Go back to John Quiggin's home page
This version 20 June 1996
The partial privatisation of Telstra: an assessment
Submission to Senate Environment, Recreation, Communications and the Arts Reference Committe inquiry into the Telstra (Dilution of Public Ownership) Bill 1996
Professor of Economics
James Cook University
FAX + 61 77 814149
Phone + 61 77 81 4798
+ 61 77 251269
The partial privatisation of Telstra: an assessment
In recent years, many public assets in Australia have been privatised and there have been proposals for more extensive privatisation. As the largest and most profitable enterprise owned by the Commonwealth government, Telstra could be sold for a substantial sum. It is not surprising, then, that Telstra has been a prominent candidate for privatisation. However, previous analyses, such as that of Brown (1995) have concluded that privatisation would reduce the net worth of the public sector and would therefore represent a bad deal for taxpayers.
A number of new issues are raised by the Telstra (Dilution of Ownership) Bill. In particular, the Bill proposes the sale of a minority of shares in Telstra in a context where the remaining shares may or may not be sold in the future. The issue is also complicated by the proposed shift to more open competition in telecommunications from 1997 onwards.
The object of this paper is to assess the desirability of partially or wholly privatising Telstra. The primary focus is on the fiscal costs and benefits of privatisation, that is, on whether taxpayers will be better or worse off in the long run. It is shown that privatisation will reduce the net worth of taxpayers. On the preferred scenario of moderate future profit growth, taxpayers will lose around $30 billion, in present value terms, as a result of privatisation. Subsequent sections of the paper examine specific implications of partial privatisation, restrictions on foreign purchases of shares in Telstra and the previous proposal for the privatisation of Telstra put forward in Fightback.
2. Prospects for Telstra profits
Telstra and its predecessors, Telecom Australia and OTC, have been consistently profitable. Assessment of past profits requires adjustment to take account of inflation and of the fact that Telecom Australia was not subject to company income tax until 1990-91. Profits since 1988-89 are given in Table 1.
Table 1 - Telstra net profits ($m 1995-96)
Telecom OTC Telstra Telstra
(current prices) (current prices) (current prices) 1995-96 prices
1988-89 973* 189 772.8 1008
1989-90 1288* 233 1005.8 1216
1990-91 963 275 1238 1401
1991-92 313 349
1992-93 904 1005
1993-94 1699 1853
1994-95 1755 1867
Source: Telstra Annual Reports
* - Not taxed
Gross profits in 1991-2 were reduced by $1.2 in extraordinary items associated with the Telecom-OTC merger and large scale redundancies, and some abnormal losses were also incurred in 1992-93. Disregarding 1991-92 it can be seen that real after-tax profits were consistently in excess of $1 billion per year even before the extensive reform associated with corporatisation. Profits have been on a steadily increasing trend.
Assuming that changes in regulation from 1997 onward do not worsen Telstra's position, the outlook for real profit growth appears favorable. Although a gradual loss of market share could be expected on the basis of the existing regulatory setup, it seems unlikely that the share of rival carriers would exceed the current Optus target of around 25 per cent, leaving Telstra with 75 per cent of the market. The history of telecommunications markets in other countries lends support to this view. Under a variety of regulatory regimes, firms which held a monopoly prior to competition have retained market shares well over 50 per cent after the introduction of competion.
A relatively small loss of market share will be offset by growth in the market. Continuing technological progress would enable maintenance of the well-established trend of reductions in unit prices. In view of the high long-run price elasticity of demand for telecommunications services, the continuing growth in population and per capita income, and the introduction of new products, sustained growth in revenues could be achieved while maintaining existing profit margins. Profits can therefore be expected to rise steadily in real terms.
In the analysis below, three scenarios will be considered. In the central scenario, profits will grow in line with real GDP, that is, at a real rate of 3.5 per cent per year for ten years and stabilise thereafter. In the high scenario, profits will grow at a real rate of 20 per cent for five years and stabilise thereafter. In the low scenario, profits will decline by 25 per cent in the first year and stabilise thereafter.
