When Australia’s $40 billion National Broadband Network (NBN) plan to build a government-funded Fiber-to-the-Premises (FTTP) network reaching 93% of Australian residences was announced in 2008, it represented the single biggest infrastructure investment in the nation’s history. By way of comparison, its $40 billion budget exceeded (0) estimates for the contemporaneously-approved Gorgon Gas project (1), which when completed will be one of the world’s largest natural gas field developments.
For the Gorgon project, both shareholder investment and government approval were contingent upon detailed cost-benefit analyses (CBAs) specifying average expected net benefits and the sensitivity of these projections to key parameters. This is normal ‘good practice’ for large infrastructure projects. Therefore, it was somewhat surprising that the NBN, with a larger budget than Gorgon, was approved without any cost-benefit analysis. Somewhat belatedly, in 2010 a ‘business case’ was released (2), but this fell far short of the sort of CBA typically required for government infrastructure investment in, for example, electricity or road construction. Government politicians at the time justified the absence of a CBA with claims that a CBA was unnecessary because no one could tell in advance what the future benefits might be. One was famously quoted as saying a CBA was an unnecessary waste of taxpayers’ money because the government was going to build the NBN ‘come hell or high water’ – that is, regardless of what any analysis might indicate.
The most ferocious critic of the then-government’s lack of empirical support for the NBN was opposition politician Malcolm Turnbull who, in 2013, was appointed Communications Minister following the landslide electoral defeat of the Labor-led government who were responsible for implementing the NBN. Mr Turnbull immediately commissioned a review of the NBN, with a view to downscaling it to a more-modest ‘mixed technology model’ (MTM) network utilizing existing copper and cable networks (which were decommissioned in the NBN) in addition to a slower FTTP deployment driven by demand rather than political imperatives. Unlike his predecessor, Mr Turnbull did commission a CBA of the entire MTM, overseen by an ‘independent’ panel of experts led by Public Finance expert and former Secretary to the Victorian State Treasury Dr. Michael Vertigan. The Vertigan Committee CBA (3) was publicly released on August 27.
The report makes for an interesting read, both in its content and the process under which it was developed.
Given the politically-fraught history of the NBN, the terms of reference for the Vertigan Committee required them to call for public comment on the conceptual basis and analytic framework (4) to be used for the CBA. This rather novel procedural approach was intended to garner both political buy-in for the process and to ensure that as wide a base of information as possible was collected to set the parameters and inform the sensitivity analysis undertaken in the subsequent report. Furthermore, the framework used for the CBA was developed by Griffith University’s, Dr. Alex Robson, and in standard academic fashion, was reviewed by Professors Nicolas Curien (Conservatoire National des Arts et Metiers, Paris), Kenneth Flamm (Lyndon B Johnson School of Public Affairs, University of Texas, Austin), Jonathan Pincus (University of Adelaide) and Cliff Winston (Brookings Institute). However, these procedural safeguards do not appear to have satisfied certain critics, who have damned the August 27 report on the basis of the supposed bias of its ‘independent’ members (5) – notably academic and consultant Professor Henry Ergas, whose expertise in both the theory and practice of CBA had led him to publicly criticize the lack of such supporting analysis for the NBN. The Vertigan critics also appear to have problems with the committee including some inputs from NBN critics – perhaps because, like the tailors who sewed the Emperor’s new clothes, they know that the articles of faith upon which the NBN was constructed cannot withstand empirical scrutiny.
After reading the full Vertigan report, it is hard not to conclude that the critics’ ad hominem attacks are because, no matter how the numbers are sliced and diced, the FTTH NBN comes out as a costly waste of taxpayer funds. Indeed, doing nothing – that is, the government leaving the private sector to manage the rollout of fast broadband to the vast majority of Australians, with government subsidies applying only for deployment of infrastructure in non-economic rural areas – is the most cost-effective approach. Starting from the current point in time, the analysis finds that, compared to a base case of an unsubsidized roll-out, there is a net cost to Australia from both the down-scaled MTM and the original FTTP (Table 1). The net cost of the FTTP network is $20 billion – nearly four times the cost of the MTM at $5.7 billion. Private sector-led deployment of fiber delivers the greatest net benefits to the Australian economy (Chart 2).
Furthermore, the sensitivity analysis shows that the more modest MTM outperforms the NBN/FTTP option over a very wide array of parameter settings. For example, FTTP will outperform MTM only if the willingness to pay (WTP) for ultra-fast broadband increases by around 250% over 10 years (that is, 13% per annum for each of those years). The reason is that the MTM scenario allows faster rollout of improved speeds on existing technologies, leading to immediate benefits for those valuing speed highly. In comparison, FTTH will take much longer, and it crowds out incentives to do anything other than decommission existing copper and cable infrastructure. Indeed, under very high increases in WTP, FTTH performs worse than MTM, simply because of the lost benefits from those who could have used the faster technologies but cannot if they have to wait for an FTTH connection (Chart 5).
The report also notes the effect of the declining value of additional broadband speeds under all methods. Importantly, it recognizes that growth in bandwidth demand is not the same thing as growth in demand for data traffic. It is not axiomatic that increases in demand for traffic require increases in bandwidth if the applications using that traffic do not require the higher speeds offered by FTTH. Whilst traffic may grow rapidly (and most likely at more than 13% per annum), willingness to pay for more speed (bandwidth) rapidly drops off at speeds above 40 Mbps, and is not expected to grow at all for speeds greater than 50Mbps downstream and 9 Mbps upstream for residential users unless new residential applications requiring these speeds become available. As most of the high bandwidth applications cited as necessitating the NBN (e.g., health) are for business purposes and not residential, it is not expected that the residential WTP growth of over 250% over ten years sufficient to allow FTTP to dominate MTM will emerge. Thus it is most unlikely that a FTTP network would ever have been more cost-effective than the alternatives.
Moreover, the MTM offers an option value not present in the FTTH network that renders it more ‘future proof’ than FTTH. This is especially important if demand for fast broadband grows more slowly than expected, as it avoids the over-early sinking of high fixed costs in an under-deployed network. At the same time, it does not close out the potential to accelerate the FTTP build if demand reaches the turning point earlier than expected. These benefits are not included in the CBA because they are difficult to value, but are nonetheless real.
The report thus provides a real challenge to the Vertigan (and Turnbull) critics – who still cling to the ‘faith’ that new high-bandwidth applications will emerge out of the ether, and that burgeoning demand for video streaming (i.e., data traffic volume) is a sufficient justification for scrapping copper and cable for a ‘future-proof’ fiber network. Malcolm Turnbull’s MTM is still more costly in net terms than ‘doing nothing’, but will ensure that ultra-fast broadband will be delivered faster and more equitably to Australian residences than if left to the private sector alone. It proposes that Australians will overall be better-served than citizens in other countries where government funding is less generous. The report challenges them to show why short-term benefits of faster speeds for those who value them should be sacrificed for a particular (government-owned) network vision. In Australian politics, the NBN model owes arguably more to a desire of legislators to break the market power of a troublesome telco than to principled policy for industry development. (8) The Vertigan CBA reveals that a $20 billion net disbenefit to the Australian economy is the price one government was prepared to pay to renationalize the fixed line telecommunications industry.