Go back to John Quiggin's home page

Options for privatisation of the HEC: An economic assessment

 

 

Professor John Quiggin

Australian Research Council Senior Research Fellow

James Cook University

 

 

 

 

EMAIL John.Quiggin@anu.edu.au

FAX + 61 7 47814149

Phone + 61 7 4781 4798 (bh)

+ 61 7 47251269 (ah)

 

 


Options for privatisation of the HEC: An economic assessment

The purpose of this paper is to provide an economic assessment of the proposed partial privatisation of the Hydro-Electric Commission (HEC). The analysis differs from those presented by Bain/Salomon and the Treasury in two important respects. First, the previous analysis have been based on the cash flow conventions of the traditional budget framework and have therefore confused capital and current expenditure and failed to take appropriate account of retained earnings. The present analysis deals with the full economic impact of privatisation in a manner consistent with the principles of accrual accounting.

The second main feature of the analysis presented here is an analysis of risk allocation based on the fundamental principle that risk should be borne by the party best able to manage it. It is argued that regulatory risk is best managed within the public sector and that the transfer of regulatory risk from the public to the private sector results in a loss of economic value.

The paper is organised as follows. Section 1 sets out the economic principles governing the appropriate allocation of risk. Sections 2 and 3 discuss recent developments in the electricity industry, and particularly the disaggregation of integrated enterprises into generation, transmission, distribution and retail components in the light of the theoretical analysis presented in Section 1. Section 4 is a discussion of mechanisms of privatisation. Section 5 sets out the theoretical framework for the assessment of privatisation proposals, and discusses common fallacies associated with attempts to assess privatisation using cash flow measures. Section 6 presents estimates of the retention value of the HEC and provides comparison with the effects of sale at prices ranging from $1.5 billion to $2 billion.

1. The allocation of risk

A critical issue in evaluating the desirability of public and private ownership is the appropriate allocation of risk. The central principle in the allocation of risk is risk should be borne by the party best able to manage it.

Government business enterprises in Australia have a long history of undertaking risky investments, in general with a high degree of success. Advocates of privatisation have claimed that it will free taxpayers from the risk associated with ownership of business enterprises. However, they have often confused enterprise-specific risks, which are best handled by the owners of enterprises, with regulatory and systematic risks, which are best handled by governments.

Enterprise-specific risk

Enterprise-specific risk is derived from factors specific to a given enterprise, such as the effects of good or bad management. Most uncertainty about operating costs is enterprise-specific, as is much uncertainty about market share.

Enterprise-specific risk for one enterprise is uncorrelated with the enterprise-specific risks facing firms in general or with fluctuations in the economy as a whole. Because enterprise-specific risks are uncorrelated, when a large number of enterprises subject only to enterprise-specific risks are combined the aggregate rate of return will display very little risk. The capacity to pool risk through private mechanisms, such as insurance and portfolio diversification, mean that enterprise-specific risk, qua risk, is unimportant in evaluating the market value of a enterprise or enterprise.

However, the existence of enterprise-specific risk gives rise to agency problems, which are analysed by economists using principal-agent theory (Laffont 1989, Chapter 11). The theoretical framework for principal-agent theory is based on the idea of a productive activity characterised by enterprise-specific risk. Agency problems arise when one party (the agent) undertakes a enterprise on behalf of another party (the principal). The agent has private information about the outcomes of enterprise-specific risk. If returns are uncertain and the agent has private information or some other strategic advantage, it will be preferable, other things being equal, for the agent to bear the risk associated with the venture. If the risk is assumed by the principal, the agent will have an incentive to shirk or to divert some of the assets to private uses, then to claim that the bad outcomes of the venture were simply the result of bad luck.

The central implication of the principal-agent literature is that, where possible, the party that has most control over risk should be the owner, that is, the recipient of the residual income. In cases where enterprise-specific risk is associated with response to firm-specific market conditions or with management skill, the implications of principal-agent theory support private ownership.

Regulatory risk

Regulatory risk is the risk arising from changes in government policy, including public authorities established to regulate particular sectors of the economy. For firms operating in competitive markets, regulatory risk is normally a fairly minor concern. However, for firms in natural monopoly industries, such as electricity transmission and distribution, or in closely related activities such as electricity generation, regulatory risk is crucial. The returns to electricity distribution will be determined directly by the decisions of regulators. Although regulation does not directly determine the returns to generation, decisions regarding the structure of electricity pool markets are the primary determinants of the long-run average returns to participants in those markets.