The profitability of Telstra may be either greatly enhanced, or greatly reduced, by the decisions made on the structure of the telecommunications industry after 1997. The outcome most favorable for Telstra would be a 'light-handed' regulatory model along the lines of that adopted in New Zealand. This would probably result in Telstra gaining market share on the basis of the economies of scale, scope and network density associated with its operations.
The least favorable outcome would be one in which Telstra was regarded as a dominant enterprise and subjected to discriminatory requirements to grant its competitors access to its network at low (for example, marginal cost) prices set by regulation. However, the loss of profits would presumably be moderate, since any substantial loss of market share would undermine the basis for a finding of market dominance.
From the viewpoint of society as a whole, the net effects of changes in regulation will be relatively small. Regulatory outcomes that deliver benefits to Australians in their capacity as consumers, in the form of lower prices, will involve offsetting losses to Australians in their capacity as taxpayers, through reduced profitability for Telstra. Conversely, the costs to consumers of higher prices will be largely offset by higher profits for Telstra.
Hence, in assessing the value of Telstra in continued public ownership, it is not necessary to take detailed account of regulatory risk. Rather it is sufficient to value Telstra on the assumption that regulatory changes will have, on average, no effect on profitability. Should it turn out that the optimal regulatory policies are adverse to Telstra, the loss of profits will be more than offset by consumer benefits.
From the viewpoint of owners in a partially or wholly privatised Telstra, however, the risk associated with regulatory change implies a significant reduction in the value of their asset. First, the increase in risk per se implies an increase in the premium demanded by holders of private equity, and hence in the asset value associated with a given expected value of profits. Second, once Telstra is privatised, governments will have an incentive to make regulatory decisions which restrict Telstra's natural dominance of the market and favor the interests of its competitors.
Principles of valuation for publicly owned assets
As has been argued in more detail by Quiggin (1995) and Brown (1996), the appropriate valuation procedure for a publicly-owned asset such as Telstra is based on the saving in public debt interest represented by the flow of profits of the enterprise. The asset value is equal to the present value of the flow of post-tax real profits discounted at the real rate of return on public debt. That is, the valuation is given by the amount of debt that could be sustainably serviced on the basis of the flow of profits of the enterprise.
A number of misconceptions need to be clarified here. The first is the idea, implied by the treatment of asset sales as negative outlays in the Budget papers, that the proceeds of the sale of assets such as Telstra are equivalent to revenue raised from taxation. In fact, asset sales involve the loss of a stream of future income, and the proceeds should not be treated as current income. The present Treasurer correctly criticised the previous government for using asset sales to conceal Budget deficits and has announced an intention to focus on the underlying Budget deficit, which excludes the proceeds of asset sales and repayments of State and government business enterprise debt. Thus, it is the inappropriateness of treating the proceeds of asset sales as revenue has been clearly recognised. Unfortunately, the issue has been muddied by the announcement of an environmental spending package of $1 billion, contingent on the partial sale of Telstra. Since the sale or retention of Telstra will have no effect on the underlying Budget deficit, the question of whether the proposed environmental package is desirable and affordable is independent of the sale of Telstra
Next, many analysts focus attention on the flow of dividends remitted to the Budget sector, rather than on the flow of profits. That is, these analysts disregard retained earnings reinvested in the enterprise. This is an error in the evaluation of either a private or a public enterprise. A basic result in the theory of finance is the Modigliani-Miller theorem which states that dividend policy is irrelevant to the value of an enterprise. This result is true for both private and public enterprises.
Finally, it is frequently assumed that the value of an enterprise in government ownership must be equal to its value in private ownership. This is not true for two main reasons. First, there is the issue of regulatory risk discussed above. This reduces the value of an asset such as Telstra to private owners, but not to the public. Second, there is the problem of the 'equity premium'. 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'.The equity premium puzzle may be explained by the observation 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).
