Ros Eason
The Shadow Minister for Communications, Lindsay Tanner, released his telecommunications policy discussion paper, "Reforming Telstra", on 23rd May. The document has subsequently received considerable attention in the media and provoked sharp responses from sections of the industry. Despite the range of issues identified in the paper, public comment has understandably centred on the possibility that the ALP in government may seek to "structurally separate" Telstra, along either geographic or product lines. Although Tanner is careful to stress that the paper simply canvasses options, the amount of space devoted to a consideration of a wholesale-retail break-up of the company suggests that this scenario is under active consideration within sections of the Party. The Communications Electrical and Plumbing Union (CEPU) believes that the structural separation of Telstra would be bad public policy.
This paper argues that any of the models for structural 'reform' considered in the paper would create inefficiencies and would retard innovation and investment in this critical sector of the economy. Nor do we consider policies that would radically disadvantage Telstra vis-a-vis its domestic and international competitors to be in the national interest. We have strongly recommended that the ALP reject such options.
More broadly, we have argued that the Tanner paper has approached telecommunications policy from the wrong direction. Echoing the complaints of Telstra's competitors, Tanner asserts the key problem facing the industry is the company's "dominance" of the Australian scene. This assertion leads in turn to a focus on how Telstra might be restructured to produce a more competitive market ("Opening up Telstra"). This approach begs a number of questions which we will consider more closely in this paper - on the one hand, the extent to which the natural monopoly characteristics of the industry limit the scope for competition; on the other the degree to which bad business decisions, rather than regulatory failure, may be responsible for any given company's poor performance.
More critically, however, the paper's approach obscures the ongoing tensions between market dynamics and social objectives that constitute the major challenge to telecommunications policy makers today. A more "competitive" market in Tanner's terms (i.e. one where more firms are profitable) will not necessarily deliver lower prices and better services to all sections of the population. Skewed patterns of investment, favouring highly urbanised areas, have characterised competition to date. For competition to take root in the more thinly populated areas of Australia, at least some retail prices (i.e. line rentals) would have to be allowed to rise.
But irrespective of these issues, the level of competition in the Australian market (and especially the number of firms operating successfully within it) cannot be, in itself, the main object of Labor policy. Rather, its aim should be the fostering of an innovative telecommunications industry within a regulatory framework that ensures all Australians equitable access to services, both old and new. Telstra, as the only company capable of acting as a national universal service provider, has a central role to play in the realisation of such an objective. Policy should ensure that it has the technical capabilities, the revenues and the incentives to do so.
Has competition policy failed?
The issue of 'dominance'
The fundamental mistake that the Tanner paper makes is to pose the question of future telecommunications policy in terms of a supposed need to 'reform' Telstra, rather than to review the regulatory regime and, or, the industry as a whole. This approach arises, in turn, from a loosely based set of assertions about Telstra's continuing dominance of the Australian market.
It is true that Telstra has maintained high levels of market share in traditional product areas (access, local calls, domestic long distance) although not in international (where it was deemed "non-dominant" even before full market liberalisation) or in mobiles or in the newer service areas (data transfer, internet access, pay-TV). Why this is so and, just as importantly, whether this puts it is a position to 'dominate' these product markets are other questions. The Tanner paper offers no clear analysis of these issues. As Tanner should know, 'dominance' of a market cannot be established simply by looking at market share. It is more a question of whether a company is able to shape overall market behaviour (e.g. pricing) by virtue of its position. An efficient company may well retain a high market share if it responds effectively to competitive pressures. This does not mean it controls the market - let alone that it is a virtual monopoly, as the paper continually asserts.
In reality, Telstra's power in the market is quite constrained - by competitive pressures in most of its sub-markets and by price regulation at both wholesale and retail levels in key product areas. 1 Telstra's command of the lion's share of industry profits, which the paper also laments, is indisputable. But is this due to an abuse of market power or does it reflect the inevitable logic of market economics? Or the bad commercial judgement of Telstra's competitors? Or both?
Telecommunications remains a high fixed cost industry, where investment is 'lumpy' and time-lines for returns on investment can be long. Rapid technological change such as has occurred in the last two decades exacerbates this problem, by compressing the time available for amortisation of investments and hence increasing risk. It has always been a moot point how many facilities-based companies could succeed under such conditions, especially in a market the size of Australia's. The fact that new entrants are not yet (and may never be) profitable should come as no surprise.