Regulatory risk is most appropriately borne by the government of the relevant jurisdiction. The general principle that risk should be borne by the party best able to manage it applies in this case. Regulatory decisions are necessarily complex, and the best guarantee of desirable outcomes is that the agent making the decision should bear both costs and benefits. This condition applies when governments regulate the returns to publicly owned monopoly enterprises. The benefits of higher prices are received by citizens as taxpayers in the form of increased profitability while the costs are borne by citizens as consumers. Governments represent citizens in both capacities and must therefore take account of both costs and benefits. That is, regulatory risk is internalised. By contrast, in the case of privately owned monopolies, there is a direct conflict between the interests of shareholders and those of citizens.

The fact that governments do not bear any net risk may be further illustrated by the point that, if they choose, governments can fully offset any variation in the net earnings of transmission and distribution enterprises by varying the levy paid by those enterprises, which currently takes the form of a 5 per cent contribution to consolidated funds.

Systematic risk

Systematic risk refers to the risk associated with fluctuations in aggregate output. Because systematic risk is correlated across projects, it is not eliminated by risk pooling. Nevertheless, market mechanisms such as portfolio diversification mean that systematic risk can be spread. A number of studies have shown that the private capital market does not spread systematic risk perfectly. There is a large divergence between the rate of return demanded by private equity holders and the real rate of return on public debt or good quality private debt. This divergence cannot be explained as a pure risk premium, at least on the basis of standard life-cycle consumption models, and hence has become known as the 'equity premium puzzle'. As a result, the equity premium (the difference between the rate of return on equity capital and the rate of return on low-risk assets, such as high-quality bonds) is larger than it would be under the 'efficient markets' hypothesis.

The efficient markets hypothesis, which is equivalent to the assumption of zero transactions costs, is implicit in much current advocacy of privatisation and private sector infrastructure provision.

The equity premium may be explained as the result of imperfect risk-spreading by private sector capital markets and the existence of transactions costs which restrict borrowing and lending (Mankiw, Lucas and Heaton), so that capital markets do not work costlessly or perfectly efficiently. In particular, it is not possible to take out insurance against losses incurred as a result of recession (Mankiw 1986, Constantinides and Duffie 1996). Hence the premium required for bearing systematic risk is substantial, whereas, under perfect capital markets, it would be negligible (Quiggin 1995). Public ownership implies a capacity to spread risk through the tax system, thereby providing a degree of insurance against systematic risk. The superior capacity of governments to bear systematic risk implies the desirability of public ownership of capital-intensive enterprises such as those in the infrastructure sector.

2. Recent developments in the electricity industry

In the last few years, the structure of state-owned electricity enterprises in most states, including Tasmania, have undergone radical changes. In large measure, these changes are a response to the development of a national electricity market and to the requirement for competitive reforms under the Competition Principles Agreement of 1995 and the associated Competition Policy Reform Act 1995 (Cwlth), usually referred to as the Hilmer reforms, after the report of the Hilmer Committee (Hilmer et al. 1993). Under the Competition Principles Agreement, failure by state governments to implement competitive reforms to the satisfaction of the National Competition Council may lead to the withdrawal of financial assistance grants from the Commonwealth.

Hilmer and others have stated on many occasions that the implementation of the Hilmer reforms does not require privatisation. Indeed, the terms of reference of the Hilmer Committee required them to identify policies that would enable public and private enterprises to compete on equal terms. Nevertheless, the need to comply with competition policy has been used as a pretext by governments wishing to privatise government business enterprises in the electricity industry and elsewhere.

Under the National Electricity Market, which is now coming into operation, generators bid to supply electricity to State pools. Large consumers may purchase electricity directly, while residential consumers contract with retailers who purchase electricity from the pool. Transmission and distribution of electricity remains a monopoly activity subject to price regulation.

It is important to note that, despite the description of the new system as a 'national' market, it is in reality a set of interconnected state markets. Because in the past, each state operated an independent system, interconnections allowing the transmission of electricity from one state to another have limited capacity. Because of the large distances between major cities in Australia, expansion of interconnection capacity is an expensive option.

It remains unclear how the National Electricity Market will work in practice. In particular, it is not yet apparent:

(i) Who, if anyone, is responsible for the reliability of supply of electricity;

(ii) Whether market prices will give appropriate signals for new investment in generation and transmission capacity;

(iii) Whether the generation and distribution enterprises created over the past few years are large enough to achieve economies of scale (or, equivalently, whether mergers resulting in horizontal reintegration will be required); and

(iv) Whether the separation of the industry into generation, distribution and retail components is economically sustainable, (or, equivalently, whether mergers resulting in vertical reintegration will be required).

