Welcome to the latest in our In Conversation series, between Opposition Spokesman for Communications and Broadband, Malcolm Turnbull and economist, Professor Joshua Gans.
This week, attention has again turned to the future of Australia’s largest, and most controversial, infrastructure project – the NBN. The Australian Competition and Consumer Commission (ACCC) released a statement saying it was unable to accept a crucial aspect of Telstra’s move to structurally split its retail and wholesale arms.
In this interview, conducted last month before Professor Gans took up his new University of Toronto post, Turnbull says: “I’m not here to defend Telstra … but it has been literally trench warfare between the ACCC and Telstra for a long time.”
The wide-ranging discussion also touches on:
The NBN will lead to higher prices – and just where did the cost-benefit analysis go? What can we learn from New Zealand’s experience?
Is the NBN about social benefits – such as e-health – or will it mostly provide video entertainment? Turnbull says: “Why are we spending $50 billion so that over-the-top companies like YouTube, Google, Netflix or Australian equivalents are going to be able to deliver more high definition streaming video?”
The digital economy is pressuring the retail sector and forcing job losses. Turnbull says: “I’ve talked to Australian online retailers who are moving their distribution, their headquarters, their warehouse … moving it to New Zealand so they can take advantage of [cheaper postage]”.
Retail versus mining. Turnbull says: “I was at a dinner recently where … one of the leading figures of the Australian business community said, ‘I don’t care about retail workers, I don’t care if they lose their jobs, they’re not as productive as mine workers’.”
E-books won’t engender romance. Gans says: “I once thought: who cares that … a book is no longer going to exist? We’ll have a digital copy. Then my wife said: do you remember the first time we met? … I saw that you had all these books on the shelves …”
Telstra, the NBN and market power
Joshua Gans: So I think what we should probably talk about first is broadband. In particular, I was quite excited two or three years ago when the government announced it was going to build it all itself, primarily because for so many years, we’d been stuck in the morass of market-power dominated telecommunications and at the very least it looked like something that could provide some competition.
A few years hence, I’m not so sure anymore. And I was just wondering what your take was on what’s happened; how it is that aspirations of competition seem to have disappeared, and how you get around that?
Malcolm Turnbull: Well, the government identified a couple of problems – and it was not a brilliant flash of insight on the part of them to do this – but the problems were the vertical integration of Telstra.
They [Telstra] owned the customer access network and they also owned a retail business, and they were the biggest customer – and user – on that customer access network. And that provided a conflict of interest. So one policy goal was set to be the structural separation of Telstra.
The other one was the fact that many Australians have poor broadband and they want better broadband services. The NBN allegedly will deal with both of these – but at a gigantic cost.
If the vertical integration of Telstra is the problem, then the solution is structural separation, right? You don’t need to build an entirely new telecommunications network and render Telstra’s existing customer network obsolete to do that.
If you want to get everybody better broadband, why would you not have a technologically agnostic approach that says we will use whatever mix of technologies that will deliver the broadband speeds that people need or are prepared to pay for, and at the lowest cost to both consumers and indeed to taxpayers?
Now, all of this would have come out if they had done a proper cost-benefit analysis – and I know as an economist, you will be sympathetic to that – but the failure to ever ask that question, to ever define the objective, then ask what is the most cost effective way of getting there, that’s the core problem.
Gans: To reflect on that, I wonder how easy it was to come up with other plans? Having sat there in meeting after meeting where they propose different things – originally it was fibre-to-the-node, which was going to end up with two sequential monopolies essentially connecting to Telstra.
There were earlier plans from the Coalition, which was not to worry about urban broadband at all, but to worry about how to provide services to regional areas. Around the rest of the world, the only places where you seem to have progress on high-speed broadband are ones where the government really did take this larger role. I’m thinking of countries such as Japan and South Korea.
Turnbull: The governments are not building broadband networks nationally in those countries.
Gans: Not nationally, but they did have substantial inducements to do so.
Turnbull: But there is nowhere, Josh, in the world where this amount of money is being spent, so this off the charts. There actually are charts which show the amount of government money supporting broadband in Australia compare to other countries – it’s one of those bars which has to have a little break in it because you can’t fit in all on the page, it’s so high and so extravagant.
And the second one is, of course, that nowhere else in the world are they actually building a new government-owned monopoly, because from a competition point of view, in the rest of the world, in most countries you do get competition at the facilities level. And indeed it is a tenet or an objective of government policy to encourage facilities-based competition and typically it is between fibre, where it is present, between hybrid fibre coaxial (HFC), in other words, pay TV cable, and between copper and some versions of ADSL and VDSL. And you get that competition and that’s true just about everywhere.