These uncertainties are compounded by regulation. Tanner notes that "local call services in many parts of Australia are still completely dominated by Telstra", a problem he attributes (somewhat circuitously) to the fact that "Telstra is the dominant player across the board". But may it not be that the absence of competition in the local call market reflects the existence of natural monopoly and, or, the impact of legacy pricing structures still held in place by regulation? It is by no means clear how much of the Customer Access Network (CAN) can be economically duplicated, both because new technologies (wireless) are challenging the traditional economics of the 'last mile' local network and because retail access prices do not reflect historic costs. These questions do not arise because of Telstra's 'dominance' of the local call market, although they may explain its basis.
Finally, it should be acknowledged that poor business decisions are to blame for many of the problems now facing the industry, both domestically and abroad. Optus for instance, which has just announced a $402 million loss for its last financial year, still carries the albatross of its pay-TV network around its neck - a legacy from an earlier phase of industry history when it was believed that hybrid fibre-coaxial cable (HFC) would provide the best platform for the transition to 'full service' (i.e. integrated voice and data) networks. Such platforms now appear likely to be superseded by xDSL technologies (high speed data over copper pairs) and, or, by wireless (3G). In the meantime, relatively low levels of pay-TV penetration and the high costs of programming ensure ongoing losses to all operators.
Despite opportunistic attempts by some in the industry to attribute the collapse of One.Tel to high access charges, it is generally acknowledged that that company's demise stemmed from the shortcomings of its operations. Deficiencies in the company's billing procedures were a major contributor to its cash-flow problems. Equally debilitating was the strategy of acting as a loss leader on local calls in order to capture mobile market share - a strategy likely to come to grief in a highly saturated market.
Boom and bust: the world Telco slump More generally, it is obvious that the telecommunications industry worldwide is currently undergoing rationalisation in the wake of the dot.com-telco crash. The 1980s and (even more) the 1990s saw the rapid expansion of telecommunications activity worldwide, spurred by new technological and regulatory openings. This expansive phase of global activity saw:
Increased cross-border activity by the major telecommunications operators;
The creation of international alliances between such companies;
Investment in media and internet assets by telcos;
Massive investment in transmission capacity in both domestic markets and internationally;
The growing role of the WTO and IMF in determining national policy frameworks (regulatory structures, interconnection principles, foreign ownership restrictions).
It is now widely acknowledged that the activities of this period have directly paved the way for the worldwide slump in the telecommunications industry:
International alliances (e.g. the British Telecom/AT&T company concert) have proven to be both unstable and costly.
The high prices paid for both cross-border and domestic acquisitions have weakened the balance sheets of many of the leading European and US companies. (In the case of several European telcos, these weaknesses have been exacerbated by the high prices paid for third generation (3G) mobile licences).
The creation of excess capacity in international transmission (submarine cables) has led, inevitably, to the collapse of several of those companies which sprang up to serve a demand for bandwidth which never materialised (Global Crossing, Level 3).
Over-investment in domestic markets has been followed by the collapse of dozens of new entrants, most notably in the US, where over 60 alternative access providers have disappeared in the last 12 months. Bankruptcies among their counterparts in Europe (e.g. Jazztel, Versatel) and the UK (Energis, NTL) are expected to rise this year.
This crisis in the industry, which has raised the hitherto unthinkable prospect of a major world telco (e.g. AT&T or BT) facing insolvency, has both fed into and been heightened by the wider slowdown of world economic activity. Lower rates of revenue growth, combined with the existence of excess capacity, have led to a sharp cut back in spending in the industry, the effects of which have flowed through to other sectors such as manufacturing. In response to these new conditions, several clear trends have emerged:
A re-consideration of global strategies by major operators. British Telecom, for instance, has been selling off many of its overseas assets in an attempt to reduce debt.
A rising cost of capital for those companies which survive and hence a continuing squeeze on earnings
More insistent calls for regulatory relief by new market entrants, largely in the form of pressure for more favourable terms of access to former monopolists' infrastructure, particularly the local loop.
The downturn in Australia
It is in this context that arguments about the success or failure of telecommunications policy in Australia need to be considered. The Australian majors, notably Telstra, have not incurred the crippling debt levels that now threaten the viability of many of their European counterparts. Nevertheless, the problems identified above can be observed in Australia, especially the creation of excess capacity in certain sections of the market.