On points (iii) and (iv), representatives of the private component electricity industry are already advocating reintegration, primarily through mergers across state boundaries. It appears that the competitive phase of the national market will be short-lived.

From the viewpoint of suppliers of generation, transmission and distribution services, the question of responsibility for reliability raises concerns about possible legal liability in the event of large-scale breakdowns such as those that occurred in Auckland and, on a smaller scale, in Queensland earlier this year.

All of these difficulties raise doubts as to whether National Competition Policy, and, in particular, the regulatory structure currently planned for the National Electricity Market, will prove politically and economically sustainable. A number of recent developments suggest that substantial regulatory change may take place:

(i) the design of the WestLink connection between Queensland and New South Wales reduces the capacity for exports from New South Wales to Queensland relative to the original EastLink plan

(i) the newly elected government of Queensland has announced reintegration of the industry, a policy directly contrary to that advocated by supporters of National Competition Policy

(ii) there is considerable pressure to review the disaggregation undertaken in New South Wales following the unsuccessful move to privatise the electricity industry in that state

(iii) the South Australian government has announced that the RiverLink connection to the New South Wales electricity grid will not proceed

(iv) the draft decision of the Victorian Regulator-General, proposing a 7 per cent real rate of return for gas distributors has raised concerns about the implications of similar decisions being applied to the electricity industry

 

The absence of a stable regulatory structure for the electricity industry implies that participants in the industry face significant regulatory risk. Returns will be affected at least as much by changes in regulatory policies as by fluctuations in demand and changes in costs.

3. Generation, transmission, distribution and retailing

The generation activity is potentially competitive, though the extent of competition in any given state market depends on the number of separate generators that can operate efficiently and on the capacity of interconnections between state markets. Under the National Electricity Market, generators bear significant enterprise-specific risk as well as systematic risk. Given the possibility of changes in competition policy or in the structure of National Electricity Market generators also face significant regulatory risk.

The transmission activity is a natural monopoly, which will be subject to rate-of-return regulation under the new market arrangements. Hence, the operators face no significant market risk, but substantial regulatory risk. As has been argued above, regulatory risk is internalised and no risk premium should be associated with the returns to a publicly owned regulated monopoly.

The distribution and retail activities are combined into a single enterprise in the disaggregated HEC. However, in economic terms, the two activities are quite dissimilar. The physical distribution of electricity is a natural monopoly activity, similar to transmission. Whether the charge for distribution is included in the distribution-retail margin of an integrated distribution-retail enterprise, or is set as an access charge for independent retailers, it must be determined by regulation. On the other hand, retailing is a low-margin activity with modest capital requirements. Retail activities may be undertaken directly by the distribution enterprise, or contracted out, or undertaken by independent retailers. Whichever outcome emerges from the new market structure, the profitability of the distribution enterprise will only be marginally affected.

This distinction has not been appreciated in analysis of the likely effects of competitive entry on the profits of the distribution-retail enterprise, which have suggested that a large reduction in the profit margins of the distribution-retail enterprise would arise from entry. As argued above, the distribution-retail margin is determined by the implicit or explicit charge for access to the physical network, and this charge can be varied by regulation with or without competitive entry.

The current proposal for privatisation involves maintaining public ownership of the generation activity, while privatising transmission, distribution and retailing. Since market risk is borne primarily by the generating activity, the proposal will not reduce the market risk faced by the Tasmanian government. The privatisation of transmission and distribution will create regulatory risk which is currently internalised through public ownership. The privatisation of retailing is of secondary importance, since any benefits obtained in this manner could be matched by contracting out retail operations.

4. Methods of privatisation

A number of mechanisms have been employed for privatisation. The most common are: sale to an existing firm; public flotation; and long-term leases.

A number of privatisations have taken the form of sale of a GBE to a private enterprise, usually either a domestic competitor or, more frequently, a multinational firm active in the same industry. Examples of the latter include the sale of 25 per cent of Qantas to British Airways, the sale of Aussat to Optus (partly owned by a consortium of overseas telecommunications companies) and the sale of New Zealand Telecom to two United States telecommunications companies.

The main motivation for sale to an existing firm is that it often yields the highest available price, and hence the greatest revenue for governments. If there exist economies of scope in the enterprise concerned, then it will be worth more as a part of a larger multinational enterprise than as a stand-alone firm. In some circumstances, the combined enterprise may exert increased market power, and hence potentially extract monopoly profits.