But here, not only is the government building a new network itself, but it’s going to decommission the Telstra copper network and more incredibly still, legislate and contract to prevent Telstra and Optus from using their HFC cables, which pass around a bit less than a third of Australian households, to prevent them from using that to carry broadband and voice, for no reason other than preventing competition with the NBN.
Now this, from a policy point of view, seems to me to be so retrograde. And it casts my mind back to when you were either babes in arms, or not born at all, when state governments used to own railways – and they still do – and would impose all sorts of regulatory and licensing restrictions on the trucking industry to prevent and discourage competition with state government-owned railway.
Now that was all struck aside by a combination of High Court and microeconomic reform but we’re going back to that. The government builds a big new monopoly, they’ve overcapitalised it and the only way we can get even an anaemic return is to prevent anybody competing with it. What’s that going to lead to? It’s going to lead to higher prices isn’t it?
Gans: It’s completely outrageous. There is no doubt about that.
I wonder where the Australian Competition and Consumer Commission (ACCC) was in all of this? This has been operating as a business – just because it is a government-owned business doesn’t mean it’s not subject to that (scrutiny). These things seem extraordinary to me.
Turnbull: I’ll tell you where I think the ACCC has been, under Graeme Samuel, it’s been a great supporter (of the NBN) because of its extraordinary animosity towards Telstra. I’m not suggesting that is not justified – I’m not here to defend Telstra, but it has been literally trench warfare between the ACCC and Telstra for a long time …
Turnbull: … Which is one of the reasons Telstra should have structurally separated long ago. The other thing is that in the NBN legislation, this arrangement between Telstra and the NBN is essentially deemed to conform with the requirements of the Consumer and Competition Act, formerly known as the Trade Practices Act.
Gans: Right, still, nonetheless, in other times, the ACCC would have come out even vocally against that sort of thing and that hasn’t happened.
Turnbull: You would think so, but they have much more limited scope because the agreement that was entered into with respect to the HFC is about as anti-competitive an agreement as you could imagine.
It is avowedly anti-competitive, because they (the government) has said we’re doing this to protect our financial position, which is what monopolists always do.
Gans: If we come back to the cost-benefit analysis, you’re right, as an economist, I want to see a cost-benefit analysis. But something tells me that we already know the answer to this one. That, certainly on a commercial basis, the private sector has already spoken, the cost-benefit analysis for putting this in wasn’t there. It wasn’t going to stack up.
Also, in terms of how we think about this, if you wanted to do a social cost-benefit analysis, if you put aside the government’s statements that this is all going to be commercial and you tried to say, “ok, let’s do this as a public good …”
Turnbull: … Just so we don’t gloss over the history, when this was announced in 2009, Kevin Rudd said this would be a commercial business, it would be 49% owned by the private sector …
Turnbull: … mums and dads would be lining up to invest in it, it would be such a hot deal. They’ve forgotten all about that and they’ve now said, no it’s not going to be commercial … they’ve got a business plan with heroically optimistic assumptions in it which tells them it is going to develop a 7% return.
Now, you and I know, and we all understand the optimism bias which always comes with infrastructure. It’s hard to think of any public infrastructure project of any complexity that has not run significantly over budget. I am sure there are some that haven’t, but they are the exception.
So, the probability is that this will be even more expensive than it is forecast to be and there are assumptions about take-up and usage – and the assumption that people will pay a significant premium for higher speeds – is not borne out by experience either here or anywhere else.
So the probability is that you’ll end up with the taxpayer spending $50 billion, $60 billion, or whatever it will cost to build it and you’ll end up with an asset worth a fraction of that. Whether than fraction is a half, or a third or a quarter, it’s hard to say, but it’s a huge haircut for the taxpayer. And that is a loss to the common wheel and the public good, and net detriment to the public good.
Private over public
Gans: But part of this is also in the way that it’s been done. It’s been to maximise the private benefit so that it looks like the public supply of a private good as opposed to … one can imagine a lot of things. Had they done something about competition, really, that would have lead to prices pretty much plummeting to zero and all of us going on subscriptions. That would have been a benefit in terms of consumer surplus, and outweighed any profit loss.
The other things they could have done were reforms and other things to actually improve the adoption of services to do with broadband. So we talk about e-government … there are myriad consultancy reports all over the place about the potential benefits of being able to digitise government records, government engagement, things like the Census, are an example.