During the duopoly period (1991-1997), opportunities for investment were limited by regulation. Full liberalisation has seen the number of licensed carriers expand from three to 89.2 While many of these companies depend largely on existing infrastructure to provide services, a significant number have embarked on network build, buoyed by the expectation of a rapid rise in demand for bandwidth and fuelled by the ready availability of finance.
In 1999, the National Bandwidth Inquiry reported that the potential capacity of installed backbone networks in Australia exceeded usage by between 100 and 100,000 times. Since that time, more fibre optic cable has been rolled out along the east coast and across the continent. Yet, as the Productivity Commission was told during its recent inquiry into competition in the industry, advanced technologies have:
... the capability to enhance the capacity of a single fibre so that it would be capable of efficiently carrying the entire forecast load between Melbourne and Sydney for the next five years.3
Meanwhile, investment in rural and regional Australia has languished. The post-1997 period has seen the appearance of a small number of niche operators which have targeted the larger regional centres (Neighbourhood Cable in Albury and Mildura, Agile in the Coorong), but it is doubtful whether such entrants will ever achieve the scale necessary to sustain their operations. Certainly, neither their activity, nor that of other government-funded regional entrants such as the now defunct Green Phone, offers any plausible alternative to the role of Telstra, on whose ubiquitous network such companies continue to depend. Yet that network is suffering from the effects of long-term under-investment, especially in these same regional areas.
In short, liberalisation has to date offered no answers to the question of how to guarantee socially optimal patterns of investment in telecommunications infrastructure and services. In the meantime, it has produced an industry structure which is highly brittle and which is likely to undergo further rationalisation, especially given the higher cost of capital that its members now face:
Proposals for sharing Pay-TV programming may well be a prelude to a fuller rationalisation of the Telstra and Optus HFC networks, which currently act as a drain on both companies' profitability.
The high costs associated with the transition to third (or fourth) generation mobile networks are likely to prompt further rationalisation in this sector. This may take the form of network sharing or other joint venture activity (as with the current TCNZ-Hutchison agreement) but could also involve one or more of the present operators exiting the Australian market.
Some rationalisation can be expected among the current backbone network providers (Flowcom, Nextgen, Amcom), given the degree of overcapacity in this section of the industry.
A failure of policy? Do these trends, then, point to a failure of competition policy? Certainly, they will disappoint anyone who may have believed that 'managed' (i.e. highly regulated) competition could make a highly capital intensive industry resemble the market for haircuts i.e. one characterised by a goodly number of small to medium sized producers.
In fact, however, Australian telecommunications policy has never been based on the creation of such a market structure. While it has, since 1997, allowed companies to make their own decisions about the amount of infrastructure replication that might be profitable, its centrepiece has been an access regime designed to allow Telstra's competitors use of the national network at regulated rates. Tanner's arguments for the break-up of Telstra are based not only on his (flawed) analysis of Telstra's dominance, but on the contention that this regime has failed. So it is to the operations of the access regime that we now turn.
The access regime As the number of facilities-based competitors to Telstra is always likely to be small, the regulatory regime has, since 1991, made provision for companies to enter the market through access to Telstra's network. As might be expected, both Telstra and its competitors continue to argue about the terms and price of such access. However, neither the existence of such disputes nor the commercial difficulties of many of the current carriers is evidence that the access regime in not working.
Tanner quotes the instance of the long-running dispute between AAPT and Telstra over access pricing as proof of the short-comings of the system and of Telstra's capacity to delay satisfactory access outcomes. "Access delayed", he argues, "is access denied".
But AAPT was not, in fact, denied access to the Telstra network - the company was just not satisfied with the deal it was being offered and held out for something better, while continuing to do business as usual on the basis of interim rates set by the ACCC. Given the disputed nature of these rates, which were set below Telstra's commercial offering, the situation presented few incentives for Telstra to delay proceedings. The matter has since been settled by negotiation. Without judging the rights or wrongs of this particular dispute, it should at least be acknowledged that both side are capable of regulatory gaming.