Computation of the real interest rate
Over the last hundred years, real interest rates in developed countries have averaged around 1 per cent. However, it seems likely that this low real rate is due in part to unanticipated inflation and that the equilibrium value of the real interest rate is between 3 and 4 per cent. If it is assumed that the likely rate of inflation in the United States over the next ten years is around 2 per cent, the prevailing US 10-year bond rate of 6 per cent implies a real interest rate of 4 per cent. Interest rates in Australia are somewhat higher. This is presumably due in part to the Reserve Bank's choice of a target inflation rate around 3 per cent and in part to the judgement that there is some risk of a return to higher rates of inflation. In this paper, a real interest rate of 4 per cent will be used as the basis of analysis. For the purpose of sensitivity analysis, rates of 2 per cent and 6 per cent will be considered.
The valuation of Telstra in public ownership
In this section, the principles set out above will be applied to derive a range of values for Telstra in public ownership. Three scenarios will be considered. In Scenario 1, profits grow in line with real GDP, that is, at a real rate of 3.5 per cent per year, until 2001, and stabilise thereafter. In Scenario 2, profits grow at a real rate of 20 per cent until 2001, and stabilise thereafter. In the low scenario, real profits decline by 25 per cent in 1996 and stabilise thereafter. The resulting profit streams are presented in Table 2.
In Table 3, the streams of profits under the three scenarios are valued using four different discount rates. The preferred rate, based on the analysis presented above, is 4 per cent. For the purpose of sensitivity analysis, rates of 2 per cent and 6 per cent are considered.
Table 2 Profit scenarios for Telstra
Table 3 Valuation of Telstra in public ownership
The preferred estimate for the valuation of Telstra in public ownership is $54.4 billion, obtained by applying a 4 per cent real discount rate to the stream of profits in Scenario 1. This is approximately twice the value envisaged in most discussion of the possible sale of Telstra. Assuming a realised sale price of Telstra of $24 billion, privatisation would reduce the net worth of taxpayers by around $30 billion.
Under Scenario 2, where the past rate of profit growth is sustained for a further five years, the value of Telstra in public ownership rises to $107.9 billion. This is approximately equal to the value of all outstanding Commonwealth government debt. That is, at current interest and inflation rates, the profits of Telstra under Scenario 2 would cover the entire real interest on Commonwealth government debt. The remaining nominal component of interest could be rolled over while keeping the real value of public debt constant and allowing the ratio of public debt to GDP to decline steadily. Thus, the debt burden on taxpayers would effectively be eliminated, since no contribution to debt service would be required from general revenue. There is no guarantee that Telstra's past profit growth will continue as envisaged in Scenario 2. But by undertaking privatisation now, the government guarantees that taxpayers will bear a substantial burden of public debt for years to come.
In the least favorable Scenario 3, Telstra suffers a substantial loss of market share and profitability as a result of competition. Even in this scenario, however, the stream of earnings generated by Telstra is greater than the saving that would arise by privatising Telstra and using the proceeds to repay debt.
Variations in the real interest rate used in the analysis within the range from 2 to 6 per cent do not change the basic conclusion. For all three scenarios and all three discount rates, the value of Telstra in public ownership is greater than the sale price envisaged by most analysts. The costs of regulatory uncertainty and the premium rate of return demanded by holders of private equity outweigh any efficiency gains that might be expected under private ownership.
If the new private owners of an enterprise can introduce substantial efficiency improvements, the increase in the flow of profits may offset the higher rate of return demanded by private equity holders. If this is so, the public will benefit from privatisation. But in the case of a partial sale, as proposed in the Telstra (Dilution of Ownership) Bill, there is no change of management and hence no possibility of efficiency improvements beyond those that would have taken place anyway. The public suffers the loss associated with the equity premium but gets no efficiency benefit.
The best option, therefore, is either to retain the asset or sell it in one go. If it is believed that equity markets are too thin to absorb the asset in one go, the best way of selling is to commit in advance to a sale staged over several years. However, where equity markets are thin the equity premium is likely to be larger than usual, and the case for privatisation correspondingly weaker.