Public floats are typically motivated by a mixture of political concerns and a desire for competition. The political concerns include a desire to build political support for privatisation by creating a class of beneficiaries from the issue of shares at prices less than their true value. This typically motivates the generous terms under which privatisations have taken place while the concept is new and controversial. A related motivation is a desire to build a large class of shareholders. This is seen as politically promoting support for conservative parties, since shareholders are generally believed to be more likely to vote for them. The partial gift approach based on underpricing and special concessions has been an even more prominent feature of the privatisation program in the United Kingdom, where the vote-buying element is quite frankly acknowledged. A wide distribution of shareholding is also sometimes seen as reducing the large scale concentrations of economic and political power normally associated with big business.

The use of long-term leases as a method of privatisation has been motivated partly to avoid political opposition to outright privatisation and partly by a desire to maintain some control of the assets being leased. The fact that the buyers become lessees rather than outright owners implies that the value of the asset will be less than in the case of an outright sale. The extent of the discount will depend on the terms of the lease, and on the perceived likelihood that those terms might be unilaterally varied in the future.

In general, the more restrictions are imposed on buyers, the lower will be the sale price of the asset. For this reason, trade sales are, in general, the most desirable method of privatisation. If the asset in question is too important to be sold to a foreign corporation, it is probably too important to be sold at all.

 

5. Comparing sale price and retention value

Whenever the owner of an asset contemplates selling that asset it is necessary to consider whether the sale price that can be realised is more or less than the value of retaining the asset. In the absence of non-economic considerations, sale of the asset will be justified if and only if the sale price exceeds the retention value.

In general, the sale price is a lump sum, while the retention value depends on estimates of future income flows arising from ownership of the asset. Comparisons of sale price and retention value may therefore be made in one of two ways. First, the estimated value of future income values may be converted to a lump sum using an appropriately chosen discount rate. Alternatively, the sale price may be converted to an income flow, for example, by assuming that sale proceeds are used to repay debt, and this income flow may be compared to the estimated flow of earnings under continued ownership. Although the two procedures are logically equivalent, the analysis based on flows is more transparent, and allows for the use of a range of projections of future earnings.

The method discussed above is equally applicable to privately owned and publicly owned assets. Before developing the analysis further it is necessary to deal with two common fallacies which have been put forward in the debate over privatisation of publicly owned assets. These will be referred to as the current income fallacy and the dividend valuation fallacy.

The current income fallacy

The first countries to adopt policies of large scale privatisation in the 1980s were the United Kingdom, Australia and New Zealand. In all three countries, it was assumed that the proceeds of privatisation could be treated as if they were current income. For example, in his memoirs, former Prime Minister Hawke (1994, p. 391) says:

When it came to privatisation, Brian [Howe] was able to argue that one dollar simply could not do two jobs. Putting a dollar of equity into the running of an airline was paid at the expense of programs for the unemployed or the single mother.

The governments of the United Kingdom and New Zealand used privatisation proceeds to finance large cuts in income taxes. In all cases, the result was the same. Once the supply of assets for sale was exhausted, there was no source of revenue to finance the increased expenditure or reduced taxation. Not only were there no further privatisation proceeds, but the flow of income formerly generated by the assets was lost. In all three countries, the initial privatisations contributed to the growth of intractable budget deficits.

The fallacious assumption that privatisation proceeds were equivalent to current income was based on the conventions used to draw up the Budget, under which the proceeds of asset sales are treated as negative outlays. In assessing privatisation, it is important to take account of economic reality rather than accounting conventions. Since the fallacy of treating sale proceeds as current income has now been generally recognised, attention is now generally focussed on measures of the underlying Budget deficit, which excludes the proceeds of asset sales. Unfortunately, the use of part or all of the proceeds of asset sales to finance 'pork barrel' expenditure remains widespread.