Tanner is equally naive in quoting recent complaints by the US Trade Representative about Australian access processes. The opening of telecommunications markets world-wide has been accompanied by intense rivalry between nations in the form of demands for ever-easier access to one another's national networks. Former US Trade Representative, Charlene Barshefsky, was tireless in this cause, particularly in relation to Japan whom she repeatedly threatened to take to the WTO over NTT's interconnection rates. More recently, the US has been at loggerheads with Mexico over this same issue. These are the trade wars of the "new economy", in which US positions are no more disinterested than they are in relation to steel or wheat.
That said, it is true that all sections of the industry have at times expressed frustration with the current administrative processes. The relevant issues have been canvassed extensively by the Productivity Commission in its recent report, which contains recommendations for creating a more efficient and equitable regime. However, it should be recognised that it is the highly artificial ('managed') nature of competition in this sector which gives rise to the complexities of the current system and the opportunities for regulatory gaming (by all parties) that it provides.
Access prices will always be a subject of contention. However, it should be noted that the recent Productivity Commission inquiry into telecommunications competition concluded that once the access deficit component of current interconnection rates was set aside, Telstra's actual access prices were at about the mid-point of the international spectrum, while the ACCC's proposed prices were the lowest in the international sample.4 This latter fact may go some way to explaining the protracted dispute between Telstra and the ACCC over access undertakings.
Structural separation: A non-answer to a non-problem? It is Tanner's misreading of the dynamics of the Australian industry and his too cursory analysis of the access regime that leads him to entertain the idea of structurally separating Telstra. The "vertical integration of delivery of infrastructure and services is at the heart of the problems in Australian telecommunications..." he argues before moving to consider options for "opening up" Telstra. Prime among these, in terms of the consideration it is given, is the separation of Telstra along wholesale-retail lines.
Calls for structural separation of incumbent telcos are neither new nor confined to Australia, yet since the dismemberment of AT&T in the early 1980s, which divided the company along long distance vs local service lines, no such radical splitting up of an incumbent has been carried out by a government or regulator.5 This restraint perhaps stems from an awareness of the high administrative and legal costs likely to be associated with such moves, but it also reflects scepticism about the actual benefits of such interventions. Certainly, the passing of the 1996 Communications Act in the US was designed to reverse the thrust of the 1982 order, by allowing the US long-distance carriers into 'local' markets and the Regional Bell Operating Companies (RBOCs) into national long distance.
Recent US experience
Nevertheless, the present downturn in the industry worldwide has stimulated a renewed interest in the structural separation option as a way of reviving the flagging forces of competition and (supposedly) facilitating the more rapid deployment of high bandwidth services. Thus in the US, the crisis of the Competitive Local Exchange Carriers (CLECs), which had built their recent business cases around the roll-out of fast data services (xDSL) over the RBOCs networks, has been accompanied by moves to force structural separation of RBOCs into separate wholesale and retail companies in order to allow readier competitive access to copper 'last mile' loops.
To date, however, no such moves have been successful. In 1999, the Public Utilities Commission of the state of Pennsylvania ordered Verizon to structurally separate its retail and wholesale businesses, but the order was subsequently modified (after a state court challenge by Verizon) to require accounting separation only.6 In March 2001, the PUC acknowledged that the structural separation it had originally supported would involve substantial costs to implement and would require at least as much ongoing regulatory monitoring as the existing access arrangements.7
Similar moves in other US state jurisdictions have also faltered. In March 2001, the Florida Public Service Commission was asked to order the structural separation of Bell South to facilitate competition in local markets. The request was rejected on the basis of the "costs and inefficiencies" it would create, as well as on the grounds that such drastic remedies were premature, given existing access provisions:
Each additional regulation imposed on BellSouth creates costs and inefficiencies; may interfere with other regulations previously imposed; and brings uncertainty to an industry in which stability is necessary to foster competition. Not only is it premature to judge the efficacy of our earlier efforts, but it is also premature to determine that another solution is necessary.8
Crandall and Sidak have specified the nature of the costs that structural separation along wholesale-retail lines is likely to produce.9 These include not only the administrative and regulatory costs involved in establishing and monitoring the new corporate structure, but also the costs created by efficiency losses. Most obviously, vertical separation involves the loss of those economies of scope which exist in multi-product firms such as telecommunications companies. A vertically integrated company may achieve lower cost structures, for instance, by spreading billing costs across a wide range of services. Similarly, it can produce service packages ('bundling') at a lower cost than a firm producing the same services on a stand-alone basis. Crandall and Sidak also argue that vertical integration allows the firm to best coordinate production and investment decisions by minimising external transaction processes and their attendant costs and delays. Such a mode of operation is particularly necessary in an industry operating on the 'technological frontier', where internal processes and structures need to be highly responsive to change.
