Save Search

Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
ENVIRONMENT, COMMUNICATIONS, INFORMATION TECHNOLOGY AND THE ARTS REFERENCES COMMITTEE - 12/11/2003 - Competition in broadband services

CHAIR —Welcome. Thank you for your time today; it is much appreciated. I think you are all old hands at committee appearances so I will skip the usual formal statement to witnesses. Do you want to make any alterations or corrections to your written submission at this stage?

Mr Scales —No, we do not.

CHAIR —I now invite one of you to make an opening statement before we move to questions.

Mr Scales —I would like to thank the committee for giving Telstra the opportunity to appear before you today. Telstra is committed to encouraging Australian businesses and consumers to experience the benefits of high-speed Internet access. Telstra's future growth potential really does depend to a large extent on broadband services, so we have every reason to share the Broadband Advisory Group's vision of Australia becoming a world leader in the effective use of broadband technologies and we see broadband as a key growth engine for our company. Indeed, Telstra is confident that the outlook for broadband in Australia is extremely encouraging.

Telstra believes that the essentials are now in place for Australia to take its place as a world leader in the productive use of broadband. According to the McKinsey quarterly report No. 2 of 2003, which was titled Making sense of broadband, Australians already have better access to broadband technologies than almost any other consumers in the world, with all Australians being able to access one or more forms of broadband at prices that are comparable with international averages. Telstra's initiatives, such as the ADSL demand register, will continue to expand the range of broadband options available to Australian consumers.

Telstra would now like to make a comment on some recent prominent studies that have focused on broadband take-up and price since our submission was lodged with the committee. As you know, it is sometimes suggested that the price of broadband in Australia has inhibited its take-up when compared to the take-up in other countries. A recent A.T. Kearney report examined broadband prices in Australia and found that Australian prices are at par or near par with prices internationally. Significantly, according to A.T. Kearney, entry-level DSL prices in Australia are cheaper than in both the United Kingdom and the United States. Telstra is aware of commentary based on a report by the International Telecommunications Union that Australian broadband prices are expensive by international comparisons. Telstra is currently seeking to clarify this finding with the ITU as the report seems to be based on an assumption that entry-level broadband packages in Australia are $US91.77 per month, a figure that seems to be quite significantly higher than we understand prices in Australia and that is particularly significantly higher than Telstra's entry-level ADSL price of $A59.95 per month once connected.

Last month has delivered a number of favourable reports around broadband in Australia. A recent comprehensive survey of the Australian manufacturing industry by the Australian Industry Group showed that use of broadband amongst Australian businesses is significantly higher than residential use. The survey found that 58 per cent of the businesses surveyed had fast Internet connections—a figure that is expected to increase to 72 per cent in the next 12 months. Thirty-eight per cent of the businesses surveyed by the Australian Industry Group used what the survey referred to as second-generation broadband. Importantly, the results showed no difference in the rate of broadband usage between metropolitan and regional businesses.

There has also been good news about broadband growth in Australia more generally, with the ACCC's most recent snapshot of broadband deployment showing that broadband penetration in Australia doubled in the 12 months to June 2003. The number of DSL connections increased by an impressive 166 per cent. All of these reports support our view that broadband penetration will continue to rise rapidly as more consumers and business owners see a convincing value proposition from broadband. Interestingly, the most frequently cited reason given in the Australian Industry Group's survey for businesses choosing not to upgrade to broadband was not price or lack of availability but a perception that it was not needed or would not be of benefit to business performance.

Telstra believes that industry and governments must work together to continue to convince the Australian community of the significant value of broadband technologies. Telstra's roll-out of ADSL technology has been one of our nation's biggest and most expensive engineering projects, and it has been delivered in a very short period. I am sure we will come to that in the discussion. Telstra will continue to invest in enabling exchanges in circumstances where there is a reasonable basis for recouping the costs of the significant investment involved. Generating demand for broadband services will encourage further industry investment in broadband infrastructure. Telstra supports the view of the Broadband Advisory Group, which recommended that governments consider funding initiatives, investment initiatives and public-private partnerships to develop services in areas where the deployment of broadband is not supported by sufficient demand to make its deployment financially viable for providers, including Telstra.

Although Telstra is optimistic about the future of broadband in Australia, we do recognise that there is still much work to be done to increase the broadband options for some Australian consumers. Telstra continues to work on strategies for improving the affordability of two-way satellite for people living in remote and some regional areas. In addition, Telstra has introduced a series of initiatives to optimise broadband access for customers whose telephone service is supplied using pair gain technology.

In conclusion, Telstra has set ambitious targets for broadband. We aim to have one million broadband customers by the end of 2005 and $1 billion in broadband revenue by the end of 2006. I am pleased to say that we are on track to reach both of these quite aggressive targets. Telstra will continue to work with local communities and governments at all levels to ensure that all Australians are introduced to the benefits of the broadband technologies available to them. Broadband penetration rates will continue to rise as we move from an earlier adoptive phase into a more mature mass business and consumer market. Senators, thank you very much for listening and for giving me the opportunity to make these introductory comments.

Senator LUNDY —I would like to go to the issue of the ACCC's report on emerging markets, and the view expressed in the report that there is a declining state of competition in the broadband area and that one of the factors in that decline is the interest Telstra has in two significant forms of access to broadband services—namely, the HFC cable and the copper access network. It seems that Telstra has used the Ergas report to try to respond to those findings. I want to ask Telstra specifically to explain the formula behind the Ergas report that attempts to counter those findings and, in particular, challenge some of the substantial research that was done by the OECD, the ITU and others, and get to the nitty-gritty of the formula used in the NECG report.

Mr Scales —And your question?

Senator LUNDY —Can you explain to me the formula used in the NECG report that comes to the quite extraordinary conclusion that, if Telstra were to divest itself of ownership of the HFC cable, in fact competition would decline—which I find to be quite an extraordinary statement. Can you take me through the formula that allows that conclusion to be reached?

Mr Scales —I might ask Dr Warren to go through that. I think it might not be completely accurate to describe it as a formula.

Senator LUNDY —In the attachment to the report, it looks like an algebraic equation to me.

Mr Scales —It is modelling. I am happy to go through the details with you of how modelling is done, if you wish. As you know, modelling brings together a series of simultaneous equations, with a whole range of assumptions associated with those simultaneous equations. It tries, through the process of modelling, to look at the interactions between various behaviours implicit in the assumptions within those simultaneous equations. It would be helpful for me to know which particular simultaneous equation you are referring to, then we might be able to address it. In general, what Henry Ergas does, and what the particular report you are referring to does, is try to ask a series of questions based around certain assumptions. Both of those reports generally come to the conclusion that there seems to be no causation or correlation between the question of ownership and the roll-out of broadband.

Senator LUNDY —Telstra's submission relies on this report to say that the OECD finding that there is a correlation between slowness of penetration and take-up and co-ownership of the HFC cable is just plain wrong and references this modelling which has been created by Henry Ergas as proof, if you like, of the counter argument. I am interested because, within the Ergas report, there is a particular figure which shows that the performance in Australia is still half that of comparable nations. So, even putting the modelling aside, I am at a loss as to why Telstra is saying that, even under its own modelling, Australia still does not compare particularly favourably relative to other countries.

Mr Scales —Thank you very much for that question. It was one of the issues we tried to address in our submission. In our submission we tried to address many of the questions which we gathered from the community, including of course the questions that came out of Senate estimates that you were involved in. One question was: `To what extent has take-up of broadband been slower in Australia than it has been in other comparable nations?' What our submission tries to do—and if I am hearing you correctly, your question goes to the heart of that issue—is to argue the case that says you have to look at the period from the beginning of the application of broadband to be able to make those comparable analyses. You will have read in our submission that we said that, for all intents and purposes, broadband has been a technology available for around three years in Australia. We go into some of the details of why it was delayed in terms of its introduction, then we ask the question, as other people have: `Is that reasonable compared to other countries?'

