web analytics

Measuring Broadband New Zealand Report May 2020

Demand was off the scale. Yet New Zealand’s fibre and copper broadband networks almost never skipped a beat during the 20200 Covid-19 lockdown.

Both technologies did well enough to earn a tick from the Commerce Commission

To no-one’s surprise, fixed wireless broadband did not fare as well. It was never going to.

The Commerce Commission uses UK-based SamKnows to track broadband performance. The Measuring Broadband New Zealand Autumn Report, May 2020 shows there was no significant decrease in download speeds on fibre or copper networks.

Fixed wireless speed drops

SamKnows reports average fixed wireless download speeds fell by around 25 percent.

There is a simple reason for the difference between the technologies. On a fibre network a customer has a direct line from their connection point back to the local roadside cabinet. In effect the same happens with copper.

Fixed wireless broadband users share their connection with others. That means speeds drop at busy times. During the lockdown the networks were busy for a lot more time than would otherwise be normal.

Latency

On a similar note, the report found fibre responsiveness consistently outperformed all other technologies. Fibre latency came in at under 20ms 90 percent of the time. This is important for applications like Zoom video calls and makes a huge difference to the online gaming experience.

The report says:

“Fixed Wireless connections will be more likely to experience issues with latency-sensitive applications such as online gaming or video calls.
”

SamKnows reports that games hosted in New Zealand had lower latency than games hosted overseas.

Fibre to the max

Telecommunications commissioner Dr Stephen Gale says: “Chorus and other providers reported record levels of online activity. But despite that increase, the latest report from our independent testing partner, SamKnows, shows that copper and Fibre 100 plans continued to perform well, with average download speeds unaffected.

“We’re pleased to see that Fibre Max speeds have again increased, but there is still a significant variation of results on these plans. We are working with the industry to understand the causes of this, which involves looking at hardware and the performance of individual networks.”

The report, published by the Commerce Commission, notes the average download speed of Fibre Max plans, that is gigabit wholesale fibre, increased by around 50Mbps since the earlier report. The jump is down to improved performance by a single service provider.

A medium-sized consolidation play for the Australian-owned telco.

Vocus will grow its customer base by 10 percent when it absorbs Stuff Fibre. The Australian telco picked up the media company owned ISP for an undisclosed sum.

The deal will add roughly 20,000 broadband customers to Vocus’s existing 200,000 customers.

Vocus remains New Zealand’s third largest internet service provider behind Spark and Vodafone. After the deal it will have around 13 percent of the market.

Stuff Fibre is part of the Stuff media group, which is also up for sale.

Unusual ISP

It is unusual in the telecommunictions sector because it is a media company with a broadband business. More commonly broadband companies have forayed into media with projects like Spark Sport and Vodafone TV.

The other unusual aspect of Stuff Fibre is that it is, in effect, a virtual ISP. Stuff Fibre doesn’t own and operate its own infrastructure. Instead it buys these as a service from Devoli.

This was a smart strategy for a business with a recognisable brand, but without in-house broadband expertise.

Stuff Fibre’s virtual nature will make it easy to integrate into the Vocus business. That’s often a sticking point with broadband company acquisitions. Some well known broadband brands remain famously unintegrated years after acquisition.

Vocus New Zealand chief executive Mark Callander
Vocus New Zealand chief executive Mark Callander

The ISP was, in its own small way, innovative in other ways. Mark Callander, who heads Vocus New Zealand made a point of mentioning the in-home Wi-Fi troubleshooting services. He says he plans to make this available for Vocus’ existing customers.

He went on to suggest other acquisitions could be in the pipeline. For years industry observers have expected to see consolidation among ISPs. New Zealand is significantly oversupplied with service providers, many other markets only have a handful of players, there are 90-odd here.

It’s possible the Covid–19 disruption could trigger a wave of more acquisitions, but don’t hold your breath. The shape of the market has been consistent for years now with more new entrants joining than mergers taking brands out of contention.

Callander also says Stuff Fibre staff to will transfer to Vocus.

Ultrafast Fibre’s owners have agreed to sell their shares in the business to First State Investments, an Australian asset management company.

The business was owned by electricity distributors, who bid and won the UFB contracts over a decade ago. WEL Networks Limited owned 85 percent and Waipa Networks Limited controlled the other 15 percent.