The dividend valuation fallacy

The idea, derived from Budget conventions, that the proceeds from asset sales can be treated like current income is now generally recognised as fallacious. However, a subtler form of the fallacy remains influential. By convention, government business enterprises are located in the 'non-budget' sector of government. The earnings of these enterprises are therefore not counted as part of the government's budget income. Instead, budget income includes only dividends paid by government business enterprises into the government budget. Hence, analysts have frequently compared the flow of dividends foregone with the interest saved by using privatisation to repay debt. This approach to a comparison of sale price and retention value may be referred to as the dividend valuation fallacy. To be more precise, the dividend valuation fallacy is the claim that an enterprise should be valued by its owners solely in terms of the flow of dividends it generates and that retained earnings are in some sense 'locked up', and inaccessible to the owners,

In general, dividends are less than earnings since some earnings are retained to finance future investments. However, there is no reason why dividends cannot exceed earnings. In 1995-96, for example, Telstra paid a special dividend of $3 billion, in addition to its ordinary dividend of $1 billion, even though earnings for that year were only about $2 billion. A number of state government corporations, in the electricity sector and elsewhere, have paid similar special dividends. These occurrences illustrate the point that dividends can be set at any level, either above or below the level of earnings.

This simple observation ought to be sufficient to refute the idea that public enterprises such as the HEC should be valued in terms of dividends rather than earnings. However, since the fallacious analysis in terms of dividends has been repeated widely, it is worth analysing in more detail.

Until the 1950s, the dividend valuation fallacy was common in valuations of private enterprises. This fallacy was refuted by the economists Modigliani and Miller (1958), both of whom received the Nobel Prize for their work. Modigliani and Miller showed that, in the absence of differential tax treatment, and assuming that capital markets work smoothly, the value of shares in an enterprise is unaffected by the dividend policy adopted by that enterprise. The interests of shareholders are not affected whether the enterprise pays out all its earnings in dividends, using new issues of equity and debt to finance new investments, or pays no dividends, using retained earnings to finance new investment. In the latter case, shareholders who wish to realise the income associated with retained earnings can do so by borrowing against the increased value of their shares.

The analysis of Modigliani and Miller forms the basis of the modern theory of finance. In particular, it leads directly to the Capital Asset Pricing Model and its generalisations such as Arbitrage Pricing Theory. In these models, firms are valued according to the expected flow of future earnings to shareholders, with an adjustment for risk. Dividend policy is irrelevant to the valuation of the enterprise. A policy of paying high dividends and financing new investment through debt will result in a higher debt-equity ratio, commonly referred to as a higher level of 'gearing'. This will imply a higher, but riskier rate of return to equity. The Capital Asset Pricing Model shows that the effects of higher earnings and higher risk cancel out, confirming the basic result of Modigliani and Miller.

In practice, considerations of corporate governance lead firms to avoid very high levels of gearing (which arise if all earnings are paid out as dividends) and very low levels of gearing (which arise if all earnings are retained). The relationship between corporate governance and dividend policy has been analysed in recent years using concepts of agency theory. With these minor modifications, however, there has been no successful challenge to the Modigliani-Miller analysis since it was put forward forty years ago. The standard finance text by Fama and Miller (1972), presents the same analysis, as do Australian texts such as Juttner (1987) and analyses of market valuation in Australia such as that of Higgins, Johnston and Coghlan (1976), who suggest that on average, a valuation based solely on the flow of dividends represents less than half the value of a business enterprise.

A representative discussion is the article on dividend policy by Brickley and McConnell (1987, p. 120), who observe that:

After a brief flurry of debate, the Modigliani-Miller irrelevance proposition was essentially universally accepted as correct under their assumptions [of neutral tax treatment and full information]'.

Brickley and McConnell go on to discuss reasons why shareholders might not be totally indifferent to dividend flow, such as the considerations of corporate governance mentioned above. However, the idea that retained earnings should be disregarded in the valuation of an enterprise, or treated as being less valuable to shareholders than earnings paid out as dividends, is generally recognised as a fallacy.

After some debate, the majority of economists in Australia have accepted the point that retained earnings should be taken into account. The change in the views expressed by the Commonwealth Department of Finance is of particular significance. In evidence to the 1996 Senate inquiry into the partial privatisation of Telstra, the Department claimed that the Modigliani-Miller proposition was 'an academic theory with no practical relevance to the real world' and that retained earnings were 'locked up for ever and never used'. The Department therefore endorsed the view that privatisation would lead to an improvement in the Commonwealth's fiscal position.

In the 1998 inquiry into the proposal for full privatisation, the Department's Office of Asset Sales and Information Technology abandoned the position taken in 1996 and presented a position consistent with the Modigliani-Miller analysis, saying that, in the presence of perfect capital markets, the sale price would be equal to the present value of future dividends and retained earnings. Therefore, the Office of Asset Sales said, under these conditions, privatisation would be neutral. That is, there would be neither a benefit nor a loss to the public. The Office further conceded that the public could be worse off if capital markets were imperfect or if the privatisation procedure were mishandled.