Senator LUNDY —Can we start at that point, then. Why do you say three years? That might be the case for something like ADSL, but it is not the case across other platforms.

Mr Scales —It is broad application of broadband. I think most of the reports that have been done, including the OECD report that you referred to earlier, have tried to get an apples for apples comparison. With respect to the three-year figure, in the spirit of openness and debate in this sort of forum, if you were to say it is two years and nine months or three years and six months, we would not have an argument with you about that, but we are certainly not saying it is five years; we are saying it is a relatively shorter period than applies in some other countries. The general point that we try to make in our submission is that you do have to get the appropriate starting point to be able to make appropriate comparisons. That is really the point that we are trying to make.

Senator LUNDY —But given that one of the contentions is that the co-ownership also had the effect of delaying introduction—and this, again, was an argument presented in the ACCC report—how do you factor that element into your formula, given that your formula is based around three years? Are you not factoring in that contention by the ACCC in your model?

Mr Scales —Let me address some elements of the ACCC report. We can come back to it at any time and in any way that you want.

Senator LUNDY —The point I am making, Mr Scales, is that factors like that have been excluded from the modelling, so we then have to look at your modelling in that context. It is a situation of looking at the problem through one window, as opposed to looking at the whole problem. I think the ACCC has gone to a great effort to consider that greater climate, rather than trying to look at it through any particular interest's window.

Mr Scales —I understand that. Let me just focus on the ACCC report for one minute, because it does relate to at least one of the questions that you have just asked. As you know, we have been critical of the ACCC report quite publicly—not privately—because we think it simply did not address some of the issues which you have just raised. For example, implicit in the issue which you have just raised is almost complete substitution between one form of technological delivery and another. Our argument is that the ACCC were simply wrong in fact, that they are quite different technological delivery mechanisms and therefore that is why, for our purposes, that has been generally excluded. Just on that point, that is the reason why we have taken that view.

Dr Warren —I suppose—and this is what the Ergas report was meant to do—we had seen some discussion that came out of the OECD, as you said, where people looked at relative positions on the OECD league table of broadband penetration and asked, `Is that correlated with any structural factors?' One of the structural factors they thought it was correlated with was co-ownership.

Senator LUNDY —Do you think that is a factor?

Dr Warren —We tested that statistically.

Senator LUNDY —Through the Ergas model?

Dr Warren —We said to ourselves, `Is that the explanation?' Professor Ergas said to us that all technology, be it mobile phones, video recorders or whatever, has an underlying S-curve of penetration: a slow take-up, a boom and then it flattens off. That is, if you like, the model that is in there. That is the logistics formula that he has in there. That is the statistical formula he tested against. Then we got data from the OECD going back over several years—the penetration rate year on year on year—and we said to ourselves: `Okay, what is going to explain differences in penetration across these countries? That is what we are seeking to explain.' The first thing you always put in there is GDP per capita, because that explains a huge amount of any variation across countries. You put in there, basically, the relative affluence of the country. As always, with any OECD comparison, that is a pretty strong indicator—for example, Japan picks up more than Turkey.

Senator LUNDY —Australia would have compared well in terms of GDP through that era.

Dr Warren —Australia certainly do very well. We are mid-rank; we are not Lichtenstein, Japan or wherever, but we do pretty well. So you put that in there. The other thing—

Senator LUNDY —Which three years are we talking about?

Dr Warren —The last three years—actually, I think it is the last five years of broadband penetration. It is since the index started. We got that from the OECD. We then asked the question: what else explains the penetration rate? One variable we put in there was time. We can go to your point about delay in a tick. But one thing was: how long has the technology been put in? Obviously, we had to split the sample size between cable and DSL because, as you quite rightly point out, cable has been here longer than three years; it has been here about six years. So we looked at that, and that proved to be an extremely strong explanator of differences across countries. So GDP, plus how old the technology is, explains the length of time. We put in a dummy variable—a one or a 0—and asked the question: did the incumbent own the largest cable network? It was found to have no statistical relationship whatsoever with penetration. Interestingly, the sign went the other way, but there was no statistical relationship. So whilst the sign going the other way suggests that, if you have the incumbent owning both the platforms, penetration will be greater, it was not statistically significant. So you cannot claim one way or the other that incumbent ownership of the platform matters.

Senator LUNDY —So, if you have not factored in the time relativity of the introduction of ADSL compared to cable, doesn't that blow your formula out of the water?

Dr Warren —No, we did do that. We did not come down in the last shower. Basically, we split the formula in two, so we had two separate formulas: one for cable and one for DSL.

Senator LUNDY —Hang on: the whole point is the competing tensions between the two forms of infrastructure. That is the point the OECD report makes: jurisdictions where you have co-ownership of both the cable and the copper access network have less penetration—and demonstrably less penetration, in fact—

Dr Warren —That is what this statistically proves is not the case.

Senator LUNDY —Statistically—

Dr Warren —We did not just do a correlation—

Senator LUNDY —I do not understand a model that does not factor in—

Dr Warren —We did factor it in.

Senator LUNDY —You say you split the cable and the ADSL. The whole point is the competitive tension between the two. If you split the model then you cancel out the effect of any competitive tension in the model, so it becomes irrelevant to the argument that is being presented. It is an irrelevant comparison to the argument that is being presented in the OECD study because it reinforces the argument that there was a delay in the introduction of ADSL; therefore, there is no competitive tension, cancelling out any possibility of being able to accurately assess co-ownership anyway.

Mr Scales —Before I ask Dr Warren to answer that question, I should say that the OECD study was a draft. The report was a draft. It has gone through a series of iterations, as you probably know. My understanding of it is that it has not yet been finalised. I understand that, like all OECD reports, it will go through a process of being drafted and it will be sent out for comment—as would happen with any reasonable paper. So I think it would be quite wrong to infer or imply that the OECD report is somehow a final report that, of itself, necessarily has to become the bible for how this issue should be addressed.

Senator LUNDY —So you are lobbying the OECD to change their findings, are you?

Mr Scales —No, I am not saying that.

Senator LUNDY —Why are you making this point?

Mr Scales —I am making the point because you said `the OECD report'. I am simply making a factual report for the committee's consideration that says—

Senator LUNDY —But are you challenging the OECD's findings?

Mr Scales —Yes, we are.

Senator LUNDY —Are you saying that there is going to be a change in the OECD's data?

Mr Scales —No; I do not know that. We do know that there is some reconsideration of this data, and the reconsideration is because we understand that the OECD is trying to make sure that it understands the circumstances in each country that would lead it to a particular set of public policy prescriptions for that particular country. So the general point I would make about that is that one size does not fit all. I am simply saying to the committee, to answer your question directly, that that needs to be taken into account when the committee makes its final consideration.

Senator LUNDY —Indeed. But I also think that is the business of the OECD—to try to turn apples into apples when doing global comparisons between jurisdictions. I have to say that I have a little more faith in their model than in Henry Ergas's model. I would still like to focus on that, and I have some specific questions, Dr Warren.

Dr Warren —It is an interesting question. With regard to your point about the splitting, I am not sure what is the best way we can discuss that, because what you have suggested does not have the effect that you are suggesting. It is kind of like saying, `Let's sit down in front of a whiteboard and work it out.' Maybe that is a better way of doing it. Maybe we can take it on notice. But it does not have that effect; I can tell you that.

