This version: 2 June 2000
Privatisation
John Quiggin
Australian Research Council Senior Fellow
Department of Economics
Faculty of Economics and Commerce
Australian National University
Quiggin, J. (2000), 'Privatisation', Dissent Summer(4), 39-43.
EMAIL John.Quiggin@anu.edu.au
FAX + 61 2 62495124
Phone + 61 2 82494602 (bh)
+ 61 2 62578992(ah)
http://ecocomm.anu.edu.au/quiggin
Like many popular policies, privatisation has mostly been advocated and adopted for bad reasons. The most common reason governments in developed countries have privatised assets is because of the illusory belief that the money raised in this way will allow them to increase public spending, cut taxes or repay debt. This illusion has its basis in the way governments have, until recently, presented their financial accounts.
Traditionally, the main object of the Budget was to ensure that ministers were accountable for public money, rather than to present an accurate picture of the government's financial position. Hence, accounts were presented in cash flow terms without any distinction between current and capital outlays. As a result, the proceeds of asset sales were treated exactly like current revenue (or, in some cases, as a reduction in expenditure) and, as far as the Budget was concerned, available for spending in the year in which they are realised.
This kind of dodge was employed extensively by governments in the 1980s, including the Thatcher government in the United Kingdom, and the Hawke-Keating government in Australia. It was gradually recognised that a policy of selling assets to finance current expenditure was unsustainable. A simple ad hoc response was to replace the cash measure of Budget balance with an 'underlying' measure, which excluded asset sales.
The exclusion of asset sales from the Budget balance diminishes, but does not eliminate, the spurious fiscal benefits of privatisation. The most important problem is that the Budget takes account of dividends received from government business enterprises, but not of earnings that are retained and reinvested in those enterprises. Although this practice flies in the face of all standard principles of valuation (it would imply, for example, that the Microsoft Corporation, which reinvests all its earnings and pays no dividends is worthless) it has been defended by the current government and, until recently, by the Department of Finance.
A more systematic response to the defects of the cash Budget balance measure has been to adopt a system of accrual accounting. The basic idea of accrual accounting is to separate current and capital expenditure and to recognise revenue and expenditure as they accrue, rather than when they are realised as cash payments. Accrual accounting is not perfect, but it prevents the use of some of the devices by which governments have fudged their accounts in the past. For example, the Keating government used changes in the timing of company tax to collect five quarterly tax payments in one year, thereby hiding the fact that the One Nation tax package was fiscally unsustainable. Under accrual accounting only one year (four quarters) of revenue would be recognised, regardless of the timing of payments. Similarly, the purchase price of capital assets is amortised over the life ofthe asset, rather than being lumped in with current expenditure in the year of purchase.
In principle, accrual accounting should have eliminated the apparent improvement in the budget balance arising from privatisations and other asset sales. Unfortunately, this has not been the case, largely because the principles of accrual accounting have not been applied in cases where they would undermine the attractiveness of policies dear to the hearts of Finance departments, such as privatisation.
The critical problem is the failure to value assets properly. In Australia, for example, government business enterprises like Telstra are valued at their historical cost, which is virtually zero. Hence, the sale of such enterprises, even at derisory prices, produces an illusory improvement in measures of the government's net worth. The appropriate procedure would be to use the principles of benefit-cost analysis to estimate the value of these enterprises in continued public ownership. For a partially privatised enterprise like Telstra, the sharemarket capitalisation provides an estimate of value which, though not ideal, is preferable to historical cost. If such a valuation were adopted, the sale of assets at market prices would have no effect on measures of the government's net worth. The sale of public assets at a discount, as with the first Telstra float, would be correctly assessed as reducing the government's net worth.
It is not surprising that governments that are ideologically committed to privatisation have been slow to implement their stated commitments to accrual accounting. Nevertheless, the illusory nature of the supposed fiscal benefits of privatisation has been generally recognised, both by economic commentators and by the general public. The only group who have clearly missed the point are the political commentators who continue to refer to the supposed 'war chest' of income from a Telstra sale that could be used to 'finance' electoral bribes.
The collapse of the fiscal case for privatisation does not imply that the issue is resolved. Among economists, the supporters of privatisation have always played down the fiscal arguments, preferring to argue that privatisation is desirable because it will increase the efficiency of the enterprises concerned, and of the economy as a whole.
The intellectually serious case for privatisation begins with the claim that governments should, wherever possible, leave risky activities to the private sector. Since all productive activities involve risk, this implies that, although governments may be purchasers of services, they should never be providers. The notion that governments are poorly equipped to handle risk is diametrically opposite to the view that prevailed during the thirty years after World War II when governments routinely undertook projects seen as too big or too risky for the private sector.
