When talking about what he faced as the minister responsible for NBN, Prime Minister Malcolm Turnbull was fond of recalling a joke about asking for directions from a pub in Ireland and being told “I wouldn’t start from here”. It is a jab at what he believed to be the deficiencies of the former Labor government’s broadband policy, but telecommunications policy has been a problem for governments dating back even to the Hawke era.

One of the most popular theories around why the Coalition was opposed to fibre-to-the-premises (aka fibre-to-the-home, or FTTH) for the National Broadband Network is that it would render the Foxtel pay TV network obsolete. While it is probably a bit much to blame only Foxtel for the change in NBN policy after the last election, pay television’s link to Australia’s broadband woes goes back a long way.

The Hawke cabinet papers of 1990 and 1991 reveal how telecommunications policy in Australia started to go all wrong.

In 1990 and 1991, while Foxtel was just a glint in Rupert Murdoch’s eye, the Hawke government was finalising the introduction of a second telecommunications carrier via the privatisation of Aussat (now known as Optus) and the introduction of pay TV services in Australia. Ultimately pay TV in Australia would be delivered two ways: through hybrid fibre-coaxial cables installed by Telstra and Optus, and through satellite services to areas outside Sydney, Brisbane, and Melbourne.

In a cabinet discussion paper prepared by the Department of Transport and Communications on the delivery of pay TV services in 1991, the department argued against the very HFC cable NBN is now buying, suggesting fibre-to-the-home (i.e. Labor’s model of the NBN) would be vastly superior, even in the early 1990s.

“FTTH is the technology for next century. Recent advice from Telecom [the government-owned company that would eventually become Telstra] is that they do not expect the economics to be right to commence installation to homes until 1997 [The Prime Minister argues the economics are still not right for it in 2016].”

The cost for a fibre rollout is, even with adjusting for inflation, significantly lower than the current estimates for either the full fibre-to-the-premises NBN or Prime Minister Malcolm Turnbull’s “multi-technology mix”. According to the discussion paper, the rollout for “homes in urban areas” would be $5.3 billion, and could be rolled out to between 3% and 10% of homes every year. HFC could be rolled out faster, but the paper states that it would “still be expensive and be made obsolete by FTTH”.

The internet, not being what it is today back in 1991, was not a major factor in the reasons for rolling out FTTH. The discussion paper suggests if FTTH were to be built, then there would need to be other applications like “home shopping, banking, and security” to make it worthwhile.

The report was fairly scathing on the HFC networks that Optus and Telstra eventually would go on to build, and NBN would ultimately buy from them:

“A hybrid network could be started now but it would be obsolete before the end of the century and could be a one-way system that would not significantly add to the development of Australia’s telecommunications infrastructure.”

A parliamentary committee at the time suggested that Telecom (Telstra) could be the company to build the network, and that if it did, then it would not be able to be a pay TV operator, but would instead offer other companies to sell pay TV over the cable. This is how NBN operates today in not selling broadband services directly to consumers.

It should be noted that no one decision in this period can be blamed for the mess that telecommunications policy has been in Australia for the past 30 years. This discussion paper was well before a series of poor policy decisions (privatising Telstra as a single company that is both a wholesale network operator and a retail service provider was probably the worst) throughout the ’90s and into the current century from both Labor and Liberal governments that ultimately led to the Rudd Labor government deciding to pursue a government-owned and built fibre-to-the-premises network from 2009.

The Coalition argued before the last election that advances in technology for new uses of legacy copper networks, from the ADSL broadband connections we use today to the VDSL connections over fibre-to-the-node, have meant that the economics to upgrade the existing networks rather than rolling out brand new fibre networks make more sense, and could be done faster. In government, however, switching to a fibre-to-the-node and HFC network has been slow progress with most of the NBN still consisting of fibre-to-the-premises connections, and an estimated $8 billion added to the cost of the network just to switch from Labor’s policy.

NBN has also rejected a recent analysis suggesting fibre-to-the-premises would be ultimately better value for the company than fibre-to-the-node.

The 1991 report predicted that in 2010 Australian homes would have access to broadband providing “high-capacity communications services” including HDTV, video telephone, home shopping, and energy management”. Most of us are still waiting for that.