Senator LUNDY —Let me put some of the statements in the NECG Ergas report to you. One of the statements in this report is as follows:

In 2002 Australia's broadband penetration rate was 1.8, twice the estimate of the lower bound.

I think his formula had established this as being at 0.9. The report continued:

Thus although the actual average for 26 countries was 3.5 in 2002, one cannot conclude, based on a simple economic model and formal statistical criteria, that the Australian penetration rate is significantly lower than [the OECD] average.

Can you explain to me, now that we have established that Australia's broadband penetration rate was 1.8 and that the statistical average of 26 countries was 3.5, how on earth you cannot conclude that Australia's penetration rate was significantly lower? This statement says that Australia's penetration rate was effectively half of the average, yet the statement says that this was not significantly lower than the average. I don't get it; it does not make sense to me. Can you explain it?

Dr Warren —The big problem is that the OECD league table, as it stands, has a whole lot of countries that are at different phases of development in terms of broadband. So the statistical average is the statistical average for countries that have had broadband for lots of years and for very few years. In other words, the table as we always see it—that one where we are at 21 or whatever it is—

Senator LUNDY —And falling.

Dr Warren —And falling—is very much a non apples with apples comparison. The Ergas statistical modelling has said, `Taking account of the start date, what would be the average,' and essentially the average is at three years, rather than asking the question, `What is the average given this hugely dispersed nature of start dates?' So the statistical average is very different from the numerical average, if you like.

Senator LUNDY —Following that through, what you are suggesting, for the purposes of comparing apples with applies, and you establish your statistical S-curve or whatever you call it—is that right?

Dr Warren —Logistics curve.

Senator LUNDY —You are just adjusting the start dates across that table so that they all match up at the start.

Dr Warren —In practice that is what we have done on page 14 of our submission.

Senator LUNDY —But that is the core of the model, isn't it?

Dr Warren —No. The dependent variable is penetration. So penetration equals a function of GDP per capita, plus the age of the technology, plus `Let's see if the incumbent owning both platforms has any implications,' plus an error term.

Senator LUNDY —So let us focus on the age of that technology. Is that when that technology was introduced into a given jurisdiction?

Dr Warren —Yes.

Senator LUNDY —So what I am saying is correct: if you are looking at your logistics curve or your S-curve, so it is GDP and all those other variables, what you are saying is that because one of those variables is how long that technology has been in a given jurisdiction, you are actually aligning the S-curve or the logistics curve.

Dr Warren —What we have done is run the model the other way. You get the parameters and then you say to yourself, `Okay, Australia is this GDP per capita; it has the technology of this age; it has the incumbent owning both platforms; where would you expect penetration to be given those factors?' And that is where it is.

Senator LUNDY —Fine, and that is where you have come to. Adding the additional variable, the point I made earlier about the relative time of the introduction of ADSL compared with cable and the competition tension between the two, I put it to you that you are not able to add in the factor of co-ownership of HFC and the network, because the timing of the introduction completely distorts that potential for competition tension. I am just making the point again.

Dr Warren —We asked ourselves the question: what is it that explains DSL and what is it that explains cable. They are very different issues when you focus in on those.

Senator LUNDY —Hang on—the whole context of the ACCC report is that they are absolutely not, and that it was the lack of competitive tension because Telstra had an interest in cable that was a contributing factor to the delays in the introduction of other technologies. That is fundamentally the assertion.

Dr Warren —We do not want to lay blame on the DSL delay. We are not doing that. But, if you look at the history, it was very clear why—

Senator LUNDY —But your model removes that from its considerations.

Mr Scales —With any model, you are trying to get a baseline where you can then vary that baseline to try to understand what are the explanators. That is all we have been doing with that model. We should not try to overcook the model, by the way, because that is really all that it is meant to do: help us understand what might explain the differences. I certainly do not expect you personally to recognise that we would have done it for this reason, we did it for the purest of reasons—that is, people were making claims that broadband take-up in Australia was lagging behind the rest of the world, so we wanted to honestly test that. But to test it you have to get some things that are static, and you have to ask yourself: what is the appropriate starting point? Then you vary certain variables to accommodate what that might help you to do. That is all we have done here.

Senator LUNDY —I put it to you that the conclusions in the Ergas report effectively counter the reality that the rest of the world has observed. I know that in the countries where there is co-ownership of the cable and the access network, all of them except Switzerland, have declined on the OECD rankings. Australia dropped from 19th to 20th; Austria, from 10th to 11th; New Zealand, 21st to 22nd; and Portugal also declined. Switzerland was different, but I think that was because it was in the midst of trying to sell its cable network. That was because competition was introduced there. But in each of those areas where it did actually decline that common factor of co-ownership exists. To me, that is statistical evidence that cannot be denied.

What I can see is a report by Ergas that not only tries to say that is not the case but tries to argue in the modelling that, if there were not co-ownership, competition would reduce. So it has gone the complete inverse with the use of this nominal zero factor in your coefficient in those formulas, to say that there would actually be greater competition with co-ownership. I, for the life of me—excuse the pun—find that whole aspect of the modelling quite impenetrable. I genuinely want to know whether that is what Telstra thinks. Does Telstra think that, based on your modelling around the world and in defiance of evidence to date, if there were greater co-ownership of cable networks and of copper access networks, competition would increase in those markets? That is what your modelling says.

Mr Scales —No. That is too simple an analysis of what any modelling does. Modelling tries to understand correlations and then we—you and I and other fair-minded individuals—have to draw causation. We have to understand the very significant difference between causation and correlation. No modelling will give you causation—none—and do not let anybody tell you any different. What we have to ask ourselves is: what are the appropriate causes that might help us to understand what might be the outcomes of a particular set of circumstances. I think what we are saying with this particular report is that we cast doubt on what has been an assumed view of the world. That is what we have done—unashamedly cast doubt on that.

Senator LUNDY —When Telstra launched its DSL services in Australia, why did it drop its basic cable modem service package from 512 kilobits per second to 256 kilobits per second?

Mr Scales —Help me to understand what is behind the question.

Senator LUNDY —In my view, it is an example of how Telstra has capitalised, if you like, on co-ownership of both cable and DSL and effectively dropped the speed of the basic package to match the new entry into the market.

Mr Scales —That might be a very good example of my point about correlation and causation. What you have done is seen a correlation and then assumed a causation; whereas, if we just pick up the general point that you have raised about speed, we take a—

Senator LUNDY —It is a fairly easy assumption to make, though.

Mr Scales —Not necessarily. The reason is that when we think about speed we think about demand for speed in the community because speed is not costless.

Senator LUNDY —Sorry?

Mr Scales —For us to create technologies that create high levels of speed for the delivery of data is not costless to us. So, as a company, we have to keep asking ourselves the appropriate question: what is the trade-off between the demand for a particular speed and the cost for that speed? That is going on all of the time. If there is a causation, as distinct from a correlation, with regard to that issue, the causation is that this is a business question based around the demand for higher and higher levels of speed. Within the company that is not a burning issue because when we get customers who come to us and say, `We require high levels of speed,' we sit down with them and work through what their demands are and we accommodate those demands in all sorts of ways.

Senator LUNDY —Would that be for corporate customers?

Mr Scales —Generally for corporate customers, yes.

Senator LUNDY —Not for residential customers or for residential DSL subscribers?

Mr Scales —Certainly there is a more direct relationship between our corporate customers and us because we are able to talk more precisely about their needs. When we are talking about what is effectively six million individual customers, it becomes much more difficult—hence the quite correct point that you have raised with us a number of times at Senate estimates about why we should have, for example, something like a demand register so that we can aggregate that knowledge in a way that helps us to accommodate that. I readily admit that we cannot quite as easily have that conversation with six million or so customers. The only thing with my general point about it is that, if we are concentrating on the speed, we look at it in terms of the demand for that and the investment that needs to be made to accommodate that investment.

