New Zealand appears to be facing its own productivity paradox. On a personal level, we understand the benefits of Ultra-Fast Broadband at work and at home, but for the economists and researchers who study productivity, identifying significant gains and causal connections is hard.
Take heart, folks. We’ve been here before.
Between the 1970s and the 1990s a similar effect was seen as companies ramped up investment in information technology. Where investment in farm and factory automation resulted in productivity gains of over three percent, investment in IT struggled to register at around one percent.
Economist Robert Solow observed the puzzling effect, and many people suggested there was a problem with measurement. The debate culminated in an eight-page Harvard Business Review essay by Nicholas Carr, in 2003, entitled ‘IT Doesn’t Matter’.
The essay questioned whether the sheer ubiquity of IT meant it no longer delivered a strategic advantage to businesses. Any strategic
advantages, Carr argued, were fleeting and easily replicable. IT mattered a lot to the economy, Carr said, but only when it became a commodity; only when it became basic infrastructure, like railways and roads.
Even before Carr’s article was published, however, productivity growth had begun to accelerate. While not all of this could be attributed to IT investment alone, such investment was almost certainly necessary for it to be realised.
Flash forward to 2017 and New Zealand researchers face a very similar conundrum. Despite their best efforts, productivity gains attributable to Ultra-Fast Broadband are hard to nail down.
One reason for this is because to prove productivity benefits you must prove causation. Finding correlations is easy. Proving causation is hard. Very hard.
This month, for instance, the OECD released what on the surface appears to be a powerful chart showing a strong correlation between network speeds and gross domestic product (GDP) but that doesn’t pass the high bar of proof of causation, says one researcher.
“Unfortunately, these macro correlations really reveal nothing about the relationship between GDP per capita and broadband penetration – other than that there is an association between the two. ”
“Unfortunately, these macro correlations really reveal nothing about the relationship between GDP per capita and broadband penetration – other than that there is an association between the two,” economist and productivity researcher Arthur Grimes says.
“It is highly likely that higher GDP per capita enables countries to roll out better broadband. We have no idea from this type of data whether the converse is true – that is, whether greater fibre also leads to improved GDP per capita.”
Grimes, a senior fellow at Motu Research, the New Zealand research institute, a professor of economics and former chair of the Reserve Bank, has been involved in two studies that found clear causal links between Ultra-Fast Broadband and increased productivity.
Two recent Motu studies, ‘The effect of fibre broadband on student learning’ and ‘Picking up speed: Does Ultra-Fast Broadband increase firm productivity?’ tackled the thorny issue by controlling for multiple other variables that could impact on the results.
In the first study, Motu used the fact that UFB was rolled out at different times in different regions to track changes in educational performance that may have flowed from the roll-out. Education was, after all, one of three key sectors targeted in the UFB plan, along with health and business.
Motu looked at school results for national standards and the NCEA to see if the arrival of UFB had any effect on the performance of schools that received fibre ahead of others in their region.
“What we find is that there is a quite clear, statistically significant improvement in national standards results for primary schools, and that's for mathematics reading and writing, each of the three,” says Grimes.
The effect is small but “highly significant”, he says. The pass rate for the number of students at national standard or above national standard goes up about one percentage point. Exactly which schools benefited most is less clear statistically, but it appears to be decile 1 to 3 schools, which saw a roughly two percentage point gain.
That’s exactly where you would hope to see such an improvement, says Grimes.
Apart from suggestions of improvement in numeracy at NCEA Level One, Motu didn't find much effect at the NCEA level.
Grimes says it was a lot more difficult to find an effect at secondary level because the data is thinner: there are a lot fewer secondary schools than primary schools.
The result is a bit of a first internationally, Grimes says. Offshore studies tend to find no improvement in educational outcomes from ICT-related investment.
As for businesses, Motu’s study of firm productivity found that having fibre broadband by itself doesn't improve firm productivity. A company switching from, say, ADSL to fibre does not become more productive overnight.
“When you think about it, it's not too surprising, Grimes says. “What we did find, though, is that firms that made complementary investments in the ICT area, to make the most of broadband, did improve their productivity.”
The opportunities for such complementary investment are many: staff training; implementing new work practices and management techniques; relocation of some business activities; investment in new capital stock that may make use of UFB; extra research and development, and redesigning processes.
Grimes says these were conscious decisions that had to be made by the firms around the same time as they installed fibre.
“Firms that did these sorts of complementary activities and also had fibre-to-the-door did improve their productivity,” he says.
What Motu didn’t find was any particular effect for different industries. The effect, and therefore the opportunity for improvement, was seen across the board.
“We didn't find high ICT, high-tech industries or anything like that benefiting more than low tech industries,” Grimes says.
Importantly for a discussion specifically about UFB, Grimes says the gains seen for users
“Some firms are getting started on transforming their business processes - this is where the value comes.”
of Ultra-Fast Broadband were much lower than those detected in a 2009 Motu study, where productivity gains of round 10 percent were detected.
“We're not talking about going from dial-up to broadband, which was a big change,” says Grimes. “It was a big change, and we found big effects from that in an earlier study.
“So, we're already talking about people who can access most things with broadband and now they get something better. You would think you'd have to do something on top of that rather than just faster connectivity making you more productive.”
To a significant extent, the UFB investment was also made to support new applications and business models, and these take longer to arrive than fibre.
Other research on the business benefits of connectivity was conducted by Sapere for the Innovation Partnership, in 2014. Using Statistics NZ data from 5589 firms, along with interviews with 76, Sapere found high users of internet services and technology in tourism, professional services, agriculture and retail were six percent more productive than others in their industries.