The reality is more complex than either of these views. Risk arises from many different sources, and has many different effects. The problem of defining the boundaries of the public and private sectors is, in large measure, a question of how different sources of risk should be handled.
Among other possibilities, risks may be borne directly by individuals and families, reallocated through financial markets and corporations, or managed collectively by governments. A central theme of the social-democratic welfare states developed after World War II was the need to reduce the risks faced by individuals and families arising from illness, economic fluctuations and so on, and to transfer those risks to society as a whole, through government. This process has been partly reversed by the attacks on the welfare state over the past two decades.
A closely related theme of the postwar welfare state was dissatisfaction with the way in which private markets handled risk. Private businesses were seen as often being unwilling to take the necessary risks to develop public infrastructure and as demanding excessive returns when they did so. This perception was reflected in the key finding of the Royal Commission set up in 1948 by the conservative Playford government in South Australia to examine the performance of the privately-owned Adelaide Electric Supply Company
Over the period of the last 24 years [to 1948], the Company has paid in dividends and interest nearly 2 million pounds more than if the Treasury rate had been paid. Future capital costs at Treasury rates would result in reduced capital costs and lower charges.
On the basis of this and other findings of inadequate performance, Playford nationalised the industry. For the following fifty years, the Electricity Trust of South Australia (ETSA) supplied electricity efficiently, met a range of social objectives and yielded returns greater than or equal to the cost of capital to the Treasury.
In 1998, however, Playford's conservative successor, John Olsen, announced plans to privatise ETSA. The necessary legislation was later passed with the assistance of Labor defectors, and with the tacit support of key figures in the Labor Opposition. Although the usual spurious fiscal arguments were presented, the crucial argument put forward in favour of privatisation was that public ownership exposed the government to unnecessary risks.
To assess the issues properly, it is necessary to consider the sources of risk in more detail. First there are operational risks, for example, risks that some particular project will fail to deliver the expected benefits or will cost more than expected. In general, such risks are best borne by the person or group with the greatest capacity to manage them. For example, in the construction of a road or building, efficiency will be maximised under a fixed-price contract, where the builder bears the loss from any cost overruns, and the benefits from any cost savings.
The matching of risks and rewards is easier in small private enterprises than in large corporations or government enterprises. Where small-scale operation is feasible, the superior capacity of small firms to manage operational risk is a crucial advantage that often enables them to survive actual or potential competition from much larger businesses. The difference between large private firms and government enterprises with respect to the capacity to manage operational risk is less clear-cut, in part because a wide range of organisational structures is available for both private and public enterprises. Nevertheless, matching of risks and rewards is easier in a private firm than in a government department that is organised on traditional bureaucratic lines. The differences are greatly reduced when government departments are replaced by commercialised or corporatised government business enterprises, but, in the process, many of the advantages of public ownership may also be lost.
A second important source of risk relates to market demand for particular goods and services. Where demand varies in response to changing tastes or local market conditions, those who respond most rapidly and appropriately will be rewarded. Hence, as with operational risks, small firms will generally outperform large ones, and private enterprises will outperform public ones. By contrast, where market demand fluctuates because of movements in commodity prices, risk is best managed through publicly-operated buffer funds or through financial market instruments such as futures markets.
While markets handle some sorts of risks very effectively, they perform poorly in responding to risk associated with fluctuations in the aggregate economy. The return demanded by investors in private equity in the average large company includes a risk premium of around 6 percentage points to compensate for the company's exposure to aggregate economic risk. That is, if the rate of interest on government bonds is 5 per cent, investors in a typical stock will expect a return of around 11 per cent. The equity premium is smaller for companies with stable returns or for those that are only weakly correlated with the economy as a whole, and larger for those with highly cyclical returns such as companies involved in construction.
The market's aversion to risk is reflected in the difference between the return demanded by investors in private equity and the rate of return on government bonds or good-quality corporate bonds. This difference is called the equity premium, and its size represents a long-standing puzzle for economists. Most economic models imply that, if capital markets spread all relevant risks efficiently and at low cost, the equity premium should be no more than one percentage point, and probably less. A variety of explanations for the 'equity premium puzzle' have been offered, most of which incorporate the failure of private capital markets to spread risk as well as is assumed in neoclassical models of the financial sector.