Vertical integration also allows a closer control over product quality:
A cost of vertical separation is the loss of a single point of accountability. It is difficult for a customer to hold multiple vendors accountable for some form of product failure. Without this single point of accountability, consumers are left 'calling firms' service departments and searching for the party responsible for the failure.10
Competition based on resale has already introduced these problems into telecommunications markets. That is no reason, however, to compound them through further regulatory interventions.
European and UK views
Much has been made in recent industry debate of the OECD paper, Structural Separation of Regulated Industries.11 This paper finds merit in structural separation in industries characterised by natural monopoly elements, and tentatively suggests that in many countries there remains considerable scope for such interventions. As the OECD is forced to acknowledge, however, the only example of mandated structural separation (as opposed to line-of-business restrictions, which the paper loosely includes in its discussion) in telecommunications is the AT&T break-up, a decision which is now in the course of being unwound. In Europe, countries which have examined the option have to date rejected it.12
The OECD reports, for instance, that the Norwegian Parliament voted down a proposal for separation of the Telenor network into a separate company in 1999. Similarly, the UK regulator Oftel has resisted structural remedies to address competition issues arising as a result of vertical integration.
As Oftel explained in an April 2001 report:
some commentators have suggested that a means of addressing ... competition concerns is to prevent the creation of vertically-integrated companies, and thereby forcibly separate content and carriage markets. In some cases, vertical integration enacted through merger and acquisition may be adjudged to be against the public interest. But Oftel believes an all-encompassing prevention of vertical integration would be unjustified, since it may hamper innovation in new services, damage competition across different platforms and hinder UK firms competing in world markets.
Rather than precluding vertical integration altogether, it is more appropriate to address any competition concerns through action by the sectoral regulator...
The potential problems which might emerge when vertically-integrated operators have market power are not new. More importantly, the solutions to such problems are well-established. For instance, BT is subject to obligations relating to the provision of access to its network on non-discriminatory terms. These obligations help prevent market power in one market from being leveraged into another market.13
Despite ongoing pressures from BT's rivals for such separation, a recent House of Commons Committee report simply recommends that Oftel (or its successor OFCOM) "take account" of such propositions.14 The Committee was no doubt wise to handle the proposal (from Cable and Wireless) in a gingerly fashion. The UK experience of structural separation in networked industries has not been a happy one, with higher prices in water, gas and electricity businesses being the norm. The spectacular failure of Railtrack, the UK railways network provider, has cast a particularly deep shadow over these Thatcherite policies, preceded as it was several major rail disasters.
The Railtrack experience highlights the difficulty of ensuring adequate levels of investment in network infrastructure once providers are denied the revenues and consequent incentives that flow from vertical integration. This problem is all the more acute in the telecommunications industry, where technological change is rapid and where investment demands are therefore all the more pressing.
Breaking up is hard to do: technical difficulties Moreover structural separation of telecommunications networks poses particularly intractable problems at a technical level, given the growing complexity of modern systems and the presence of intelligence in different network layers. Where, in evolving networks, do 'wholesale' carriage services end and 'retail' value-added services begin? Where, for instance, would regulation locate such intelligent network services as caller line identification (CLI) and call waiting-forwarding, and where would the future incentives lie for the further development of such network capabilities for a pure utility operator?
Such difficulties have always bedevilled regulatory attempts to draw a neat line between different telecommunications functions. The original AT&T divestiture order, for instance, also required the RBOCs to offer "enhanced services" (i.e. services other than simple voice telephony) through separate subsidiaries. Within two years of divestiture, the requirement had been abandoned, the Federal Communications Commission (FCC) having concluded that:
Structural separation imposes opportunity costs by discouraging the BOCs from designing innovative enhanced services that utilize the resources of the public switched network. Such innovation losses, resulting from the physical, technical and organisational constraints imposed by the structural separation requirements, directly harm the public, which does not realize the benefits of the new offerings.15
The Quiggin model
Finally, the Tanner paper raises again the spectre of a structural separation along the lines recently advocated in the Australian context by John Quiggin.16 Under this model, Telstra would be stripped of its 'potentially competitive' activities, and the revenue raised by such divestiture used to bring the natural monopoly core business back into public ownership. Such a proposal may appear, at first blush, to offer a way out of the difficulties that arise from Telstra's present ownership structure. But as Quiggin himself acknowledges, it is no easy matter, in a period of rapid technological change and increasing global economic integration, to determine what activities might reasonably be regarded as 'core'. Nor, given the continuing price distortions and regulatory constraints that characterise the industry, is it altogether clear where the natural monopoly elements lie.