Approach to valuation

The discussion above leads to two critical conclusions about the approach to valuation. First, the analysis must be undertaken in terms of net worth, rather than in terms of budget conventions and cash flow. Second, the principles of risk allocation must be taken into account.

The need to base the analysis on net worth means that both flows to the public sector and earnings retained within the government business enterprise sector must be taken into account. If this principle is not observed, policies that involve a transfer of net wealth from the government business enterprise sector to the budget sector will appear beneficial, even though they may reduce net worth. This problem arises in the Treasury analysis of the impact on the consolidated fund of the partial sale of the HEC. Because this analysis is based purely on cash flows to the budget sector it yields the conclusion that

(i) withdrawal of equity from the HEC is necessarily beneficial

(ii) repayment of general government sector is preferable to repayment of HEC debt

Neither of these conclusions is valid. Since both HEC debt and budget sector debt are publicly guaranteed it makes no economic difference whether debt is located within the HEC or within the general government sector. If conclusion (ii) were valid, it would be possible to address the problem of excessive debt levels in the general government sector simply by requiring large special dividends from the HEC or other government business enterprises, financed by the issue of additional debt. It should be evident that such transactions would make no difference to net worth. Once it is recognised that (ii) is invalid, it should be clear that (i) is also false.

As shown above, the valuation of risk for publicly owned enterprises is not the same as the valuation of risk for private enterprises. If this were not the case, it would make no difference whether enterprises were publicly or privately owned. Three principles should be used in the valuation of risk

(i) regulatory risk should be disregarded in the valuation of enterprises in public ownership since it is fully internalised

(ii) systematic risk is less costly for public enterprises than for private enterprises because of government's capacity to spread risk through the tax system

(iii) enterprise-specific risk is more costly for public enterprises than for private enterprises

The effect of privatisation on net worth may be undertaken by computing a present value for returns under privatisation, and comparing it with sale proceeds. This procedure is relatively simple, but not particularly transparent as it depends crucially on the choice of risk-adjusted discount rates. A more transparent procedure is to compare future flows of income under retention and sale. In making this comparison, it is crucial to focus on net worth rather than cash flow.

6. Estimates of retention and sale value of the HEC

Attention in this analysis will be focused on the proposal to sell or lease the distribution, transmission and retail components of the HEC. Since the status of the generation enterprises will be unchanged, the most appropriate approach is to exclude these enterprises from consideration. However, to facilitate comparison with other analyses the results will also be presented in the form of a comparison of the value of the entire HEC under the options of retention and partial privatisation. Three projections will be presented. The first is designed to be comparable to those undertaken by the Treasury and Bain/Salomon. It is assumed that revenues contract in real terms by around 10 per cent over the next decade, but that this contraction is offset by reductions in operating costs and interest payments. In the second projection, it is assumed that revenue remains constant in real terms. The third projection is based on the assumption that revenues for the transmission and distribution sector are determined by rate of return regulation, with the real pretax rate of return to capital being set at 7 per cent, as in the recent draft determination for the Victorian gas industry.

A number of additional assumptions, which do not affect the economic comparison are made to derive projections of income flows. The allocation of assets and capital to the distribution, transmission and retail enterprises follows that proposed by CS First Boston, with the book value of assets being approximately $1.4 billion and debt of $350 million. For simplicity, capital expenditure is set equal to depreciation. A 70 per cent dividend ratio is assumed. This is consistent with a gradual retirement of debt within the enterprises. The annual rate of inflation is assumed to be 2 per cent and the nominal interest rate is 7 per cent.

The comparison of net worth under retention and sale is unaffected if debt is transferred between the generation and transmission enterprises. Similarly, as has already been noted, the value of the enterprise to its owners is independent of dividend policy. Provided the real interest rate is constant at 5 per cent , changes in the rate of inflation have no real effect. Further, provided additional capital investments are appropriately taken into account in setting the regulated rate of return, differences in assumptions about capital expenditure also make no difference to comparisons of net worth under retention and sale.

This invariance to arbitrary choices about financial structure is the key advantage of an analysis based on net worth over one based on cash flows in the general government sector. As has already been observed, the cash flow approach leads to spurious conclusions such as that of the desirability of transferring debt into the government business enterprise sector.

As noted above, the first scenario is based on the assumption that real revenues contract over time. It is important to recall that an outcome of this kind for the transmission and distribution enterprises can arise only as a result of regulatory decisions designed to reduce the cost of electricity.

