Few economists would accept this proposition as a generalisation. Tax increases can have significant incentive costs (although even here there is controversy^[4] ), with the so-called ‘deadweight’ (choice distorting, welfare-reducing) costs of higher taxes as high as 20 cents in the dollar. But the net economic cost of tax increases depends on:

With a careful choice of tax instruments and well designed spending programs, the economic benefits of public spending often outweigh the costs, especially in a low tax country like Australia. That is why it is hard to find a significant statistical correlation between size of government (levels of government spending and taxation) and a nation’s economic performance.

People who want governments to spend less can mount a general argument based on ‘government failure’ and the need for a ‘disciplined and restrained approach’ to public spending. Equally, those who want governments to spend more on infrastructure, education, health care or early childhood intervention can make a general argument for additional spending based on increased wellbeing and social capital gains. But, at this generalised level, the arguments are based on ideology — not economics. Each proposal requiring higher taxation needs to be assessed on its individual merit without any prior presumption for or against.

This claim is plainly ludicrous. Net public debt — the difference between the government’s stock of financial (mainly debt) liabilities and its financial assets — is not an appropriate measure of a government’s balance sheet strength. The focus should be on net public worth — all financial and non-financial assets minus all liabilities. If governments borrow money to invest in real ‘productive’ assets with a comparable life to the debt, this adds to net public debt but it does not detract at all from net worth and, depending on the investment, could even increase net worth in the long term.

The only prudential requirement on governments should be ensure that, over the medium term, they run a net operating surplus (an excess of current revenue over current expenses) and borrow only to invest in physical or human capital projects that have met the standard cost-benefit criteria and have the potential to increase the revenue base in the long term. This should ensure net government worth is stable or rising and that public debt levels are kept at a sustainable level in terms of capacity to service. This is the stance now adopted by state governments and by most other OECD governments.^[6] It is only the Commonwealth government that is out of line.

The irony is that Australia’s public debt levels are among the very lowest in the developed world (less than five percent of the OECD average, relative to GDP). All Australian governments have very strong balance sheets^[7] and our credit rating agencies are generally relaxed about some increase in government borrowing for investment purposes.^[8] This should give Australia more — not less — freedom to borrow than other OECD countries.

As a generalisation, this proposition is simply untrue. While there is little doubt that the private sector is generally better than the public sector in design, construction and operation of infrastructure, capturing these benefits does not require private ownership. Governments can and do out-source most operational matters to private companies and consultants.

The efficiency case for private ownership of infrastructure is based on a number of premises which may or may not be correct.

Firstly, it assumes that the equity risks of the infrastructure project are largely commercial in character. But the equity risks are often more regulatory and political in character and in such cases the private sector is likely to demand a very high risk premium.

Secondly, the government is not always able to effectively transfer to the private sector the ultimate risk of default. Whatever the formal contracts might say, if a privatised hospital, school, road or railway network fails to perform, the government is held responsible.

Thirdly, private ownership is able to deliver benefits to users only if there is sufficient contestability in financial and service markets. The market for infrastructure finance has now matured and the up-front transaction costs such as fees to financial intermediaries, while still high, are now more reasonable than they were. But it remains hard to avoid quasi-monopoly market power in many infrastructure service markets. In such circumstances, privatisation would require close regulation and monitoring to ensure prices are reasonable and there is adequate accountability and transparency. This could nullify many of the efficiency advantages of private participation.

A fourth assumption is that private ownership will lead to improved managerial incentives. But will it? Government agencies are often derided for their lack of modern management expertise but in recent years they have developed ways to auction out community service obligations to avoid opportunistic political interference while also giving managers clear goals and well-structured performance incentives. There are many successful state-owned enterprises across the world, one example being Singapore Airlines. Private ownership of infrastructure, especially in listed companies, comes with its own problems such as the overriding desire to satisfy the short term demands of financial markets.

Finally, assigning infrastructure ownership risks predominantly to the private sector can lead to a misallocation of capital resources. For example, it tends to create a bias in favour of infrastructure investments with good commercial potential and against social infrastructure with high social returns (an issue I return to later). As well, in the case of new roads, privatisation can distort patterns of usage (forcing motorists to take less time-saving alternatives).

A recent OECD study of country experiences finds there are only ‘limited’ efficiency gains from extensive involvement by the private sector in the ownership and provision of non-self-funding social infrastructure and in social protection.^[9]

In short, while private equity ownership of infrastructure will often be able to save taxpayers money and offer a service that is cheaper and more responsive to consumer preferences, governments should not start with a universal presumption that it is always superior to public ownership (which is what the embargo on net public borrowing implies). Each case needs to be assessed on its merit.

This view about the implications for interest rates and inflation is widely held, even by people who should know better, including the former Treasurer, Peter Costello.^[10] Yet it has very little validity. Here it is important to distinguish between interest rate pressures stemming from capacity constraints (‘real crowding out’) and those stemming from financial market reactions (‘financial crowding out’).

