We voted for it and we’re about to get it — what promises to be a huge boost to New Zealand’s broadband capabilities — not just fibre to the node, but fibre to the home.
The government is planning to spend $1.5 billion on fibre, saying it will boost productivity and drive economic growth. But no matter how hard you look, a comprehensive business case for broadband investment is missing.
It is taken as read in business that any significant investment, let alone one of this scale, has to be supported by a business case, including solid return on investment analysis.
But when it comes to broadband, that requirement seems to have flown out the window. That’s a question that has been nagging IBRS analyst Guy Cranswick.
“A liberated telecommunications market is not necessarily the same thing as an engine for growth, but in this debate the two ideas collide. On the one hand making broadband more accessible and better will introduce more people to using it, just like cars and highways, which is often the analogy in this debate. The trouble is the real equation of cost and output is ignored or misunderstood,” he wrote in a briefing paper, “Making sense of Broadband”, published in May 2007.
Growth of broadband users does not correlate to economic growth per se, just as the liberalisation of telecommunications markets is not a catalyst for growth, he continued.
It’s a point TelstraClear CEO Alan Freeth made, to some ridicule from the proponents of government investment, when National announced its broadband policy last year. Freeth suggested the main result of faster broadband may be more downloads of pornography and movies rather than improvements to productivity.
Cranswick questions the methodology of some of the papers that purport to show the economic benefits of broadband investment.
An Accenture’s study from 2001, he says, produced both a high and a low estimate for the economic benefit to Australia of widespread broadband. The high estimate of A$30b (NZ $37b) a year, quoted by the Australian Labor Party, “appears to be a measure of the projected spending on broadband rather than the benefits received”.
Cranswick says it is derived by adding the projected spending on computer equipment per household per year, to the projected spending on broadband subscriptions.
The low estimate, A$12b a year, is based on “Accenture propriety analysis” but appears to be derived by assuming the economy will benefit by one-third of GDP growth, he writes, before turning his critical attention to a New Zealand study.
“The methodology in the analysis quoted by policy makers, and as a backbone for various players in the industry to make decisions, is flawed. A New Zealand study published by the interest group Hi Growth produced this finding: ‘...if broadband diffusion in New Zealand is accelerated to a level of 50 broadband subscribers per 100 of population within five years, compared with the baseline forecast, nominal GDP would increase by NZ$889m by 2010, NZ$5.9billion by 2020 and by NZ$13.1billion by 2030’.”
Cranswick argues the analysis is suspect, as it is based on a “growth accounting methodology that commits the same mistake as the Australian study”.
It includes inputs across 10 variables of ICT consumption and increased broadband adoption rates to produce modelled figures of growth, and the consequent impact on the economy.
“The information produced is the most likely to be favourable to an advocate’s position, despite its conceptual problems,” he writes, before recommending policymakers ensure that research used to formulate policy passes “critical tests of adequacy”, and that they don’t use research prepared by technology companies or other interest groups.
Instead, they should prepare small studies to understand and quantify the effect of broadband at levels of the national economy.
In June last year, Cranswick revisited his criticisms when analysing the supposed benefits of Australia’s troubled National Broadband Network (NBN) project:
“A business case [has] never been put forward to explain and demonstrate how the NBN will produce an economic benefit. The responsible minister has not published any policy, or research to support the policy with empirical data or arguments.”
Cranswick had not studied the New Zealand Institute’s influential 2007 papers on broadband when Computerworld contacted him last week. After looking at these, he had a mixed reaction.
“From page one it’s an advocate’s argument. The potential benefit is unevenly spread across the economy,” he wrote in an email before adding that the “methodology seems prudent and balanced and also adapts [a] range of data, not [a] single source.”
However, he says the institute’s assumption that growth benefits are limited to a five-year period, with zero benefit after this, is a poor one.
The institute says this was done to generate a conservative estimate of the economic benefits of broadband investment, but Cranswick says it helps produce a positive outcome.
“Because there is inevitable positive correlation to growth with investment in the first phase, five years, so this already finds a positive outcome. It means the results are already prepared,” he says.