 

Table 1: Projections of revenue and earnings for HEC Transmission and Distribution

Low projection

a: Net of payments for electricity

b: Operating expenses excluding depreciation, amortisation and abnormals

c: Earnings before interest, tax and abnormal and extraordinary items

In Table 1, the first column shows the financial year, the second shows the project revenue net of payments for electricity. The third column shows expenses excluding deprecation. The fourth column shows EBIT (earnings before interest and tax) which is obtained by subtracting expenses and depreciation from revenue. The fifth column shows interest payments, and the sixth shows pretax profit (EBIT - interest). The last two columns show the breakup of pretax profit between payments to the general government sector (dividends and tax equivalent payments) and retained earnings.

In a comparison with the option of sale the critical variable is EBIT, since under a sale, the government forgoes both the company tax equivalent payment and the earnings of the company, while retaining the obligation to pay interest or, equivalently, to use part of the sale proceeds to require debt. A break-even sale price must yield annual interest savings of around $130 million per year. If the real interest rate is assumed to be 5 per cent (on the assumptions used in this analysis this implies a nominal rate of 7 per cent), the break-even sale price is approximately $1.9 billion.

As noted above, the EBIT projections derived here are similar to those used in the Treasury and Bain/Salomon analyses. The difference in the results arises from the fact that the analysis presented here is an analysis of full economic value rather than cash flow. The Treasury analysis erroneously excludes retained earnings from consideration. Since the dividend ratio in that analysis is set at 50 per cent, only one half the net earnings of the enterprise is taken into account. The Bain/Salomon analysis erroneously treats capital expenditure as if it were current expenditure. This error is particularly egregious in the analysis of an enterprise subject to rate of return regulation, where new capital expenditure enters the capital base and therefore directly ensures an increase in future revenue.

A smaller part of the difference arises from the fact that no discount has been made for risk. As has been argued above, risk in the earnings of the enterprise relates almost exclusively to the decisions of government regulators and is therefore internalised under public ownership.

The assumption of declining real revenue used in the projections presented above is fairly pessimistic, bearing in mind that fluctuations in the market price of electricity will primarily affect the generation enterprise. An alternative projection based on constant real revenue is presented in Table 2.

 

Table 2: Projections of revenue and earnings for HEC Transmission and Distribution

Stable projection

a: Net of payments for electricity

b: Operating expenses excluding depreciation, amortisation and abnormals

c: Earnings before interest, tax and abnormal and extraordinary items

Because EBIT grows steadily under this projection, a break-even sale price must yield interest savings in excess of initial EBIT foregone. The excess of interest savings over EBIT foregone can be used to retire further debt so that the present value of savings is equivalent to that from retention. Depending on assumptions about earnings after 2009, the required flow of interest savings will be between $150 million and $160 million implying that the break-even sale price is in the range $2.1-2.3 billion.

Table 3: Projections of revenue and earnings for HEC Transmission and Distribution

Rate of return regulation projection

 

a: Net of payments for electricity

b: Operating expenses excluding depreciation, amortisation and abnormals

c: Earnings before interest, tax and abnormal and extraordinary items

The final simulation is based on the assumption that earnings are set by regulation fixing the real rate of return at 7 per cent, with revenue being determined accordingly. Break-even requires initial interest savings of around $150 million or a sale price of around $2 billion.

Analysis of sale options

The analyses of retention options presented above provide a simple basis for the estimation of a break-even sale price. However, it may be useful to provide a more detailed projection of the implications of sale options. Two options are considered with sale prices of $1.5 billion (12 times initial EBIT) and $2 billion (18 times initial EBIT). As noted by CS First Boston, the Victorian distribution assets yielded an average sale price of 12 times EBIT. Since then, interest rates have fallen, but this is offset, or more than offset by the withdrawal of infrastructure tax concessions. Hence, a sale price of around $1.5 billion appears realistic, and, as indicated above, such a price is unlikely to compensate the government for the loss of income flows from the transmission and distribution enterprises. However, in considering asset sales it is important to avoid a dogmatic position in favour of, or against, sale at any price. The analysis of the $2 billion sale price is included to show that, at high enough sale prices, privatisation will yield fiscal benefits.