Turnbull: E-health, e-education.
Gans: All of that.
Gans: The big objections are normally – what about the people who don’t have access, or who don’t have access to broadband? This proposal on the way is still going to have that problem, because of the high prices.
So this government has not taken the opportunity to say, “OK, we are building a public good – if that’s the rationale – then we should work out how to make it publicly accessible.”
What I worry about is that we’re losing opportunities to move forward. If you’re going to spend this cost, and if we took that as a given, you should also spend it and structure it in way that’s going to maximise social return, not just the private return. It seems like we’ve swept all those things away.
Turnbull: Let me pick you up on a few things. You’re about dead right on the importance of the affordability. The biggest barrier to internet access is not technology or geography, it’s lack of household income and that’s quite clear from the statistics.
So why would the government spend tens of billions of dollars on doing anything with the broadband if it’s not going to make it more affordable?
The one thing that is quite clear is what this will actually not do: it will not make the broadband more affordable, at higher speeds it will be more expensive and the history, here and abroad, is that people won’t pay that premium.
Secondly, of course we’ve seen prices coming down really dramatically over the last five years. Dramatically coming down for all forms of telecommunications. Because this is a monopoly there won’t be the competitive tension to continue that decline.
The second point however, all of those social benefits – which I don’t minimise or diminish in any way, you know, e-health, e-government, e-education – is you don’t need enormous bandwidth to deliver that.
Yes, if you have a medical centre in a remote area and you want a surgeon in Sydney to be able to give directions to a doctor, then it’s good to have very high-definition bandwidth. But people are not going to need that degree of high definition for e-health at home.
When you actually go through the applications you talk about, the ones regarded as productivity-enhancing, they actually are not ones that need enormous bandwidth and they need only a reasonable upgrade of broadband – whether it’s 20 megabytes or 25, or 15 – if that was universally done, say if you had 12mbs everywhere in Australia.
I challenge you to identify which one of those productivity-enhancing applications would not be very available.
The final thing I’d say is that when you look at high-speed broadband everywhere else in the world, the only thing that drives its take-up, in so far as this is being taken up, is video entertainment. And so this why the triple play is so important – this is why some countries, such as Singapore, are moving to essentially make it impossible for one telco to have exclusive rights to video content.
If high definition video is going to be the selling point, you don’t need 100mbs to access that.
But if it is, then why are we spending $50 billion so that over-the-top companies like YouTube, Google, Netflix or Australian equivalents are going to be able to deliver more high-definition, streaming video.
I’m not suggesting that’s a bad thing, people are free to watch whatever they like as far as I’m concerned. But is that worthy of such a massive subsidy?
Gans: No, even if it was, the copyright laws are not harmonised appropriately with the US and there’s a lot of exclusive content deals, and it’s not clear the consumers will get that anyway.
And even if we take the medical examples, I once talked with some MBA students of mine about the idea of providing in-house ability to take to a photograph of the inner ear and transmit it over the internet.
So instead of going down at a bad time of the evening to work out if your child has an ear infection, all the doctors do is look in an ear you could shoot them a photograph and nine times out of 10 they could determine whether there’s an ear infection or not.
No one has to wait and you can just do it and the big barrier to that was liability; medical liability was not going to cover that. Medicare was not equipped to think about that as a procedure.
And these are things under the government’s control, there are rules and regulations under the government’s control that if they were thought about and relaxed could actually enhance the value of broadband socially. It seems there has been all this emphasis on laying down the pipes but nothing about all the other things that could be going on.
Turnbull: Amen to that. I mean this is the point I made actually in my opening remarks at the National Press Club (in August), that we are spending a lot of time talking about the pipes, but not enough time talking about the digital economy per se and the applications that are going to be developed.
Not all upside
We should not imagine that the digital world is all upside. If you are a fashion retailer in Australia, the digital world is not cheering you up at the moment because a lot of your customers are buying their goods overseas and they, and this is the point Rupert Murdoch made – not about fashion but generally – the internet generally will destroy more profitable businesses than it will create.
But think about this in fashion – because we are a season behind the northern hemisphere. When it’s summer, for example, the items or articles that didn’t sell over there can be sold here in Australia.
So that’s been good, that’s been something that we’ve benefited from. But if you want to be really hip and access those fashions immediately now, if you’re a real fashionista, you can buy those clothes and other items online and they can be nowadays, the fulfilment is so good, that fulfilment can come from China or elsewhere in Europe or America in three, four, five days.