Quiggin unhesitatingly nominates Telstra's overseas ventures (presumably Reach, Regional Wireless Company, etc.) and its internet (Big Pond) and Pay-TV (Foxtel) activities as ripe for spin off. But what exactly (other than money raising) is the rationale and what would be the impacts of such moves? Reach, for instance (registered in Bermuda!), now holds all the international properties not only of the former Hong Kong Telecom, but also of Telstra, i.e. the submarine cables, leased capacity, international gateways and landing rights of the former monopolist. It is where OTC has gone. It might be argued that there are as many natural monopoly elements here as there are in domestic networks - submarine cable is just more fibre after all - so where is the case for divestiture?
Any sale of these assets would reverse the amalgamation of OTC and the former Telecom undertaken by Labor as part of the 1991 industry restructuring, and leave Telstra obliged to lease capacity from other international operators (SingTel perhaps?) in order to offer international services. In being obliged to do so, Telstra would face obvious disadvantages vis-a vis its major regional and international competitors, none of whom (regardless of ownership structures) face such restrictions.
Nor are constraints on Telstra's other international activities likely to be in the interests of Australian consumers, despite the political appeal of calls for Telstra to 'stick to the knitting' and put domestic investment first. The industry internationally, as discussed above, is witnessing a reaction to the rapid globalisation of the last decade, with the collapse of the mega-alliances, and the selling off of international holdings by some of the more severely wounded majors (e.g. British Telecom, Cable & Wireless). Nevertheless, the trend to increased integration of markets is set to continue, though perhaps more cautiously and with a more regional focus. In Australia, the gap left by the withdrawal of Cable & Wireless from regional markets has been filled by SingTel. Is Telstra to be denied the right to compete with this regional rival outside the Australian market? If so, what would be the implications for its long-term profitability and its ability to fund future domestic infrastructure development?
Similarly, the requirement that Telstra sell off its internet and pay-TV operations, while its major competitors presumably continue to be able to offer (and bundle) such services in the domestic market, is another recipe for declining revenues and technological inertia. Where, for instance, will the incentive come from for Telstra to digitalise its HFC network, and hence pave the way for a workable cable access regime, if it is to be debarred from offering content over this (and presumably any other) platform?
In the case of mobiles Quiggin is more tentative, but concludes it should "probably be spun off". Again, the financial and technological consequences of such a split would be overwhelmingly negative for Telstra, since:
mobiles is still an engine of revenue growth, despite approaching market saturation;
all Telstra's major competitors offer mobile products which they can bundle with other telephony and data offerings; divestiture of Telstra mobiles, without any equivalent constraints on its competitors, would again spell commercial disaster for the company;
technology trends are leading towards the fuller integration of mobile and fixed network products.
Quiggin rightly rejects models of structural separation based on geographic-product (the AT&T breakup) and retail-wholesale lines. Nevertheless, his alternative model turns the clock back some 27 years. Under these proposals, the new Telstra would resemble the Telecom of 1975 - a public utility offering standard domestic telephony services, debarred from overseas expansion and from opportunities to explore and exploit the possibilities presented by technological convergence. This is a recipe for commercial stagnation for which the public, both as customers and owners of the company, would pay dearly.
A national carrier
Telstra and universal service
For it must be remembered that Telstra still carries the ultimate responsibility for provision of both standard telephony and digital data services to the community as a whole. Labor policy currently calls for the extension of this universal service obligation to guarantee affordable access to "modern" communications services, including broadband data services. This in turn implies a major investment programme (as well as raising questions as to how the costs of such a programme are to be spread between, say, government, consumers and the industry).
Any thoughts that market forces will provide an answer to the problem of providing such services throughout the country has surely been dispelled by the experience of the last decade. The Telstra-Optus roll-out of rival HFC networks demonstrated the logic of the market in 1994/5 - duplication in the major population centres, nothing in the hinterland. More recently, the government's attempts to establish a competitive framework for the supply of USO services have ended in failure. These experiments have essentially resulted in Telstra being confirmed as national carrier of last resort, it being the only company with the technological capability and deep enough pockets to fulfill the role.