Senator LUNDY —Another issue that intrigues me about this report is the impact that data capping has on penetration. One issue that I think is particularly relevant is, again, when you look at the relationship in other countries between download caps and broadband penetration. There is some symmetry there as well; that where there are download caps there is less penetration, or a lowering of penetration, and that where there are no download caps there is stronger growth. Did you factor that into the modelling that was done by Henry Ergas, and what is Telstra's response to that issue? Do you believe download caps are impacting on penetration and in what way?

Mr Scales —I will ask Mr Mullane to cover some of the details of caps, but let me give you a sense of what we think about caps. We understand that there are countries around the world that do not have caps on downloads and therefore do not have any limit on the amount of downloads that there can be. We have taken a view, again quite unashamedly, that says it is reasonable to think about the provision of this service in terms of a very broadly based user-pays approach and that the more that somebody uses this service it does seem appropriate from our perspective that there should be a differential payment based on use.

Senator LUNDY —I understand that, Mr Scales, and I am very familiar with Telstra's argument for imposing download caps. What I am really asking is: was the presence of download caps a variable in the Ergas modelling and if so, how?

Mr Scales —You asked three or four questions and I am trying to do my best to answer each of the four questions you asked. I think you asked the very general question about how we view download caps. I am trying to address your question with regard to that.

Senator LUNDY —Please accept my interjection as trying to be more specific about the information I am looking for.

Mr Scales —So your question is not so much about download caps but about how that was built into the modelling?

Senator LUNDY —Yes.

Dr Warren —We did not specifically build that in. That is because if you have a one size fits all price, which is what a non-capped price would be, you would not necessarily have a higher price, so that would choke off penetration. It may not choke off usage, which is its problem, but we would argue quite strongly that download caps are pro-penetration because they allow you to target those very low-use customers with low-use packages. That is why it did not seem to make a lot of sense to us to put that in. Also, we did not put it in because the model was very well explained by the other factors that were in there. It was not a huge amount of unexplained error. It is an interesting question. It would be worth while looking at. It would strike me as being strange because I would have thought that download caps were actually pro penetration rather than anti penetration. They may not be.

Senator LUNDY —Just looking at my notes, with respect to the countries I went through before with those sliding scales related to download, countries that did have download caps in place—I will have to check that—experienced a decline in their OECD rankings, as did countries that have co-ownership of cable networks and access networks.

Mr Scales —That would not necessarily be surprising because if one disaggregates that into its most basic outcome it is effectively saying that the service is being supplied at a lower cost than it would otherwise be supplied if there were caps. As we know from general demand and supply curve analysis, if you reduce price you will generally get an increase in demand.

Senator LUNDY —You are saying that if you removed the download caps there would be a worse performance in penetration as opposed to a better performance even though statistically the OECD material shows that where a download cap is not in place they have improved faster—there has been greater penetration.

Mr Scales —I think I was addressing a slightly different question that I thought you were asking. I thought the question you were asking was the extent to which the removal of caps would be more likely to encourage the take-up of broadband.

Senator LUNDY —Isn't that what Dr Warren just asserted?

Dr Warren —What I said was that it did not seem to make a lot of sense to me.

Senator LUNDY —You did not assert that; you asserted the opposite. The OECD figures show that where there are not download caps there has been far greater growth as opposed to where there are, where there has been a decline in OECD rankings. Again the evidence is contrary to Telstra's view of what that impact would be.

Dr Warren —Download caps are also highly inversely correlated, so with non-download caps companies are going broke. I think it is Korea where three of the top four are in liquidation. Japan's No. 2—Softbank, is it?—went broke the other day. All round the world, carriers are saying, `We've got explosive demand. We need to price the network that reflects this demand, not have a one size fits all policy.' That is really the main game. Its relationship to penetration is probably quite secondary. It is sustainable penetration that matters.

CHAIR —That is different from putting a punitive rate on downloads. Would you say to the committee that your excess download rates are reflecting economic costs, or do they include a punitive aspect?

Mr Scales —I will let Mr Mullane cover that, but, at a more general level, we would make two points. The first point is that, if we decided to do that, we would lose significant market share, because there seems to be a bit of an assumption in some of the questioning so far that we are the only player in the market; we are not. We are one of 200 players in the retail sector of this market and, if we play loose and fast with those sorts of things, people will eat our lunch. So the first point I would make is that we do not have the luxury of being able to make those choices as clearly as I think your question might have assumed we could, because other people are out there who have the opportunity to provide services where there are no caps, and they win some of our customers. That competitive tension is going on all the time. If we were to put substantial punitive charges on some of our customers who wanted to download a large amount of data, we would lose them because somebody else would pick them up. I will ask Mr Mullane to cover the specifics of the point you make.

Mr Mullane —The issue about the price or excess data usage is one that a lot of people pay attention to necessarily. The current Telstra prices range from 15.9c down to 9.9c on the BigPond plans. Different plans are available in the business market under Telstra Direct et cetera, and wholesale prices are different again. There is a range of players in the Australian market who charge more than Telstra for downloads right down to unlimited plans. In the last couple of days, I have been looking at Telecom New Zealand. They have a similar type of pricing regime that Telstra operates. Their download charges start at NZ20c a megabyte and come down to typically the same sorts of prices as the top-end plans.

The main point to consider is that the excess charge is only for excess. It applies only to the portion of the usage that goes over your plan. Our experience is that the vast majority of our customers very quickly work out what the right plans are, with assistance from the usage meter and other inputs they have. Once they settle on their plans, most customers are pretty right. If they are not, they look at other plans or other providers. It is not the major issue that I think it is sometimes made out to be.

Mr Scales —It might be worthwhile referring you to pages 23 and 24 of our submission, where we try to put this question of excess into a bit of context. The point we make, for example, is about somebody wanting to use one gigabyte per month, and we list there what that would mean. It would mean visiting 1,000 web pages a month and at the same time receiving 1,000 personal or business emails, downloading 50 MP3 files, 100 documents, 10 software programs, 100 family photos, 10 movie trailers and 10 hours of streaming video and playing 10 hours of online games.

CHAIR —I have a niece like that!

Mr Scales —I think the general comment that we wanted to make in our submission was that, while there has been quite a bit of discussion around this question of capacity and download limits, we are talking about big limits here.

Senator LUNDY —I refer to an OECD paper titled Broadband access for business. The conclusions of this report show graphically that countries where the incumbent telecommunications carrier owns the cable network have vastly less broadband growth. In particular, this report identifies a characteristic—and I quote:

In regions where the incumbent carrier owns both platforms—

that is, the copper access network and cable—

but competes against an independent cable company, (i.e. down the same street) it may drive cable modem growth but appears to slow DSL growth and may therefore hold back the overall market.

That finding of this report correlates with the experience in Australia and again raises the issue of the delay in the introduction of that telecommunications, which characterises the problem here in Australia and why our penetration rate is slower.

Dr Warren —Senator, did you say `delay ADSL and drive cable modem'?

Senator LUNDY —It said it `may drive cable modem growth but appears to slow DSL growth and may therefore hold back the overall market'. I am sure you can understand as well as I do that this is really just another observation on how co-ownership can have a depressing effect on competition, therefore on pricing and therefore on growth in a given market.

We heard some really interesting evidence earlier from SETEL about the specific challenges facing home based businesses and small businesses. One of their issues of concern is the differential pricing for ISDN based products for small businesses as opposed to residences. What is preventing Telstra from providing non-timed calls or data services under ISDN for small businesses, given you have recently launched a whole range of ISDN based products that are not based on timed calls?