So you don’t have to modify your passion for instant gratification too much and so what’s now happening is that people can be up-to-date with fashion.
Buying it more cheaply means the ability of the local retailer to impose a cost that reflects his wage rates and rents is mitigated because instead of competing with the shop across the street or the shop across town, he’s competing with an online retailer at the other end of the world and that’s is a very, very big challenge here.
I think we need to be having more of a conversation about what we are going to do to make ourselves more dynamic and more competitive and more productive. We have a lot of industries that are traded.
Need for competition
But retail, historically has been non-traded, it’s basically only competed with other retailers in Australia and that’s why historically (and Japan’s a classic case) the traded sector is highly efficient because of competition and the non-traded sector is not terribly efficient.
When the digital world makes the non-traded sectors and the economy traded ones, in other words making them compete with the whole world, the pressure to compete and become more productive and efficient is so much greater.
And what do you think about that? You understand all this; you understand all this global economics a lot more than I do. You’re the professor.
Gans: Well my bias would be, I don’t have to go around talking to constituents and small business people. Therefore, my bias is well, that’s increased competition.
Turnbull: Well, you say that’s tough luck.
Gans: Well, I would never be as harsh as that. But a part of it is, you know, that’s the process of creative destruction, basically.
Turnbull: It doesn’t console someone if the destruction they’re enjoying is determined by an economist creating it.
Gans: Exactly, that’s the heartless economist.
Turnbull: So you are a trumpeter of trumpeter.
Gans: Ha ha, something like that. You look at those things and so when people complain, when Borders went bankrupt recently, how are we supposed to think about that? It’s not just the business itself that people reminisce about, it’s meeting their significant other in a bookstore and all those things of where will it all go.
Turnbull: I was talking to a guy this morning, a friend of mine who is in this area and he made that point. That’s an interesting thing to talk about, what is going to happen to a shopping centre or a shopping mall if the number of specialty retailers has diminished because the margin has shrunk?
Now, the consumer is keeping their money in their pocket, because they are more prudent or we assume that is a cyclical thing. But then you’ve also got the long-term trend of global competition, so what are the implications of social intercourse?
Gans: It’s very hard to imagine a way out of it, I mean I was once thinking: who cares about the fact a book is no longer going to exist? We’ll have a digital copy.
And I wrote a post about that and my wife read the post and she said: “ahh, do you remember the first time we met? I came to your apartment and I saw that you had all these books on the shelves. And that’s when she thought I was someone, with all these books and similar interests. (She thought) that’s the one for me.”
Turnbull: She didn’t know you had bought them by the yard.
Gans: I bought them at a cheap sale just for decoration.
Turnbull: Ah, so you admit it Josh, that’s amazing.
Gans: It’s very hard to imagine how these things evolved. Just in the last five years our interactions have evolved to, you know, dealing on Facebook. You just watch kids and you think are they losing something? But maybe not. And so even these sorts of things seem to change.
Miners vs. retail
Turnbull: Here’s a question for you, what is the policy response? I mean the thing that I’m troubled by is that the industry which seems to me to be losing employers at the greatest rate is the retail sector and hardly anyone is talking about it.
To the extent I was at a dinner recently where there was one of the leading figures of the Australian business community, a very, very big-wig, and I raised this issue and he said: “I don’t care about retail workers, I don’t care if they lose their jobs, they’re not as productive as mine workers.”
Well, I don’t think there are too many mineworkers that are going to lose their jobs, that industry is on a roll. But you know, are retail workers’ jobs less worthy than those in other sectors?
Gans: That’s a dangerous road to be going on …
Turnbull: I think it’s nonsense, don’t you?
Gans: Yeah well, I mean the problem is the market does place value on different jobs and that changes as things go on. So what has the internet brought in terms of destruction of some of these retail jobs? It has been the jobs of the larger chains. Some of the smaller bookstores for instance are reopening again and coming back again.
Turnbull: I think jobs have been lost. I’d like to verify that because you see the big chains because they hit the headlines – when Borders closes down and 1200 people lose their jobs. But what if there are 1200 smaller shops who all put off one shop assistant? That is a little bit below the radar. A lot of that work is part-time, this is something we have to be alert to now.
Can’t stop the digital economy
Here’s the thing, there is nothing you can do about it, you can’t stop the digital economy, we agree with that.
But how do we make ourselves more productive and more competitive? This is where issues like labour market flexibility are very important.
One assumes rents in Australian retail have always been regarded as too high relative to other countries.
They will come down and the market will deal with that, but labour market flexibility is one. There are some big issues relating to the cost of fulfilment.