Cranswick says he knows the Australian studies the institute has compared itself to and they are “quite good”, but they determine positive results without producing a control. In other words, they don’t determine what growth would have occurred without broadband.
Cranswick says other benefits found in the study seem dubious, too.
“The benefits per industry sector are predictable, but unsubstantiated in relation to actual sales and the behaviour of people, for example, tele-working.”
As for the institute’s belief that better broadband service would help small New Zealand firms exploit the “long tail”, Cranswick says the long tail is a fine concept for early digital markets but “mostly there is no profit in it”.
Benefits to exporters to overcome remoteness depend on a buoyant world economy, he says, while internal micro-economic and exchange rate advantages “take ages to determine”.
“The rest makes a case on the basis of the initial argument. It could be better and show the differential effects to the economy. Fibre will come because it’s the next stage in the technology cycle and it will be [as] important to have as paved roads are in developed countries.”
Temple Investment’s Paul Winton, who supported the NZ Institute’s study, says he is still “very comfortable” with its recommendations: that government invest $1b to 1.5b in fibre to the premise for 75% of premises.
Temple says it’s possible some of the assumptions under the model have changed or been deferred given changes in the economic outlook.
“However one would still expect the relative contribution of broadband investment to the economy to be similar. For example remote-working is just as real as it was a year ago, health benefits can still be realised and in the long term education will become more global. There seems little merit therefore in revisiting the numbers in the case,” he says.
Fibre broadband helps overcome New Zealand’s small scale and distance from key markets, he says. That hasn’t changed either.
“Fibre doesn’t guarantee success, it’s not a silver bullet, but without it New Zealand doesn’t stand a chance against other nations in weightless businesses (which still account for half of global growth) and will continue to wallow,” he says.
“National’s plan to overcome these structural impediments shows bold, ambitious leadership.”
Temple says it is impossible to state the value of such investments with certainty.
“In such complex situations it is worthwhile considering the cost of not investing and losing out. By our estimates even if the probability of our estimates being real is only 20% the government is better investing the $1.5b than risking missing out or delaying the benefits.”
Whatever the truth, some at least believe the government’s $1.5 billion promise is still in a state of flux. In its predictions for 2009, research company IDC says the government’s plan for fibre to the premises will continue — but scaled down “to affordable fibre access in central and metropolitan business centres, and ensuring that fibre is a standard part of new greenfields’ developments”.
Analyst Rosalie Nelson says this will, in part, be to avoid overbuilding existing infrastructure, which the government will see is not an economic approach.
“In this economic climate you want to be clear what the objectives are,” she says.
Nelson says the economic crunch has come at a tough time for the industry, with access and phone services under pressure and new investment expected to be paid for by demand for new services. Now that demand is more uncertain.
“That does make the business case very tough,” she says.
What about the faltering world economy?
Given the changes in the global economy over the past year, what has happened to the assumptions on which many infrastructure investment studies are based? IBRS analyst Guy Cranswick says the economy has taken a major blow and this will duly affect infrastructure projects because:
• Capital is more expensive, the cash rate is not so significant here but the availability of capital is harder and thus more expensive.
• Returns based on economic activity are pushed forward by a long way depending on the optimism or pessimism of the economic outlook; that is, whether debt deflation is a factor.
• Future world economic growth in these constrained conditions will be much smaller on the other side of the crisis or downturn, as typically, recovery post-financial crisis is slower or more haphazard
• Economies are contracting and the ‘life blood of a big infrastructure’ is greater economic activity.
• The base assumptions about returns, productivity and long-term viability in these studies will evaporate, if not now, then soon.
However, TUANZ chairman Chris O’Connell says the benefits of broadband investment will accrue through cost savings and the ability of organisations to innovate on fast networks.
“All our members want is to figure out how to do more with less,” he says. “CIOs are being asked to put meat on the promises of IT investment — and that’s about slashing costs.
He says to some extent that is an “advocates argument”, but that has been true of most revolutionary technologies “from refrigeration to railways”.