 2009, "Plans axed as pledged funds fall short", The Australian, 19 June;  and Paige Taylor and Justine Ferrari, 2009, "$250,000 hall for remote Yulga Jinna Remote Community School 'a waste' ", The Australian, 20 June.

3.  Ken Wiltshire, reported in Justine Ferrari, 2009, "Schools infrastructure fund under fire", The Australian, 13 June.

4.  Louise Watson, reported in Ibid.

5.  Kevin Donnelly, "School marks not fixed with bricks", The Australian, 15 June.

6.  Henry Grossek, 2009, "Schools' pride felled by bullying bureaucrats", The Australian, 18 June;  Justine Ferrari, 2009, "Randwick Public School P&C says it's out of time", The Australian, 19 June.

7.  Milton Friedman, 1955, "The Role of Government in Education", in Robert A. Solo, ed., Economics and the Public Interest, Rutgers University Press, New Brunswick.

8.  Fran Ferrier et al, 2007, Investigating the Feasibility of Portable Funding for Students with Disabilities, Monash University Faculty of Education.

9.  Jennifer Buckingham, 2009, "Private hurdle for kids", The Australian, 21 April.

10.  There is little information on the distribution of students with disabilities in different jurisdictions according to assessed severity.  This paper uses the (strong) assumption that a nationally observed distribution of students of disabilities, according to support need, is consistent with previously published information for NSW.  See Fran Ferrier et al, 2007, op cit.

11.  SCRGSP, 2009, Report on Government Services 2009, Indigenous Compendium, Productivity Commission.

12.  The development economist Helen Hughes has led research that shows indigenous students learning in mainstream metropolitan and major regional centre schools are generally performing markedly better than their counterparts in rural and remote areas.  See Helen Hughes and Mark Hughes, 2009, Revisiting Indigenous Education, CIS Policy Monograph No. 94.

13.  Noel Pearson, 2004, "No danger of another stolen generation", The Australian, 5 November.

14.  Julie Novak, 2006, School Autonomy:  A Key Reform for Improving Indigenous Education, CIS Issue Analysis No. 72.

15.  Commonwealth of Australia, 2008, Quality Education:  The Case for an Education Revolution in our Schools, Department of Education, Employment and Workplace Relations.

16.  ABS, Government Finance Statistics, Australia, 2007-08, cat. no. 5512.0.

17.  According to Buckingham, "these extra outlays would, at least in part, be offset by the efficiency of directing funding through families, and the downward pressure on schooling costs as a result of competition between schools." See Jennifer Buckingham, 2001, Families, Freedom and Education:  Why School Choice Makes Sense, CIS Policy Monograph No. 52, p. 32.

18.  Appropriate burden sharing may entail the commonwealth dedicating voucher base amounts for needs students in excess of those quantified in this paper.

19.  Mark Latham and Jenny Macklin, 2004, "Great Australian Schools", Australian Labor Party Policy Document for 2004 Federal Election Campaign.

20.  Michael Sandström, 2006, "School choice reforms in Sweden", in James Stanfield, ed., The Right to Choose? -- Yes, Prime Minister!, Adam Smith Institute, London.

21.  Johnny Munkhammar, 2007, "How choice has transformed education in Sweden", The Telegraph, 25 May.

22.  David Green, 2007, "Enhancing education vouchers", Prospect Magazine 140.

23.  Kim K. Metcalf, 1999, "Evaluation of the Cleveland Scholarship and Tutoring Program, 1996-1999", cited in Herbert J. Walberg, School Choice, Cato Institute, Washington DC.

24.  Paul E. Peterson, William G. Howell and Jay P. Greene, 1998, "An Evaluation of the Cleveland Voucher Program After Two Years", Harvard University Program on Education Policy and Governance, Working Paper.  This study was based on a small sample of two private schools, even though they represented 25 per cent of voucher students who transferred from public schools.

25.  Jay P. Greene, Paul E. Peterson and Jiangtao Du, 1998, "School Choice in Milwaukee:  A Randomized Experiment", in Paul E. Peterson and Bryan C. Hassel, eds., Learning from School Choice, Brookings Press, Washington DC;  Cecelia Elena Rouse, 1998, "Private School Vouchers and Student Achievement:  An Evaluation of the Milwaukee Parental Choice Program", Quarterly Journal of Economics 113 (May):  553-602;  Caroline M. Hoxby, 2006, School Choice:  The Three Essential Elements and Several Policy Options, New Zealand Education Forum, Wellington;  Rajashri Chakrabarti, 2007, "Can Increasing Private School Participation and Monetary Loss in a Voucher Program Affect Public School Performance?  Evidence from Milwaukee", Federal Reserve Bank of New York Staff Report.

26.  John R. Warren, 2008, "Graduation Rates for Choice and Public School Students in Milwaukee, 2003-2007", School Choice Wisconsin.

27.  Caroline Minter Hoxby, 2001, "Rising Tide", Education Next 1 (4):  69-74.

28.  Greg Forster, 2008, "Promising Start:  An Empirical Analysis of How EdChoice Vouchers Affect Ohio Public Schools", Friedman Foundation for Educational Choice.

29.  Under the former statewide "A+ program", the overall achievement levels of schools were graded from A to F.  Students in schools that received two F ratings in any four-year period would be offered a voucher to use at any other public or private school.

30.  Jay P. Greene, 2001, "An Evaluation of the Florida A-Plus Accountability and School Choice Program", Manhattan Institute.

31.  Patrick J. Wolf, 2008, "School Voucher Programs:  What the Research Says About Parental School Choice", Bringham Young University Law Review (July):  415-446.

32.  John F Witte, 2000, The Market Approach to Education:  Evidence from America's First Voucher Program, cited in Ibid.

33.  James Tooley, Pauline Dixon and James Stanfield, 2003, Delivering Better Education:  Market Solutions for Educational Improvement, Adam Smith Institute, London.

34.  John F. Witte et al., 2008, "MPCP Longitudinal Educational Growth Study Baseline Report", SCDP Milwaukee Evaluation Report No. 5, February.

35.  Patrick Wolf, Babette Gutmann, Michael Puma, Brian Kisida, Lou Rizzo and Nada Eissa, Evaluation of the DC Opportunity Scholarship Program:  Impacts After Two Years.