The final form of risk that is relevant in many cases is regulatory risk. Many government business enterprises are natural monopolies, or supply essential services where the government is ultimately responsible for ensuring security of supply. Even if such enterprises are privatised, governments cannot escape the necessity for intrusive regulation. Rational private purchasers of such an enterprise will require either guarantees of favourable regulatory treatment or a higher rate of return to compensate for the possibility of adverse changes in regulation.
Two Victorian incidents illustrate some of the issues associated with regulation of private monopolies. The first is the explosion at the Esso Longford plant. When gas supplies to Melbourne were cut, emergency action by government was required and the community bore substantial costs. Yet in making the cuts to maintenance services that contributed to the explosion, Esso was allowed to act as a private firm without external regulation. The only way in which Esso may be forced to bear the costs of its actions is through a possible class action lawsuit. As US experience has shown, litigation is an expensive and unreliable form of regulation, but privatisation will inevitably lead to increased litigation over failure to provide essential services.
The second example is an exception that proves the rule, namely, the draft decision (1 June 2000) of the Regulator-General, whcih reduces the rates of return allowed to the private monopoly owners of Victoria's electricity distribution system. The owners have protested vigorously claiming that the Kennett government had given them assurances of higher returns when they bought the assets in 1995. Since these assurances are not legally enforceable, it may be argued that Victorian taxpayers have made a good deal, selling the assets for a high price, but then imposing regulations that reduce their value. Conversely, the buyers were apparently under the delusion that the Kennett régime would last forever, or at least long enough for them to make a handy profit.
Warned by this example, however, future buyers of public assets are unlikely to accept a nod and a wink as a guarantee of favourable treatment. In the absence of binding commitments that they will receive favourable treatment, buyers will demand a substantial discount for regulatory risk. This discount reflects the fact that, under private ownership, there is an inherent conflict of interests between asset owners and consumers. In the absence of a substantial and expensive lobbying effort, the government will naturally prefer to resolve this conflict in favour of consumers.
Under public ownership, on the other hand, the government must take account of both the costs and benefits of regulatory changes. Consider, for example, a requirement for a publicly-owned electricity network to reduce its charges. The government benefits from such a requirement in its capacity as a representative of consumers, but, as the owner of the enterprise, it bears the cost in the form of lower dividends and retained earnings. In economic terminology, the risk of regulation is internalised.
Obvious though this point is, many advocates of privatisation have claimed that regulatory risk is relevant even though the same government that sets regulatory requirements bears the cost of those requirements. Indeed, they have claimed that government ownership of regulated firms creates a conflict of interest, which may be resolved through privatisation. This argument completely misconceives the role of government. Monopoly and similar problems requiring government intervention always create conflicts of interest. It is the role of government to resolve those conflicts, and this is most effectively done by internalising the relevant risks.
The assessment of proposals for privatisation or nationalisation requires a balancing of these risks. As a first approximation, it is possible to compare the interest savings associated with using the proceeds of privatisation to repay debt with the flow of earnings under public ownership. If the interest savings exceed the earnings foregone through privatisation, then there is a prima facie case for supposing privatisation to be beneficial. However, most Australian privatisations, and all privatisations where the asset has been sold by public float, fail this prima facie.
As has already been noted, it is an elementary error to count only dividends and disregard retained earnings. It is also important to take account of differences in the tax treatment of interest on bonds compared to company profits. In the absence of other considerations, it would be appropriate to compare interest savings to Telstra's earnings net of company tax, since both represent pretax income to the holders of bonds and shares respectively. However, the impact of company tax is almost completely offset by two major concessions to shareholders. The first is dividend imputation, which allows the company tax on the component of profits paid out as fully franked dividends to be offset against personal income tax. The second is the concessional treatment of capital gains arising from the Ralph 'reforms' of 1999.
The most costly single failure was the sale of one third of the shares in Telstra in 1997. One third of the government's shareholding, consisting of approximately 4.3 million shares, was sold at an average price of $3.40, yielding sale proceeds of $14 billion. This privatisation was claimed as a success because strong demand for shares ensured a sale price at the upper end of the range of expectations, and because purchasers of shares enjoyed strong capital gains, reflecting the oversubscription of the share issue.