If Australia’s productive capacity is fully stretched, any new government infrastructure investment financed out of borrowings will, if other policies remain unchanged, add to inflationary and interest rate pressures — at least in the short term until the investment starts to pay off. However, against this, I would make the following observations:

What of ‘financial crowding out’ — the risk that global financial markets will require an extra interest rate ‘loading’ for country and sovereign risk? This can happen in two ways. The first is by impacting on global capital markets, although this can be ruled out as Australia is a price-taker in these markets. The second is through effects on inflationary expectations. This has been a problem in the past because credit rating agencies tended to get skittish when a major new program of infrastructure investments was financed by public sector borrowing (rather than by private interests) because of unease about the lack of commercial and market disciplines in decision-making and the fear it might create useless ‘white elephants’ and make the additional debt hard to service. However, with public debt now at historically very low levels, and assuming new infrastructure decisions are made on the basis of sound cost-benefit evaluations, rating agencies and financial markets can be expected to take a more sensible and relaxed view of both public sector and external deficits.^[11]

Relying on current revenue to pay for the full cost of new social investment is not a sensible or viable alternative to government borrowing.

Firstly, we cannot count on a continuation of the extraordinary revenue windfall from the mining boom. When that boom ends, a federal government will have limited capacity to fund the up-front capital costs of infrastructure out of current revenue because of its self-imposed tax ceiling.

Secondly, there would be an economic cost if governments chose to crowd out economically desirable recurrent outlays (such as on education) or if workers were forced into higher than desirable effective marginal tax rates (Argy 2007 p. 149).

Thirdly, and most fundamentally, it is unfair to ask present Australians to pay upfront for new capital spending that will yield returns over a period of several years or even decades. Revenue should be used to pay for annual accrual expenses only (interest, depreciation, operational and maintenance costs).

This is an important issue because one of the key concerns of critics of the present fiscal straitjacket is that it has led to neglect of Australia’s public infrastructure — and especially social infrastructure. To explore this issue further, I would ask two questions:

Unless one starts with the simplistic premise that ‘governments always stuff everything up’ (as some exponents of Public Choice Theory argue), the argument that fiscal surpluses boost productivity can be turned on its head.

As already discussed, the embargo on government borrowing tends to have efficiency costs because it overloads the tax burden in the early years and encourages governments to use private financing when it is not the optimal choice.

A more fundamental productivity cost of the present fiscal stance is that it artificially constrains the ability of governments to choose between social and economic infrastructure and between private goods and collective services, even when the latter offers relatively high marginal social returns. In particular, as noted earlier, it creates a bias in the allocation of capital against many types of social investment such as public schools, hospitals, urban roads and transport, child care development, training, lifelong learning institutions and community and preventative health care facilities.

Capital markets work very imperfectly. They have a tendency to ignore wider economic benefits for ‘third parties’ (benefits not directly captured by market transactions), such as effects on the environment, travel time, accidents, and the productivity of the unpaid household sector. And they tend to under-invest in merit goods such as health, education, job search, specific training and certain kinds of infrastructure (that crowd in private investment).

While government failure is also rife and needs to be allowed for, it is ideological bigotry to assume that governments are always incapable of correcting for market failures. A number of credible studies have found national economic returns (in terms of real incomes per head) of between $2 and $10 per dollar of government outlays on early childhood disadvantage and broader access to health, education, housing, public transport and improved urban freeways,^[17] with gains coming in the form of a better educated and skilled workforce and citizenry; greater geographic and occupational mobility of labour; less waste of potentially successful entrepreneurs; higher employment participation rates; diminished health costs; lower imprisonment rates; less spending on welfare and juvenile delinquency; savings in commuting time; lower accidents and reduced pollution.

Cross-country studies show that governments (notably the Nordic countries, Ireland, the Netherlands and Austria) who choose to give relatively high priority to social investment have been very successful in reconciling high levels of social redistribution with good — or even superior — economic outcomes.^[18]

There is another, more subtle, productivity bonus to be earned from increased social investment. It arises because of the signal it gives that governments are serious about achieving genuine equality of opportunity — where everyone can achieve their full potential irrespective of the circumstances of their birth. This positive perception helps to reduce the risk of a community backlash against further efficiency-driven economic reform.

This proposition (although not a complete myth) is debatable. When Australians are simply asked if they want lower taxes, there will always be a majority saying ‘yes’. But when they are asked to express a preference for lower taxes relative to additional spending on such things as health, education and the environment, the responses are much more positive for spending.^[19]

On government borrowing, it is true that there is probably some public hostility. But this is because politicians misleadingly call it a budget deficit and equate it with bad management and higher interest rates. Australians would respond much more positively if:

The present fiscal stance poses a basic democratic problem: by setting arbitrary fiscal targets, governments are restricting their own ability to respond to the preferences and expectations of the community. Society is being prevented from exercising the full range of choices available to it.