The researchers also made a “ballpark estimate”, not a forecast or prediction, of a $34 billion gain if companies that were low users of internet services became high users.
In many ways, this result crosses over between the early and later Motu studies in that it doesn’t isolate users of UFB from
users of older broadband services.
Like the recent Motu study on firm productivity, it also notes that complementary investments are key. As Sapere report co-author Hayden Glass says, all the benefits of connectivity come from what you do with it.
“Some firms are getting started on transforming their business processes – this is where the value comes,” his report says.
Glass, a public policy and regulatory economics specialist focusing on technology, says he is no expert on productivity, however, there were many studies showing economic gains from smarter use of technology by firms, and there is plenty of anecdotal evidence of improvements in productivity at the firm level from better connectivity.
Better connectivity can enable far-reaching changes in the way organisations are structured, how they operate and how people work, he says.
“Whether or not these changes are made depends on what firms do with the new possibilities. Connectivity on its own is just one piece of the puzzle. There are lots of other investments required to get the most out of it.”
The Sapere study does not demonstrate that firms’ use of technology demonstrably leads to improved financial returns, says Glass.
“It does show that there is a connection or rather a correlation between these two things, which at least makes sense from a hypothesis point of view.”
At the industry level, as opposed to the economic one, there are still a lot of wrinkles.
Vaughan Baker, group director, government and corporate relations, for regional ISP MyRepublic doesn’t quite give New Zealand’s UFB programme a pass mark – yet. That’s primarily because of concerns about adoption by small- and medium-sized businesses.
Larger enterprise businesses have already received the benefits from UFB, Baker explains.
When the UFB contracts were awarded and the pricing was announced by Crown Fibre Holdings, the price of existing fibre services used by large enterprises were reset downwards. The prices did not fall to UFB levels because they were for dedicated networks, but they did fall.
Small and medium-sized businesses were prioritised in the UFB roll-out and most now have access to fibre. However, adoption rates are tracking behind consumer uptake.
“There's a big requirement on us, as an industry, both retail and wholesale – Chorus and the local fibre companies –
to educate businesses on what UFB is and what some of the benefits are,” says Baker.
The businesses clearly taking up UFB are those with an immediate application or need: technology businesses or those dealing with large-size files, such as architects and engineers.
Incumbents are also working to retain business customers on legacy infrastructure, says Baker. MyRepublic has responded by pushing Chorus' business grade products, Next Generation Access Business or Bitstream 3, and a symmetrical gigabit product.
“The plan is a hundred times faster and half the price. So, why wouldn't you? But even with that we haven't seen tremendous uptake.”
Even traditional PBX systems can be a block to the adoption of new cloud-based technology.
“Many of those PBXs sitting in businesses are old. They may be outside maintenance contracts and everyone's a bit scared to touch them,” says Baker.
“The plan is a hundred times faster and half the price. So, why wouldn't you? But even with that we haven't seen tremendous uptake.”
Internationally, Singapore’s national broadband network is a bit more mature than New Zealand’s, and small- to medium-sized businesses are jumping on board.
“What we're starting to see in Singapore now is nearly every Singaporean is doing something on the side and they are typically using this network to do it,” says Baker.
Start-up communities and start-up hubs are breaking out all over, as they are in parts of New Zealand. There are also new small businesses being run under the radar from residential premises. So, was it is worthwhile? Should we have done this? In 2017 that’s almost a silly question. Quite apart from the fact that rigorous studies such as Motu’s are starting to prove productivity gains are actually flowing from UFB, the total investment required was relatively small.
“It's not a lot of money,” says Glass. “In the scheme of things, it's not a huge investment.
“Would we be better off if we had slower broadband? That's a pretty hard case to make.”
Government confident UFB will drive productivity - but by how much?
If you talk to former Communications Minister Simon Bridges there is little room to doubt that New Zealand’s productivity is set for a boost from the roll-out of Ultra-Fast Broadband (UFB). The only question is: by how much?
Asked about whether the programme was delivering on its productivity promises, Bridges cited report after report to back claims there is a direct link between fast broadband and increased productivity.
He says a report by Alcatel-Lucent found the Government’s initial $1.5 billion investment in the UFB programme and the Rural Broadband Initiative (RBI) was estimated to result in GDP growth of $5.5 billion over 20 years.
“The Government’s investment in UFB also has a flow-on effect in terms of the development of broadband applications,” Bridges adds.
“The use of these applications is estimated to create an additional $32.8 billion in economic benefits over 20 years through significant efficiencies and cost savings in sectors
such as healthcare ($5.9 billion over 20 years), education
($3.6 billion), business ($14.2 billion), and dairy farming
Bridges went on to cite a 2016 report by the Sapere Research Group, which estimated that an additional 10 percent of employees gaining access to UFB in the workplace would
result in a gain of 1.62 percent of GDP. Importantly, though, these are estimates and forecasts.
“A study commissioned by The Innovation Partnership also found that New Zealand businesses that are high users of internet services are 73 percent more productive than businesses that make little use of the internet,” the former Minister says.
That report did not, however, isolate the benefits of UFB from other forms of broadband. It also contained what its authors described as “ballpark estimates”.
“Having access to fast and reliable broadband is critical to boosting our regional economies,” Bridges says.
“For example, telecommunications services are South Canterbury’s biggest growth industry for employment and the third largest contributor to its GDP. The widespread fibre roll-out is recognised as a key factor in supporting this growth.”
In education and healthcare, benefits are also being realised.
“As the UFB roll-out continues, we do not yet know the full value of the benefits it will deliver. But we are seeing strong demand for UFB as the roll-out gathers pace,” Bridges says.