That said, it must be recognised that a decade of policies of market liberalisation and, subsequently, privatisation have seriously compromised Telstra's capacity to perform the role it did under the period of monopoly public ownership. The consequences of these policies reveal themselves most obviously in the current state of the copper Customer Access Network (although the precipitous decline in R&D spending, now a mere 0.1 per cent of earnings, is equally ominous).
The Australian Communications Authority investigation following the Boulding death in northern Victoria confirmed what the CEPU has long claimed - a high degree of unreliability in the Customer Access Network (CAN) as a result of years of underinvestment, especially in regional and rural areas. In the Kergunyah exchange area in which the Boulding home was located:
On average there were 1.18 faults per service in operation (SIO) over the 13 months to the end of February 2002;
Customers that reported a fault reported 1.95 faults on average over the period; and
One customer reported nine faults.17
More startlingly, the ACA reports that these statistics, though high, are not entirely out of kilter with national statistics for similar regions. The report gives a picture of a network in disrepair, focusing as it does on the impacts that a combination of cable degradation and (in this case) use of digital pair gain systems has had on network reliability. These conditions are replicated across much of rural and regional Australia. Yet it is this same network which could potentially provide the most appropriate platform for the delivery of fast data services (through DSL technologies) on a national basis.
Broadband services and the CAN
The CEPU believes that, given the excesses of the last few years, policy makers should approach the question of broadband deployment with some caution. The collapse of so many business plans based on a vision of a bandwidth-hungry globe has as much or more to do with lack of demand as it has with supply side constraints (i.e. access to unbundled local loops). With the exception of sections of the small business community and public institutions (hospitals, schools, libraries), it is by no means clear that there is currently a significant market for high speed data transfer beyond the corporate sector.
Nevertheless, the same measures which would address the more modest data needs of the community may also help pave the way for wider availability and take-up of broadband. At present it remains the case that significant numbers of Australians cannot access the Internet at effective speeds, at least without having to take a premium service like ISDN. For instance, the Australian Communications Authority reported in 1998 that of
The 1.1 million rural and remote customers connected to the PSTN by copper wire, 55 per cent .. are unable to achieve network data rates of 14.4 kbit/s and 70 per cent are unable to achieve 28.8kbit/s.
This means that the range of application available to the majority of rural and remote households via their standard telephone service is limited and the quality of the data service available is often poor.18
The Inquiry also reported that Telstra estimated the cost of upgrading the CAN nationally to support a data rate of 28.8 kbits/s to be of the order of $4 billion.
As subsequent studies have noted, capacity constraints affecting data transfer rates may reside in areas other than the CAN - with ISPs or with customer equipment, for instance. Moreover, some areas may never be effectively served through terrestrial technologies. Nevertheless, the CEPU believes that the time is ripe for an investigation into the feasibility of a national CAN upgrade, with a view to:
Remedying the deficiencies identified in the ACA's Boulding inquiry;
Ensuring universal availability of narrowband data services at speeds of at least 28.8 kbits/s within the pricing framework of the standard service; and
Laying the foundations for the wider deployment of broadband services.
The CEPU put a similar proposal to the ALP National Industry, Infrastructure and Regional Development Policy Committee in 1999. We believe that the practicality of such a programme could now be tested in the course of a Senate Inquiry into the state of the Australian telecommunications network, which would serve the dual purpose of addressing this policy issue and focusing public attention on present network deficiencies in the run up to the T3 debate.
Such an inquiry should also investigate the incentives for and obstacles to investment that exist under the current access regime. The Productivity Commission has recognised the arguments in favour of "access holidays" for investments involving the creation of sunk assets in a climate of risk. It is likely that some such provisions will be made in relation to the digitalisation of the existing HFC (cable) networks.
But the CEPU believes that a wider discussion of the impacts of the current access pricing principles is also warranted. In the US, there is now a strong body of professional opinion critical of the unbundling requirements and accompanying pricing regulation that has been imposed on the incumbent network owners. This is increasingly being seen as a barrier to the investments needed to create broadband networks.
In the words of the most senior US economist working in this area, Professor Alfred E. Kahn: ... the FCC has imposed ... a pricing scheme ... that is manifestly hostile to the large, risky investments required for genuine competition in these markets.