Mr Scales —I will ask Mr Mullane to answer that last aspect. As I was listening to you, I felt there were a couple of points I wanted to make. First of all, the report used the word `may'.

Senator LUNDY —It is a series of observations that are being made in this report.

Mr Scales —The second point I would make about that is that the reason why, in a broad public policy sense, governments have a regulatory framework is to ensure that the `may' does not happen. That is why you have this quite elaborate regulatory framework surrounding access regimes, the way in which we operate, the way in which we operate in the market, the separation of the relationship between our retail arms and our wholesale arms and those sorts of things. By the way, I am not arguing against the general points that are made here. I am simply trying to make sure that it is seen in the context that every country tries to get that right because they know in every circumstance that that `may' is always there. So you try to put in an appropriate regulatory regime to make sure that the `may' does not happen.

Mr Mullane —On the ISDN untimed data call charge type approach for business customers, we have had the equivalent service for our consumer customer base in the market now since the end of July 2003. So it has not been there for all that long. We do recognise the importance of getting something similar there for the business sector, but we were focused initially on getting something out for the consumer sector, given the nature of the burgeoning interest in broadband in the residential market. We are working on a business proposal here, Senator. We have done some work on it but we have not finished yet. Frankly, one of the issues about developing these sorts of products is the cost of the development itself. They are generally based around large IT developments, so we are trying to balance the cost of doing it versus the benefits. But we are actively engaged on it at the moment and I hope I can tell you more next time we meet.

Mr Scales —I would like to bring to your attention an issue on page 13 of our submission. We try to pick up the point about the unconditioned local loop and the delay on that issue, where it was not declared by the ACCC under its regulatory framework until August 1999. Again, it gets to the heart of your point about the reason for the delay—why it happened and why it was not all done earlier. If you go through the points that we have made there around some of these delay issues, you will be able to see why that was the case.

Senator LUNDY —What role did Telstra play in all of the considerations of the ACCC through that period from 1997 onwards?

Mr Scales —We were negotiating with the ACCC through that whole process. Having said that, it is in our interest to be involved in extending broadband.

Senator LUNDY —Is it? I think that is a primary contention here. We all know—and I think it is extremely well understood in the sector, not least by those operating in the market in competition with Telstra—that Telstra is intent on preserving its revenues from its existing services and maximising the return from its existing infrastructure. I am getting a little bit off the track but, based on evidence before this committee and my experience, I am not able to accept some of the assertions that Telstra makes.

Mr Scales —You are entitled to make up your own mind, but I am also entitled to answer the accusation or the suggestion that we are not wanting to roll out broadband as quickly as we can—we are. We made the point—even in my opening statement—that the amount of investment we have put into the roll-out of broadband is really the equivalent of a Snowy Mountains scheme. This is big investment. The fact is that the ACCC—not us—has indicated that over the last 12 months we have had a 166 per cent increase in the take-up. These are not small numbers; these are big numbers about people being attracted to a particular technology. We have made the point that we are looking for a billion-dollar business here as quickly as we possibly can. Putting aside the public policy issues, this is a business opportunity for us. We want to be involved in first mover advantage in broadband. It is in our interest to do so.

Senator LUNDY —You say the ACCC was involved in that delay, but, if Telstra had made the choice to self-regulate and open the ADSL network, there would not have had to have been intervention by the ACCC on the unbundled local loop. I put it to you that one of the factors in that delay was Telstra's choice not to pursue that as an initiative of Telstra and to open it on terms and conditions that were considered reasonable by the ACCC. You allowed the process to develop—within the law, obviously—as is your right as a company under those regulations, but that was one of the reasons that it went on for so long. Because of your market dominance, your view and Telstra's conduct in the handling of that case were a singular determinant in the delay.

Dr Warren —What makes Australia very unusual is that the regulator required ULL to be ready and operational before we could roll out ADSL. That has not happened in almost every other OECD country. There may be good reasons for that. The ACCC did not want Telstra to get a first mover advantage.

Senator LUNDY —Isn't that a reflection on Telstra's conduct in the market to date?

Dr Warren —I do not know why the ACCC did it—

Senator LUNDY —I think it is obvious.

Dr Warren —but every other regulator in the world did not do that. Hence their consumers got DSL earlier, and hence their broadband penetration is higher.

Mr Mullane —I will just add that I was personally involved in the ADSL roll-out right from day one, and we did have a wholesale version of that in the plans right from the start.

Senator LUNDY —Does that mean that Telstra feels that the ACCC should not have compelled Telstra to make ADSL available to its competitors? That is the counterproposition to what you are saying.

Mr Mullane —No. We always had it available to the competitors. There are two pieces of ACCC action here that we are, possibly, confusing: there was the ULL approach prior to the launch of ADSL, and then there was the competition notice post the launch—a year later, or whenever it was. But it was always available; it was never not made available.

Mr Scales —Just for clarification: we did not initiate the inquiry; the ACCC did.

Senator LUNDY —Why did they do that?

Mr Scales —Because they were doing what they are required to do under the law.

Senator LUNDY —Isn't that because, if Telstra had self-regulated and provided fair access, that regulatory period would not have been needed—that intervention by the ACCC would not been needed?

Mr Scales —It is difficult to know, when you are getting these notes, where these things are coming from, but—

Senator LUNDY —I am reading from my own notes, Mr Scales, which I prepared as a result of reading your submission and the NECG report.

Mr Scales —I have seen people delivering notes to you, so I was not sure which notes you were referring to.

Senator LUNDY —I am referring to my notes, which I prepared last night—and, if you are trying to be smart, don't bother.

Mr Scales —No, I am not trying to be smart; I am just trying to get the threads of the arguments.

Senator LUNDY —I am pursuing this point you make about Telstra somehow not being responsible for the delay in the introduction of ADSL. I am contending that, in fact, Telstra had a significant amount of control over the delay and that it was tied up with, obviously, the regulatory role of the ACCC. But, had Telstra taken the initiative at various stages, the regulatory process would not have had to go to the nth degree as it did. Therefore, these technologies would have been introduced in Australia earlier rather than later.

Mr Scales —To answer your question—your accusation—no, Senator, I was trying to answer the question that you were trying to put into our mouths, which was: is Telstra now saying that it does not agree with the ACCC in terms of its preparedness to provide ADSL coverage to the wholesale level? That is what I was trying to address. You asked me that question; I am trying to answer it. As I was trying to answer your question about the ACCC: no, they were simply doing their job. That is what they were required to do. They were required to have an inquiry into this. They did that. They looked into all of the issues. They need to be fully informed about these issues so that they can make the right decision. So we do not have any complaint about that. We were simply trying to address your question about the starting date again and why it might have been delayed; it was nothing more or less than that.

Dr Warren —Unbundled local loop is an extremely, complex declaration, though. It has been in every country in the world. The difference is that, in Australia, we were required to wait for that process to work through. Other countries were not.

Senator LUNDY —How do you respond to the OECD analysis that found:

In Australia cable modem services were launched well in advance of DSL services. The reason for the earlier launch of cable modem services was that there was competition in that market segment. On the other hand, the absence of competition or local loop unbundling at that time meant that there was no imperative to launch DSL—

and that that imperative only came as the process of unbundling the local loop developed?