From the viewpoint of taxpayers, however, the existence of capital gains for buyers was not an indication of success, any more than the shareholders of a private firm would have regarded as successful the sale of an asset at a price which allowed buyers to make immediate profits on resale. The first-stage partial privatisation of Telstra exemplified the tendency for public assets to be sold for less than their market value, which in turn is usually less than the value of the stream of profits accruing to the public sector from ownership of those assets. As has commonly been the case in staged floats, the second stage was more satisfactory from the viewpoint of taxpayers. The shares were sold for $7.40, so that the government received more ($16 billion) from the sale of 16 per cent of Telstra in 1999 than it had two years earlier for 33 per cent. The share price initially rose above the float price, but later fell (to $6.92 on 1 June 2000), so that purchasers incurred capital losses, at least in the short term. The headline coverage given to this outcome in major newspaper indicates the strength of the prevailing viewpoint that privatisation by public float represents a gift of public money to buyers.
A more fundamental question in evaluating the partial sale of Telstra is the comparison between the earnings foregone as a result of privatisation, and the interest saving from the repayment of debt. Telstra's pretax profit for 1998-99 was $5.3 billion, and a recent statement indicates that 1999-2000 earnings will exhibit 'double digit growth', suggesting that earnings are likely to be around $6 billion. Thus, the earnings foregone through the sale of half of Telstra are currently around $3 billion per year, of which $2 billion are attributable to the first-stage privatisation and $1 billion attributable to the second-stage.
By contrast, the total interest savings from privatisation (at the current 10-year bond rate of 6 per cent) are around $1.8 billion, of which $850 million are attributable to the first tranche and $950 million to the second tranche. This analysis confirms the view, obvious from the jump in the share price, that the first-stage privatisation was a fiscal disaster. The second stage is likely to produce a small loss in 1999-2000. However, even assuming no real growth in Telstra's earnings from now on, the loss will grow over time because of inflation, while the interest saving will remain fixed in nominal terms.
A purely financial analysis of this kind is only a first step in an evaluation of privatisation. If privatisation is accompanied by improvements in service to consumers, or better working conditions for employees, it may be beneficial to society as a whole even if it generates a financial loss to government. Unfortunately, privatisation has rarely produced favourable outcomes for consumers, and never for employees, with the almost invariable exception of senior managers, whose salaries have increased greatly.
If the argument that most past privatisations have reduced public sector net worth is accepted, it follows that governments should consider the option of re-nationalising some privatised enterprises. This might appear to be an unthinkable proposal, but privately-owned water supply systems and profit-making welfare services would have seemed equally unthinkable a couple of decades ago. Claims that 'there is no alternative' and that 'there can be no U-turns' have been used by advocates of economic reform unwilling or unable to demonstrate to the public that the policies that they favour are beneficial. Similar, and equally invalid, claims were made by socialists in the post-war period when growth in the size and role of government seemed inevitable.
Just as a variety of methods of privatisation have been used, there are a range of possible options for re-nationalisation. In the case of enterprises such as Telstra that have been partially privatised but retain majority government ownership, the enterprise itself can make a share buyback offer, offset, if necessary, by a capital injection from the government. More generally, the government, acting either directly or through a publicly-owned holding company, can seek acquire companies through takeover bids, just as a private bidder would do.
More traditional routes to (re-)nationalisation are also feasible. Enterprises may be nationalised through legislation, subject to constitutional requirements for fair compensation. Alternatively, governments may set up new enterprises to take the place of those they have sold. This would be the most likely method for a re-entry into the industries such as banking and insurance where government and private enterprises have historically competed against each other.
Assuming that re-nationalisation is undertaken with full compensation, the public would be unable to recoup any losses arising from the fact that share issues were priced below market value, as in most privatisations by public float. In effect, the discount was a gift from the government to the participants in floats, who were predominantly upper income earners. However, the public would be better off as a result of re-nationalisation in all cases where the value of the enterprise in public ownership exceeded its private market value because of a more efficient allocation of systematic and regulatory risk.
Privatisation in Australia and other developed countries has been supported by fallacious arguments and inadequate analysis. Not surprisingly, the result is that most privatisations have reduced public sector net worth. Widespread recognition of the weakness of fiscal arguments for privatisation has contributed to the rejection of a number of major proposals for privatisation. Elections in New South Wales and Tasmania in 1998, fought mainly on the issue of electricity privatisation, resulted in overwhelming defeat for the advocates of privatisation. (In both cases, the Liberals supported privatisation and Labor opposed it.) More recently, rural opposition has forced the National Party to withhold support for the full privatisation of Telstra.
In future, it is to be hoped that debate about proposals for privatisation or re-nationalisation will be informed by sound economic analysis rather than fiscal fallacies. The central principle of such economic analysis is that the value of assets is maximised when risk is allocated appropriately. The choice between public and private ownership depends on the balance between economy-wide systematic risk, regulatory risk and enterprise-specific risks relating to operating efficiency.