Why should any new competitor assume such risks, in an atmosphere of both competitive and technological uncertainty, when regulators allow it to free ride on the investments of the phone companies at rates intended to recover only the costs of an ideally efficient new entrant writing on a clean slate? And why would an incumbent firm spend billions of dollars building a new broadband network, when confronted with the head-competitors-win, tails-we-lose prospect of having to make its successful investments available to rivals at rates below their actual costs, while alone absorbing the costs of investments that fail?19
These questions are equally pertinent to the Australian access regime. In the CEPU's view, ALP policy will need to recognise their centrality if it is to lead the country into a broadband future.
Ros Eason is a Senior Industrial officer with the Communications, Electrical, and Plumbing Union of Australia (CEPU). She serves on several industry bodies, including the Australian Communications Industry Forum and the Australian Communication's Cabling Advisory Committee, and is a contributor to the Evatt Foundation's new book, Globalisation: Australian Impacts (UNSW Press). This paper was prepared in response to the discussion paper, "Reforming Telstra, by the Shadow Minister for Communications, Lindsay Tanner. Lindsay's paper may be read on his personal site.
Notes 1. It might indeed be argued that such regulation (i.e. retail price caps) presents as much of an obstacle to the profitability of Telstra's competitors as does Telstra's purported dominance of the market. The recent relaxing of the overall retail cap from (CPI-5.5 to CPI-4.5) appears designed to give not only Telstra but the rest of the industry some breathing space in these troubled times. 2. The 100th carrier licence was issued with some fanfare in May this year, but 11 of these are no longer operative. 3. Productivity Commission, Telecommunications Competition Regulation: Inquiry Report, Canberra, 2001, p. 91. 4. Ibid., p. 394 5. It should be noted that there is no correspondence between the regulatory meaning of 'local' in the US and Australian contexts. The original Regional Bell Operating Companies created by divestiture served areas larger (in terms of subscriber numbers) than the entire Australian market. These 'local' markets have, since the introduction of the 1996 Act, been enlarged through RBOC mergers. 6. Verizon, a 'local' US carrier, was formed through the merger of Bell Atlantic (which itself had merged with fellow RBOC Nynex) and long distance provider GTE. Verizon boasts a customer base of some 134m access lines and 29m mobile customers. 7. For a fuller discussion of this case, see Crandall, R. and Sidak, J. Gregory, "Is Structural Separation of Incumbent Local Exchange Carriers Necessary for Competition?", Yale Journal of Regulation, 19:2, 2002. 8. Order Granting BellSouth's Motion to Dismiss AT&T's and FCCA's Petitions for Structural Separation, Dkt No.010343-TP, Florida Public Service Commission, Nov 6, 2001 (downloaded from http://www.psc.state.fl.us/dockets/documents/01 on 6/11/02) 9. Crandall & Sidak, op. cit. p. 30ff. 10. Alchian, A. "Vertical Integration and Regulation in the Telephone Industry", 16 Managerial and Decision Economics 323, 323-26 (1995), quoted in Crandall and Sidak, op. cit. p. 33. 11. Oecd, Structural Separation in Regulated Industries, OECD, Paris, 2001. 12. A variant of structural separation was imposed on the Japanese incumbent NTT, with its domestic operations being split between two 'local' companies and its mobiles division established as an independent entity. All companies remained, however, under the umbrella of a single holding company. 13. Oftel, Open Access: Delivering effective competition in communications markets, April 2001, at 4.13-4.15 14. The Committee's Report can be found at www.publications.parliament.uk/pa/cm200102/cmselect/cmcumeds/539/53903-h.... 15. Quoted in Crandall and Sidak, op.cit. p. 51 16. Quiggin, J. "Stick to a few principles", Australian Financial Review, 9 May 2001. 17. Australian Communications Authority, Investigation into the provision and maintenance of telephone services to the Boulding family in Kergunyah, north-eastern Victoria, March 2002, pp. 16-17 18. Australian Communications Authority, Digital Data Inquiry Report, 1998, p. 55. 19. Alfred E. Kahn, "Unleash the Broadband Economy", Policy Matters, 01-35, AEI Brookings Joint Center for Regulatory Studies, December 2001. Downloaded from http://www.aei.brookings.org 6.11.02