Dr Warren —I think, Senator—again, referring to page 15 of our submission and to the chart there—that we can hold our heads pretty high when it comes to cable modem penetration of pay TV take-up. The problem in this country, as you understand, is that pay TV has not been able to get the penetration that it has in other countries—predominantly because of antisiphoning legislation with respect to sport, which we fully understand. But if you do not have exclusive access to premier league equivalent in Australia, you are not going to get 50 per cent penetration of pay TV. If you are not going to get 50 per cent penetration of pay TV, regardless of how well you are doing—and we are the fifth-best in the world in terms of cable modem penetration per household—you are not going to get those large numbers. I know the report to which you are referring—and, again, it is a draft report—but I suspect that it does not hold any scrutiny to the facts.

Senator LUNDY —A lot of the evidence to this inquiry has identified the two primary ways in which competition can be stimulated. One, of course, is by access to the network—the access regime. The other is by investment in alternative infrastructure. We have heard evidence from a number of witnesses to the effect that there is a finite limitation to ADSL and the existing copper network. My question to Telstra is: what, if any, investment is Telstra making in new networks? We heard from one witness that the closer you can take fibre or wireless networks and so forth to the home, the greater the opportunities that can be created for affordable higher speeds, broadband et cetera. Is Telstra investing in new networks and new infrastructure that bypasses the existing copper network and will help improve penetration of broadband—and can I add another little bit on top of that?

CHAIR —It's a very long question!

Senator LUNDY —Sorry, it is a long question—but particularly in areas that are currently in that sort of digital dead zone, that cannot get ADSL, whose only opportunity is perhaps very expensive mobile data services like CDMA through the GPRS or, indeed, satellite.

Dr Warren —I think it is right to suggest that ADSL is an interim technology. It is probably the last sweating, if you like, of the old copper network assets. In copper years, if you like, we are at a sort of transition—we are at five minutes to midnight. There is quite a long delay in lead times on all this—

Senator LUNDY —Five minutes.

Dr Warren —but I think it is fair to say that everyone is thinking, `What's the next network?' and a lot of parties are trying to put down bets. Telstra is obviously asking: `Which bet do we put down? Is it wireless? Is it satellite? Is it fibre to the home? Is it whatever?'

Senator LUNDY —Isn't it a bit of everything?

Dr Warren —It probably will be a bit of everything. There are two interesting questions for us. We are the national carrier and whether or not we are forced to do it we will always have to try and do it as close to ubiquitously as possible. That is a fact of life. For us it is not a cherry-picking model. It is not: `Let's do in Brisbane,' or `Let's do it in the CBD of Sydney'—a bit like 3G with Hutchison's investment. For us, it is rather like: `What is the technology that can be economically spread out as far as possible?' That is a big question for us. The second question is: `How do we make sure we can recoup our investment and not have it effectively confiscated by access seekers?' in the sense that we put down all the bets and access seekers grab the winning bet, the one that works with consumers, go to the ACCC and say: `We want to get that just on cost base. We don't want to have to pay Telstra the cost of all those losing bets.' So for us there are a number of really difficult regulatory questions, as well as commercial and technical questions, as to how to go forward.

Senator LUNDY —Doesn't that cut both ways for other investors, though?

Dr Warren —It does, and they should be asking exactly the same question: how do they stop us confiscating their investment? Under the existing regulatory regime it is a very hard question to answer. But I think if you want to know what the big regulatory question of the next five or 10 years is, it is that. It is how you make sure that people have the right incentives to get those investments out there while we still have competitive, open access. It is not an easy balance, as you know from other inquiries we have had on competition law, for example. That, I think, is really where the next big debate is going to be.

Mr Scales —And, Senator Lundy, your point is exactly right: it is probably all of the above. The balance of all of that is a very tricky question for us. The only point of clarification, just so that there is no misunderstanding, is that when we think about the copper network we are still thinking about 10 years out. So five minutes to midnight, in this context—

Dr Warren —Doesn't mean five years.

Mr Scales —It does not. It could be 10 or even 15 years, just to get some context into that.

Senator LUNDY —Thank you.

CHAIR —Mr Scales, I have a few questions. They bounce around a bit. They come partly out of earlier evidence. Looking at the other DSLs, one of the arguments put to us this morning by SETEL, and supported to some extent by Mr Worthington from the ANU, was that a middle band would be a reasonable thing which Telstra should be developing more—the notion not of excessively fast speeds but of reliability. I think SETEL and Mr Worthington were both talking about the 100 kilobits per second type range—something which could be affordable, particularly for small business, could be targeted to small business and could be pursued along that range. One of the suggestions put to us was the development of IDSL technology, as opposed to ADSL technology, as a way of getting around the pair gains system problems. What is your response to that sort of approach from SETEL, to going for a middle band rather than pushing out fairly inflexible broadband options, particularly for small business?

Mr Mullane —I think the other part of the SETEL question that may not have been outlined there would be that it would be at a price that they think is the right price. Today, as you know, we have ISDN pretty widely available—96 per cent coverage around the countryside—but small businesses seem to have some difficulty with those timed data calls that Senator Lundy referred to earlier. Part of the answer is that when we get that untimed version of that set of products out the door that will go a long way towards meeting that very need. The IDSL type comment they have made there refers to the fact it basically uses ISDN technology to achieve an always-on type semi-broadband service. So the ISDN offer that I am speaking about would be virtually equivalent. The technology base is pretty much there in ISDN. We are doing a few other things that will complement that. We now have 1xRTT as a wireless offering, spreading—

CHAIR —I suppose the key point I was getting at was the issue of price. Their argument was that IDSL, as a slightly slower product, might be able to be delivered much cheaper than IDSN—which, as Senator Lundy said, still relies on call costs—and that the ADSL product was felt to not quite be delivering the price people wanted for the sorts of activities they were needing.

Mr Mullane —I understand exactly the point there. The point is that we do have that technology virtually available to us. It is a question of how it is packaged and offered to the customer base, and we are doing some more work on that, as I have outlined. By the way, it will not solve every pair gain situation. In a great number of our larger system—that is, the remote integrated multiplexers—we can deliver ISDN through that platform, but we have to work around the other pair gain devices when we strike those in the network.

CHAIR —Are you push for the picking up of VDSL as another option of getting around the pair gain issues?

Mr Mullane —VDSL is another deployment, really, where you push fibre out to the kerb, so to speak. It is similar to the TransACT model here in the ACT, where you have another set of opto-electronic interfaces at a kerbside point and then it runs quite a high-speed short length of copper. We are looking at that. We are looking at passive optical networks. We are looking at a range of things. We do have some active work in the Telstra research laboratories right now, examining some fibre to the home type technology. We are very engaged in that. So we are not sitting on our hands wondering what is next; we are exploring all these avenues.

CHAIR —My next question is on the satellite service. You may have heard the complaint that we had this morning from a witness from Alice Springs. He had been offered a satellite broadband service by Telstra but, because his computer system was based on Macintosh rather than on Windows, he was not able to access that. Is Telstra looking at that particular issue?

Mr Mullane —Yes. We have been in negotiations with our vendor there for quite a long period of time on this one. Our equipment is sourced from Hughes in the USA; we purchase it through Ericsson, who are their delivery agent here; and we have been directly to Hughes and through Ericsson. Finally, some equipment is going to be made available early next year, and we will just have to integrate that into our processes so that it can be offered to customers. That will enable Macintosh computers to access our satellite networks. But having said that, customers can go today and purchase a stand-alone box that will allow them to use a Macintosh on the Telstra satellite service. That equipment is available to customers through a company called Ursys. They can buy this equipment from that company.

CHAIR —Do you advise customers of that?

Mr Mullane —Yes, Country Wide apparently do make that pretty well known to the customers.

Senator TIERNEY —Roughly what is the cost of that?