The position of minority shareholders in a publicly owned enterprise of the kind proposed by the government is such as to guarantee a low sale price. Should the Liberals lose the next election, the shareholders would be at the mercy of the incoming Labor government. If that government should be genuinely hostile to privatisation, the minority shareholders would be unlikely to make large returns on their investment. In these circumstances, it would be a foolish investor who offered the same price for shares in a partly privatised Telstra as they would offer in the case of a full privatisation.
The Telstra (Dilution of Ownership) Bill includes numerous restrictions on shareholdings in Telstra, particularly applying to foreign shareholders. One effect of any restriction on bidding must be to reduce the price at which assets are sold. The exclusion of foreign buyers also undermines many of the arguments commonly put in favor of privatisation. First, privatisation is commonly seen as putting incumbent management under the market discipline arising from the threat of takeover. The only firms that could reasonably be expected to make takeover offers for Telstra are foreign telecommunications firms. Second, privatisation is claimed to permit the introduction of private sector expertise. Once again restrictions on foreign ownership preclude the possibility that the new owners will have any telecommunications expertise beyond that already available to Telstra. Finally, privatisation is based on the claim that direct public control of an enterprise is unnecessary, since a combination of general pro-competitive regulation and competitive market discipline will ensure that the enterprise acts in the public interest. But the admission that a foreign-owned Telstra might behave in a way that is inconsistent with the public interest shows that this claim is false. If Telstra is too important to be sold to foreign telecommunication companies, it is certainly too important to be entrusted to entrepeneurs like Alan Bond or Christopher Skase. But there is nothing in the proposed privatisation that would prevent a new generation of entrepreneurs from gaining a substantial shareholding or, in the event of complete privatisation, a controlling interest.
Based on past experience, it seems unlikely that restrictions on foreign ownership will ultimately be effective. The effect of the 'safeguards' in the Telstra (Dilution of Ownership) Bill will be to reduce the sale price obtained by taxpayers while obscuring the fact that the ultimate outcome of privatisation will probably be either a foreign-controlled monopoly in telecommunications or a duopoly consisting of two foreign-owned firms.
Previous proposals for privatisation
The Fightback! manifesto, released in late 1991 proposed the sale of Telstra and up to 600 other Commonwealth assets with an estimated total sale price of $20 billion (in 1991-92 values) over four years. No specific value was given for Telstra. However, the first year program (assumed to apply from 1992-93) included the sale of remaining shares in Qantas and the Commonwealth Bank, along with the AIDC, Snowy Mountains Engineering Corporation and Commonwealth Serum Laboratories, with estimated returns of $5 billion. The AOTC (now Telstra) was to be sold in three tranches, commencing in the second year of the program. It follows that the total market value envisaged for Telstra in 1991, assuming sale over the period form 1993-4 to 1995-96, was no more than $15 billion (about $17 billion in current values), even if no value is placed on the hundreds of remaining assets in the government portfolio.
The current estimated market price for a sale of Telstra commencing in 1997 is around $24 billion. Hence, even if Telstra is sold, the delay of four to five years in the sale process has meant that taxpayers have gained around $7 billion in real sale proceeds as well as more than $3 billion in dividends paid out since 1991-92. Against this, the real interest savings assuming the proceeds of sale had been used to repay debt would have been no more than $5 billion. Thus, the delay of four to five years has yielded benefits of at least $5 billion to taxpayers.
The same observation can be made about a number of previous privatisations. For example, the first proposal for sale of the Commonwealth Bank was made by Coghlan who estimated a market value in 1985 of about $1.6 billion (or $2.6 billion in 1995/96). Assuming that the final 50.4 per cent shareholding sells for around $10 per share, or $4 billion, the delayed sale of the Commonwealth Bank will yield taxpayers around $7 billion, a gain of $4.5 billion. In addition, the flow of dividends from the Bank to the Budget sector has been considerably greater than the saving in public debt interest that would have been realised if the sale proceeds had been used to repay debt in 1985.
The benefits accruing to the taxpayer from delays in privatisation simply reflects the fact that the return to these assets exceeds the opportunity cost of funds. When the asset is sold, the government effectively cashes in the reinvested component of profits. But, if a short delay in privatisation is beneficial to taxpayers, a long delay is even more so, and permanent retention of the asset in public ownership better still.
