Is the NBN value for money?

Today the Abbot government released their cost-benefit analysis of the NBN and their own, mixed-technology model. Not surprisingly (for a report commissioned by the Government), the analysis finds that the Coalition’s fibre-to-the-node NBN is more cost-effective than Labor’s fibre-to-the-premises (FTTP) model. The case for FTTP isn’t good, according to this document. The mixed-technology model comes out $16b ahead in terms of value for money.

I discussed this report on the The Project in the evening. Have a watch below.

Setting aside the issue of the impartiality of the study, one can assume they have the costs in the ball-park at least. But what about the benefits? This is where the debate will be, because some of the assumptions about the value to the country of faster broadband are highly questionable.

Here are some issues that I noted in my first reading of the document. They defy common sense and require some detailed analysis.

• The reference case used was an unsubsidised rollout of faster broadband, which incidentally had the best net benefit, namely zero. This implies that the benefits of the broadband to society are barely worth the money spent on them by industry and don’t warrant further government investment. The Coalition’s MTM model has -$6.1b in net benefits, because of the cost of bringing broadband to regional Australia which outweighs the benefits nine to one. That’s -$6890 in benefit per household. What does that say about the value of being connected in the bush?

Source: Independent cost-benefit analysis of broadband and review of regulation, Volume II

• Setting costs aside, the benefits of a full FTTP rollout are $4.7b lower than the MTM model. In other words, even if FTTP cost the same as FTTN, you’d still be better off keeping the copper network. How can this be? Because the loss of utility due to delay (up to 4 years) caused by the slower rollout. Yes, people need better broadband sooner. But I do not understand how, even with a four-year delay, the benefits of ubiquitous gigabit-ready fibre, and of disposing with the copper network, do not outweigh the benefits of the slower, more complex hybrid model with its reliance on ageing infrastructure owned by Telstra. Yes, it’s more expensive, and it’s legitimate to ask whether the extra expense is worth the money, but I question any model that yields this result.

• One of the key ways that the benefits of broadband were modelled was by predicting consumer willingness to pay (WTP) for broadband services. How much a consumer is willing to pay for something is a standard way to measure the quantitative value of its utility to that consumer, so this makes some sense. However, extrapolating out a decade and beyond is difficult when the commodity is internet service, the potential of which we are barely beginning to realise. Imagine going back 20 years an performing an analysis using only this metric – would consumers have placed the proper premium on their desire for bandwidth given the services available in 1994? The report notes that (along with some other assumptions) WTP for broadband would have to grow at 13% a year for the FTTP model to become feasible. Is this not, in fact, highly possible? At what rate has this grown in the past 10-20 years? Is it plausible, as the report states, that WTP is “not expected to grow at all for higher speeds (greater than 50 Mbps download or greater than 9 Mbps upload).” Do you see demand for upstream bandwidth topping out at 9Mbps in the next two decades? Because I just uploaded the YouTube clip above at a speed 4 times times faster than that.

• The following line raises even bigger questions for me. “The majority of the benefits from higher speed broadband accrue to private uses within households and businesses.” It goes on to say that “benefits accruing outside individual households or businesses such as in health and education… are a very small proportion of the total benefits”. Let’s tease this out a bit more because I think it’s really at the heart of the broadband debate as it relates to Australia’s future.

The Coalition have always seen the internet as an entertainment device for nerds and not as the engine of the future economy. While they can appreciate that being able to stream TV has some value, say on the order of a few tens of dollars a month, the flow on effects in terms of economic growth, productivity and opportunity afforded by ubiquitous high-speed broadband go unstated. Personally, I imagine them to be virtually limitless, but even with a more sober approach this surely must be quantified somehow.

What’s more, these benefits are amplified by the number of people that enjoy them. Communications technology is only useful if the people you want to communicate with are similarly empowered. To coin a phrase, “if you don’t build it, they won’t come”. The chicken of next-generation services in health, education, entertainment and god-knows-what-else requires the egg of plentiful broadband; and the demand for that broadband will be driven by those services. This won’t be an incremental process, struggling to reach 13% a year. It will be exponential. Unless its growth is stunted by an infrastructure bottleneck – such as the one recommended by this study.

Imagine where our economy would be if communication was still by letter and fax, photos had to be developed before they could be used and international phone calls were an expensive luxury. The way communications technology has completely remade our society and economy in the last two decades has not and cannot be captured solely in the private benefits enjoyed by individual users. I don’t know the magnitude of this multiplier effect but whatever it is I do not see it in this cost-benefit analysis.

I am looking forward to hearing from experts in industry and academia. This analysis, even if it turns out to be flawed, could kick-start a bit of a debate about the benefits of broadband. It is a debate that hitherto has been sold well short.