Mr Mullane —I think the cost of the entry-level box is around $1,000. I am told there are at least 150 of them out there, so it is not like they do not exist. They are there, they work and they are available. The company has a web site that you can go onto. It is based in Sydney somewhere. If anybody wants to know more about it and cannot find the company, they could have a chat to the local Country Wide person and they could probably steer them closer.

CHAIR —Do you know of instances where people have been forced to shift across to Microsoft products to be able to access the satellite services?

Mr Mullane —I personally do not know of any, but if people wanted to do certain applications and were not aware of this Ursys device or absolutely had to do it now without waiting until the first half of next year then that could have happened. But I do not know of it.

CHAIR —What sort of time frame would you look at for hopefully finishing your negotiations with your vendors?

Mr Mullane —They are finished. It is just a question of getting delivery of the equipment and putting it in as part of the product delivery process. It is close.

CHAIR —We shall tell that to Alice Springs.

Mr Scales —We are talking about early 2004.

Senator LUNDY —Does that mean that, through the extended zone contract, every customer of Telstra who has either one-way or two-way satellite services has had to use Windows to access the service?

Mr Mullane —That is the standard set-up. It is like when you buy any Internet service—you get something from a provider.

Senator LUNDY —If that is the case, that is the case. I am particularly intrigued—

Mr Mullane —Unless they have one of these Ursys devices, that is correct, yes.

Senator LUNDY —How many of those devices have been deployed?

Mr Mullane —Not every extended zone service takes the Internet; some of those are just voice services. Was the question about how many extended zone services there are?

Senator LUNDY —How many have used that device you were talking about?

Mr Mullane —My information is that there are around 150 out there in the field. I believe the penetration of Macintosh versus Windows type PCs out there is around five per cent.

Senator LUNDY —I think the point here is the determinism involved in requiring people to use certain types of software. Again, I am particularly mindful of what that presents to small business but also of the case of people like the witness from Alice Springs—as Senator Cherry said—as graphics and high bandwidth businesses tend to use Mac.

Mr Mullane —I understand.

Mr Scales —We are conscious of that. That is why we have been in negotiations with various suppliers on this issue. It is one of those issues. Those people who use Mac—and as you and I both know they love their Macs—

Senator LUNDY —They do.

Mr Scales —are not going to shift from a Mac to anything. My son is one of those and he loves it; he would not use anything else.

Senator LUNDY —What a fearful dilemma for those people—to have to choose between their Mac and broadband!

CHAIR —It would split our household!

Mr Scales —I know it is a tough decision to make. We are conscious of it and we are working on it. We will have a result early in the new year.

CHAIR —I am conscious of the time, but I have a few issues I would like to go through. On page 11 of your submission you talk about the initiatives to improve the availability of ADSL for customer service by pair gain systems. It all looks very good, but how many people have you managed to shift around as a result of that? I know we have discussed pair gains in many committee inquiries, but it is the first time I have seen a big long list of all the things you are prepared to do for them. I am wondering how many people you have shifted as a result.

Mr Mullane —Let me answer it this way. We have instituted quite a number of these initiatives and we are still working on probably one or two more key initiatives, but the ones we have in the market so far are transpositions. That is, if you like, a reactive approach. Where a customer requests a service, we say, `Okay, we'll go and have a look,' and we see if there is an alternative copper path in situ that we can utilise to get around the pair gain system. That process has delivered up until now—as I had a look yesterday—about 5,340 successful outcomes. We are getting a success rate of about 30 per cent. That is quite a substantial number of additional services that would not have existed prior to that initiative being put in place in May this year.

One of the other initiatives is the one that Senator Lundy touched on before, the untimed data calls on ISDN. That capability is available through both BigPond and other ISPs who want to access that capability at the wholesale level. There are approaching 10,000 services that have been sold on that basis. So, once again, I think it is an extremely encouraging take-up and a sign of a capability that is now in the market that is hitting the spot. We are getting very positive feedback on that from the people out there dealing with the customers. In terms of some of the other initiatives, we are replacing ANT1 type pair gain systems and also—

CHAIR —ANT1, was it?

Mr Mullane —Yes. It stands for `analog NT'. It is an ISDN type piece of technology that we use to provide two telephone services to a single premise. Where a customer is desirous of an ADSL service, we can take that box out and use the copper that is sitting underneath it. We will deliver standard telephony and ADSL over that copper, provided that it meets all the other service qualification requirements.

CHAIR —Going back to your transposition option, you said that you had a 30 per cent success rate. What do you do for the other 70 per cent? Do you look at other options or are they sitting out there in non-ADSL land?

Mr Mullane —We keep the information because what it then says is that we cannot, by a reactive order-by-order process, service the other 70 per cent of the requests there, but it means that as we go into that part of the network with more investment in the access network we take all that information into account now as we build the next generation of access networks or upgrade the access networks in those locations. So it is important information that we are gathering there. In a similar way, we are getting the same sort of information out of the ADSL demand register.

CHAIR —That was my next question. What sort of information are you getting out of the ADSL demand register at this stage?

Mr Mullane —It is early days. We had a couple of little teething issues we have had to sort out. But I am pleased to be able to say today that we have got 11 exchanges in the system that have more than 10 registrations against them. We have got a flow through that is now happening.

Senator LUNDY —How many do you need to build a business case?

Mr Mullane —How many in an exchange? We have several threshold levels set—at this stage, 150 is the lowest; there is one at 225 and there is one at 300, from memory.

Senator LUNDY —There will not be any more ADSL roll-out in the foreseeable future then, if that is the response you are getting.

Mr Mullane —I disagree. We have seen the same process operating in the UK, where they have had substantially higher trigger points than those I have quoted here. They have had 900 exchanges reach their threshold.

Senator LUNDY —They have competition from cable in the UK.

Mr Mullane —We have competition from cable here in this market, too. Irrespective of that, it requires community harnessing of energy, where people get behind this and initiate local campaigns to get people interested in it. We already have a couple of examples of that happening. We just have to get the momentum of this operating, and we have some marketing collateral stuff coming through that will help communities do just that.

CHAIR —I know I am bouncing around a bit but my next question is about mobiles and downloading through mobiles. Again looking at the evidence of Mr Worthington this afternoon, he was indicating that downloading through mobiles is an option which has not been explored and emphasised sufficiently, given its potential for reasonable download speeds. But he says that the fundamental problem is that the cost for data downloads is 50 times that for ADSL BigPond broadband service. He said he could not possibly see how that could be reflective of the actual costs involved. Do you have any views on why the cost for data downloads through mobiles is 50 times that of going through BigPond broadband?

Mr Mullane —Again, I think you would have to be very careful about the actual comparison rates. I have not done that comparison, but it is certainly the case that mobile—

CHAIR —It is 0.8c per kilobyte through the GPRS data and 0.159c per megabyte through ADSL BigPond broadband service. Those are the figures he gave us.

Mr Mullane —Without focusing too much on the numbers, yes, it is more expensive in general terms to download data over mobile networks. The fundamental reason is that you have got a limited resource there in the radio spectrum and you have got a lot of people needing to use that. The idea of having people doing major, always-on type downloads over mobile networks would drive the networks into very serious congestion, so that is fairly untenable.

CHAIR —Has that been considered as an option for people who cannot get ADSL at home?

Mr Mullane —Just taking a step back from the mobile network itself to wireless type approaches, there are various wireless technologies. I have mentioned one before, 1xRTT, which we think is one of the more useful wireless technologies, and we have platforms that will handle that today in our network rather than just using GSM type technology. There are many others too. You will recall that at a previous hearing by the committee we had a presentation from Dr Bradlow, who went through a number of wireless capabilities. The costs do differ with each of those platforms. But, fundamentally, the mobile phone networks that we have today have not been designed for serious data handling. That is why we have pushed the data onto the GPRS, which is a packet data type of network. Again, it uses expensive air time.