Privatisation and the 1997 reforms
The impact of the 1997 reforms on the revenue likely to be realised from privatisation has been discussed above. It is also important to consider how the likely privatisation of Telstra will affect the process of policy reform. The main impact of privatisation will be to complete the transformation of Telstra from a democratically accountable organisation with an objective of meeting socially determined needs for telecommunications services to an enterprise concerned with maximising profits subject only to external regulatory constraints. This transformation began with corporatisation, which required Telstra to focus primarily on commercial objectives. However, with complete public ownership Telstra could still be required to take account of social objectives and of the needs of national development. Once Telstra has been partially or wholly privatised, its directors will have a fiduciary obligation to put the interests of shareholders ahead of all other considerations.
Under unrestricted competition or 'light-handed regulation' of the type adopted in New Zealand, Telstra would secure a monopoly or near-monopoly position fairly easily. Following the New Zealand example, a privatised Telstra could greatly increase profits by raising access charges. Dominance in the output market would also give Telstra a strong bargaining position in the labour market, particularly in relation to technical staff who would have nowhere else to go.
Assuming such an outcome is considered undesirable, regulators will be forced to give favorable treatment to Telstra's competitors, even if the result is to reduce the efficiency of the industry as a whole. It will also be necessary to maintain price caps on Telstra while allowing much greater freedom to other firms, including direct competitors and value-added resellers.
Such a policy framework is likely to exacerbate existing contradictions. For example, since Telstra will be more dominant in some markets than in others, its profits can be increased by the practice of price discrimination, that is, setting prices well above marginal cost in markets where it is dominant while setting prices at or near marginal cost in markets where it faces competition. The various flexiplans and special pricing packages offered by Telstra since the introduction of competition may be interpreted as a step towards this kind of price discrimination, and some proposed packages have been rejected as anti-competitive by the regulatory body Austel. However, the same price outcome can arise if Telstra sells capacity at marginal cost to 'value-added' resellers who simply resell the capacity in highly competitive markets. Regulators will be faced with a conflict between the desire to encourage such resellers and their potential use as agents of price discrimination.
There is no grounds for a belief that a combination of market competition and external regulation will result in the adoption by a privatised Telstra of policies that serve the best interests of telecommunications consumers or of Australian society in general. Public ownership, even in the attenuated form of corporatisation, provides an element of democratic accountability that is vitally necessary.
Privatisation has been a popular policy in the last decade, particularly in the English-speaking countries, but there is little evidence that it produces sustained economic benefits once the illusory short-term effect of reductions in the reported Budget deficit is exhausted. In the United Kingdom, which led the world in privatisation, the much-lauded 'Thatcher miracle' collapsed in the recession of 1989-91 and the government, having sold off nearly all the assets at its disposal, found itself in severe budgetary difficulties by 1993. The pathbreaking privatisation of British Telecom is now widely recognised as a textbook example of how not to go about privatisation.
The similarly overpraised New Zealand 'miracle' appears to have stalled in the past year, with GDP growth rates falling behind those of Australia and unemployment rising. Over the past decade, GDP per capita in New Zealand has fallen by around 15 per cent relative to Australia.
The analysis presented above, along with analyses of recent privatisations such as that of the Commonwealth Bank (Quiggin 1995) and Commonwealth Serum Laboratories (Hamilton and Quiggin 1995) demonstrates that privatisation is a bad deal for Australian taxpayers. In the face of such evidence the advocates of privatisation offer nothing more than dogmatic assertions of faith in the market and appeals to fashion, based on the claim that since other countries are privatising, Australia should do likewise. This view should be rejected. Telstra should be retained permanently in public ownership.
Quiggin, J. (1995), 'Does privatisation pay ?', Australian Economic Review 95(2), 23-42.
Hamilton, C. and Quiggin, J. (1995), 'The privatisation of CSL', Australia Institute, Canberra.
Go back to John Quiggin's home page