In both cases, the comparison is undertaken relative to the projection presented in Table 1. Thus, it is assumed that, for both retention and sale options, regulation is directed towards the achievement of low electricity prices and, more importantly, low margins for transmission and distribution. To maintain comparability, it is assumed that current government expenditure is equal in all scenarios. Thus, if the interest savings from privatisation exceeds the flow to the budget sector under retention, the excess is used to repay debt. It is assumed that the debts associated with the transmission and distribution enterprise under the retention scenario are marked to market at a value of $400 million and repaid out of the proceeds of the sale. Note once again, that unlike the Treasury analysis, changing assumptions about the allocation of debt between the electricity enterprises and the budget sector makes no difference to the comparison of retention and sale options.

 

Table 4: Comparison of sale and retention options for HEC Transmission and Distribution

Based on low projection of earnings

The pretax profit shown in Table 4 is taken from Table 1 and is the income foregone by the public sector as a whole through sale of the asset. The following two columns are based on a sale price of $1.5 billion, with $400 million allocated to the repayment of debt allocated to the transmission and distribution enterprises. (It should be noted once again that different assumptions about the allocation of debt between the general government and government business enterprise sectors make no difference to the final comparison between sale and retention options).

Although interest savings are consistently less than the foregone earnings, there is an initial increase in cash flow to the general government sector which is used to repay additional debt. By 2002, interest foregone is less than dividend and tax equivalent flows to the general government sector and borrowing must increase to finance a level of government expenditure equal to that under the retention projection. The present value of the loss over the period 2000-2009 is approximately $300 million.

By contrast, with a sale price of $2 billion, there is a consistent net gain, though this gain declines over time, reflecting the fact that earnings foregone are growing in nominal terms. Over the period 2000-2009, the present value of benefits is approximately $90 million. These results are consistent with the observation derived from Table 1, that a break-even sale price under the low projection is approximately $1.9 billion.

Concluding comments

Based on the low projection presented in Table 1, the sale price required to yield a net financial benefit from privatisation is approximately $1.9 billion. The claims by Bain/Salomon and Treasury that net benefits could be achieved with a sale price around $1.5 billion are based on analytical errors. Bain/Salomon incorrectly treats capital expenditure as if it were current. The Treasury fails to take account of retained earnings, and values only dividends and tax equivalent payments. Although similar analyses have been presented in support of previous privatisation proposals, such as the partial sale of Telstra, they have now been generally recognised as erroneous. The Treasury and Bain/Salomon analyses also fail to make an appropriate distinction between regulatory and market risk in the evaluation of publicly owned enterprises.

The failure to deal appropriately with risk allocation is also reflected in an unsound choice of assets for privatisation. It is widely agreed that electricity generation is a more appropriate candidate for privatisation than are the natural monopoly areas of transmission and distribution. In the absence of economies of scope between distribution and retailing the retail component might also be a suitable candidate for privatisation, but this would not yield substantial revenue, and much the same outcome could be achieved by contracting out the relevant functions.

Appendix: Full privatisation of the HEC

As this report was being completed, a proposal for the privatisation of the entire HEC was announced. To undertake a full evaluation of this proposal it would be necessary to undertake a detailed assessment of regulatory and market risk, including the likely impact of interconnection or the construction of new generating capacity. Such an assessment has not been possible in the time available, but in the course of the analysis projections for the entire HEC were prepared on the same basis as those derived above. They are presented here for the information of the Committee. The low and stable growth projections are prepared by applying the same revenue growth assumptions to the generation enterprise as to the transmission and distribution/retail enterprise. The regulated rate of return enterprise is based on a low revenue growth projection for the generation enterprise, combined with the assumption of a regulated 7 per cent rate of return for transmission and distribution/retail enterprises.

 

Table 1a: Projections of revenue and earnings for entire HEC

Low projection

a: Operating expenses excluding depreciation, amortisation and abnormals

b: Earnings before interest, tax and abnormal and extraordinary items

0

Table 2a: Projections of revenue and earnings for entire HEC

Stable projection

a: Operating expenses excluding depreciation, amortisation and abnormals

b: Earnings before interest, tax and abnormal and extraordinary items

 

 

Table 3a: Projections of revenue and earnings for entire HEC

Rate of return regulation projection

a: Operating expenses excluding depreciation, amortisation and abnormals

b Earnings before interest, tax and abnormal and extraordinary items

 

 

Using the approach applied above to the distribution, transmission and retail enterprises, the breakeven price for a sale of the entire HEC would be estimated, on the basis of the low projection, at about $4.4 billion dollars. The stable growth projection yields a breakeven sale price of about $5 billion. The regulated rate of return enterprise yields a breakeven sale price of around $4.6 billion.


Go back to John Quiggin's home page