CHAIR —So, essentially, you are trying to discourage that because of the scarcity of the spectrum. Is that the situation?

Mr Scales —No, we are not trying to discourage it. In fact, in a broader sense, this is one of the most competitive parts of the market.

Senator LUNDY —But why is it priced so high?

Mr Scales —As Denis was saying, it is a different delivery mechanism.

Senator LUNDY —We had evidence it was 50 to 100 times more expensive than some fixed-line stuff. I do not understand why.

Mr Mullane —Bear in mind that the applications those networks have been designed for are not the same as sitting on a PC for hours doing Internet type browsing or whatever. It is more for short messages when you out and about and want to keep up with your critical emails.

CHAIR —I think Mr Worthington described it as `executive level business'.

Senator LUNDY —I still do not understand why the price point is so high.

Mr Mullane —Because the engineering of the network is such that it is designed to handle relatively short volume data messages.

Senator LUNDY —So it is high to discourage extensive usage?

Mr Mullane —There is a cost basis that sits behind these networks. Once you get into anything that is mobile, you have got all these base stations all round the place—

Senator LUNDY —Okay, but I want to understand that the motivation for the high price is based on managing usage patterns as opposed to any technical or technological requirements.

Mr Scales —No, Senator. It is very important that we make clear that we are not pricing this to discourage use. Why would we?

Senator LUNDY —Why would you?

Mr Scales —Yes, why?

CHAIR —Or you would get network congestion?

Senator LUNDY —What I am hearing is that you do not want usage other than—

Mr Scales —No, we are putting in additional base stations all the time to encourage more use of the network.

Senator LUNDY —When will the price drop for those services?

Mr Mullane —You will see different pricing on some of the more suitable platforms for higher volume data transfers. Platforms operate on different frequencies, spectrum set-ups, design arrangements. There are some networks already being set up in this country by other than Telstra to do just that.

Senator LUNDY —Who is competing against you in that 2.5 gigabit area?

Mr Mullane —I do not know about the actual frequencies that everybody is using, but there are a couple of other providers that I know about.

Senator LUNDY —Who are specifically competing against you in the 2.5 gigabit area?

Mr Mullane —As I say, I cannot comment on the frequency range because it is not my field of expertise.

Senator LUNDY —It is not a frequency. It is not quite third generation, it is not quite second generation.

Mr Mullane —I am sorry, I misunderstood what you meant.

Mr Scales —It is almost everybody—Optus, Vodafone, Hutchison. You have only got to read the technology papers to be able to see that this is a highly competitive area.

CHAIR —Again, I am sorry to be bouncing around a bit with my questions, but I know we have to finish shortly and most of the questions on the issue of infrastructure competition and the ACCC's emerging structure report were dealt with by Senator Lundy. I will say that I found that a fascinating discussion, from the point of view that the OECD, the ACCC, the National Competition Council and the ITU are wrong and Telstra is right—and A.T. Kearney is apparently right, too. I just want to throw one thing into the mix. One of my few bits of recreational reading is the Economist magazine every week. It had a survey on telcos the week before last which I thought essentially came around to a view similar to that of the ACCC, and that is that real competition in future will be between platforms—between mobiles, terrestrial and wireless. It appeared to confirm the view of the ACCC that getting that competition between platforms is really the future for telecommunications. I am wondering whether you have read through that analysis in the Economist or if you have a general comment on that, other than what you have said in response to Senator Lundy earlier.

Mr Scales —That is not new—that is very conventional, traditional economic thinking. And that is the basis upon which the whole regulatory framework has been established—that if you go back into the whole basis of competition policy in Australia, the first assumption is that you should try and encourage competition between all sorts of providers, including, in our space, network providers. Where you have what is, for all intents and purposes, a natural monopoly, then you put in a regulatory regime to try and make sure people have access to that regulatory regime at appropriate prices and with appropriate terms and conditions. So nothing in the article in the Economist—certainly from our perspective—is other than what you would expect from it, which is conventional, appropriate analysis.

Senator LUNDY —But your Ergas report challenges it. The Ergas report contends that co-ownership of two different major platforms in a given jurisdiction will improve competition.

Mr Scales —No, it does not challenge that. It does not challenge the underlying assumptions that were implicit in the Economist article. It does not challenge that at all.

Dr Warren —We are not saying the OECD is wrong. There is something I would like to add because I think it is important. I think it is clearer in the latest version of the structural separation report which has just come out, but it is something that was inexplicably missed in the ACCC report. The OECD is very clear that if the incumbent owns the only cable network then it should be—as European law suggests—divested. If, however, the incumbent has Optus overbuild equivalent, as happens in Australia and in Portugal, then it does not. That is clearly understood in the latest OECD report and it is clearly understood in European law. For some unknown reason, the ACCC decided to ignore that in its 18-month report, which went to 100 pages. We are not saying it is wrong—the ACCC report is. We are not saying the OECD is wrong or the European Union is wrong. We are just saying: read into it in a bit more detail, rather than just taking the headline.

Senator LUNDY —Overlay that with the context of that report, which was the content sharing deal. Are you saying that that is not a factor, the existence of two cable networks, and the way in which those products and content are now offered on those networks? How can you not factor that into that assertion? You cannot?

Mr Scales —No, I was not exactly sure of the question.

Senator LUNDY —It was a comment.

CHAIR —We have had complaints—and we have heard this time and time again—that Telstra's wholesale rates for ADSL and broadband often end up being higher than its post-bundling retail rate, which makes it very difficult for people to compete. Would you concede, in light of your knowledge of your pricing systems, that that would ultimately be the case once bundling is taken into account?

Mr Scales —We are not allowed to do that—it is illegal. If people believe that, then they should go straight to the ACCC and make that known. That is not allowed under the law.

CHAIR —All right, we will take it up with the ACCC. My final question relates to a common complaint we have heard, which is that customers will apply for a non-Telstra ADSL service and be refused on the grounds that Telstra Wholesale cannot provide the service over the copper line, but subsequently the same customer is offered ADSL by Telstra. Are you aware of any complaints that have come through where those sorts of issues are raised?

Mr Scales —Yes, and we have discussed this very issue at Senate estimates. There certainly have been some complaints to us, and we understand there have been complaints to the ACCC, and we are working through each one of them. Part of the dilemma is that people can go to a web site and try to understand whether they are eligible to get ADSL. Then, when you go and do the work and see how far they are from an exchange, sometimes you get the wrong outcome because it is not 3.5 kilometres from the exchange as the crow flies—it goes around in circles. So we can have a situation where, for whatever reason, people might have been given the wrong advice and then when they go back and get another piece of advice, they get different advice. We are trying to get a much clearer understanding of that so that we can advise people accordingly. But we are aware that there have been some complaints.

Mr Mullane —We take every one of those very seriously and we investigate them fully. We have tightened some processes over the last few months. It is fair to say that every one of those instances with factual issues behind them—and there are a lot of complaints we get where it is almost impossible to determine if there has been an issue—they generally relate to a manual service qualification, and we are continually tightening that process. That process happens not in the retail or wholesale parts of the company; it is always in the infrastructure services side of Telstra. If anybody has any of those complaints, we absolutely want to hear about them because, if it is happening, we want to stop it. In reality, we have found very few instances where it has happened, but there have been a couple.

Mr Scales —And we have to fix it, and we do.

CHAIR —Thank you very much. That concludes today's proceedings, and I thank the witnesses for their presentations. The committee will travel to Sydney to get more evidence tomorrow on these issues.

Committee adjourned at 3.42 p.m.