Because of the curious way the postal treaties work, it is cheaper to send a package from some points overseas to Melbourne than it is to send one from Sydney to Melbourne.
I’ve talked to Australian online retailers who are moving their distribution, their headquarters, their warehouse if you would like, moving it to New Zealand so they can take advantage of that. Now that is faintly nutty. There are some dysfunctions there, some market failures we’ve got to look at.
GST and online retailers
Gans: You do want to worry about that, as you do want to worry about incidents of (whether online retailers should pay) GST or not, that seems sensible. The Productivity Commission came out and said, “of course you want a neutral system, but it was going to cost so much to collect it”.
Turnbull: I suspect what you need to do, and this is not easy, is have some treaty arrangement where there was some kind of GST that was collected.
Gans: But think about the problem, in the United States they can’t even work out how to tax Amazon. So there are distortions there just going on within the same country.
Turnbull: Correct, that’s right. The argument there, which is really the metaphor I was thinking about. I was talking to some American online retailers the other day and what they said, and obviously Amazon doesn’t want this, but what the policy makers are starting to talk about is having a national VAT [value-added tax] in America.
But of course the politics of this, America is arguably ungovernable at the moment. But that would be the rational thing.
Gans: Right, well you lose something; there was a benefit of some degree of having local tax competition, which can keep some local governments in line. But right at the moment there is straight-out avoidance going on. It’s not fair for the local business.
Turnbull: And all these arcane arguments about whether an online vendor has a connection with a particular state and whether they have a sufficient connection to bring them within a tax net or not. I mean, it looks like a lawyers’ paradise. You could say that about much of American life.
Gans: Exactly, Exactly. So tell me how would you actually make the policy process better? I mean you’re worried about the NBN being a white elephant, a reason to worry about it. What would be the procedure to do it? Because the previous government, the coalition did not manage to get a broadband policy enacted.
Turnbull: Well we did, we had a broadband policy that dealt with the area of greatest disadvantage, which was regional and rural Australia.
This was a mix of technologies that are essentially the same as the NBN is going to use in those regions which is fixed, wireless and satellite. There was a full, quite transparent, federal government subsidiary of under a billion dollars. About $900 million.
Had that policy been implemented by Labor, those areas would have fast broadband by now, instead of having to wait for up to 10 years. So being told nirvana is coming, but that you might have to wait a decade for it, is not much consolation.
Gans: But if the electorate appears to be demanding a government response in the urban areas for faster broadband because it’s true, that instant gratification, it takes time for that YouTube video to load and this is something that is hitting people when they come to cast their votes. So if I accepted that as a constraint, how would you go about satisfying that desire, taken as legitimate and people are going to want the government to do it. How would you go about it?
Turnbull: The way I outlined in the talk I gave in Canberra was simply this: what the government should do, and this may be NBN Co, is to say to the private sector “we want to achieve national broadband speeds of not less than 12mbs as soon as possible, ideally within 12 months.
Many areas in Australia have that already, some areas in the cities do not. There are some areas where it is going to take longer to achieve.
The aim is to get everybody very quickly up until at least that level that you say ADSL2+ speeds or better and many people have got better speeds than that.
Who would do that? The logical people that would do that would be Telstra. Of course, they are vertically integrated.
New Zealand experience
So the answer is, Telstra should structurally separate, and what I’ve argued for is they should do what they have done in New Zealand.
In New Zealand, Telecom New Zealand, just the PTT (it’s the facsimile of Telstra in that country), has structurally separated so its customer network business is in a new company called Chorus which will be split. Every Telecom New Zealand shareholder will get share in Chorus and a share in the rest of Telecom, which I think they are going to call Telecom 2, the retail business. Then you have got a completely separate business.
Now what that becomes in effect is a national broadband network. It’s privately owned. It is regulated like any other utility, so it can charge prices that give it a reasonable return on its capital.
It shouldn’t be protected from competition, quite the contrary. But in areas where it cannot deliver the broadband that is regarded as equitable and socially important because of geography, then it should seek and obtain an appropriate subsidy. Now this is essentially what’s happened in New Zealand.
The experience there is a much better process, and by the way, Chorus, the Telecom customer access company, is not doing all of the rollout.
In a number of areas, there are electricity companies, what they call lines companies. They have taken up the challenge, getting a degree of government subsidy, and are rolling out fibre networks in the north or the north island, the centre of the north island, not Auckland, and Christchurch.
This is a complete transcript of the interview which has been edited for fluency and concision only.