UFF runs fibre networks in parts of the Central North Island including Hamilton, Tauranga, Whanganui and New Plymouth. The network runs past around 237,000 premises.

First State paid $854 million for the business.

OIO approval needed

The deal is subject to approval from the Overseas Investment Office.

This is where the story could get interesting.

An overseas organisation needs OIO approval to buy a sensitive New Zealand asset. In most cases sensitive means either land or a business that is worth more than $100 million. There’s usually an exception for Australian buyers.

However, a fibre network isn’t only sensitive. It is also a strategic asset of national importance.

Different rules for different fibre companies

Chorus, the largest fibre company is subject to a 10 percent Kiwi Share restriction which limits foreign investment in the business.

In 2012 the government had to decide to waive the restriction when AMP Capital moved to take a larger stake in the business.

There was something of political fuss when this happened, yet the amount of Chorus under discussion was less than the amount of UFF that’s now up for sale.

While UFF is much smaller than Chorus, it is New Zealand’s second largest fibre company. It has a wholesale monopoly in its area and covers roughly 12 percent of the homes able to connect to UFB.

So, if roughly five percent of Chorus triggered political alarm bells, 100 percent of UFF could come under scrutiny.

Exquisite timing

First State Investments is lucky that most politicians are focused elsewhere at the moment. Either that, or it timed its acquisition knowing it could slip under the radar.

If UFF was anything other than a technology company you could expect, at least, New Zealand First to take an interest in this sale.

Of course this wouldn’t be the first time overseas interests have owned strategic New Zealand telecommunications assets. Yet it seems odd that Chorus ran into problems selling a few percent to AMP while UFF can be sold in its entirety.

New Zealand’s broadband has performed well during the Covid–19 lockdown.

Our networks have been pushed well beyond expectation. For the most part, they have not been found wanting.

That’s the good news. The bad news is that too many New Zealanders are still on the wrong side of the digital divide.

Digital divides

Make that digital divides. There is more than one.

Not every household can manage the cost of a fast broadband connection or afford the devices needed to make it sing and dance.

That problem is structural, social and economic. It is beyond the remit of our telcos.

No-one can sensibly argue that New Zealand’s broadband is overpriced. It’s cheap by international standards. More so when you consider the size and distribution of our population.

Competitive

Broadband competition is so tight that telcos only make the flimsiest of margins each month. Some argue, with justification that the market is too competitive. There’s little room for them to sharpen the pencil.

One obvious fix for this is to ensure people have more money to spend on necessities like broadband. That’s not going to happen overnight the way things are now.

The other approach is to subside services for less well off customers. That’s largely down to politics.

Some great initiatives are in place to do this. We need to make sure they are taken up and, if they are not enough to meet needs, then ensure there is more money invested here.

Investment

Yes, invested is the right word. Homes with broadband have better access to education, but they also can access government services.

It is cheaper for government organisations to communicate digitally than to do so in person or by moving little packets of paper around the country.

Other households are on the wrong side of the digital divide simply because happen to be in the wrong place.

A link too far

They may be beyond the reach of fibre. If they have copper they could be too far from a cabinet for VDSL.

They may be in a RBI fixed wireless area but the local tower is full. They may be some distance from an RBI tower where performance is woeful.

We need to fix this and fix it fast. The clock is ticking. Children can’t leave their education for years. Working from home will remain important. Fast broadband is not a luxury, it is the digital equivalent of daily bread.

Fibre further

The fibre network needs to reach further into the bush.

Yes, it is expensive to connect remote homes to the network, the alternative is to accept second class citizens in our own country.

The only way to fix this is with government money. We paid for the original UFB network with a soft loan. To extend the network requires cash. It would take too long for a loan to pay back at current prices.1

Return on investment

The good news is that state investment in broadband pays a decent return.

If the idea of state control bothers you2, we don’t have to go down the Soviet route. There are plenty of opportunities to work hand in hand with local telecommunications entrepreneurs. This seems to be the preferred approach of New Zealand’s two main political parties.

There’s a point where the big telcos lose interest in extending broadband networks. That’s understandable. They are money-making businesses, not charities.

Wisps (wireless internet service providers) do a great job and they understand local needs and conditions. There should be more soft loans and subsidies to help them push broadband further up the remote valleys.

Good work

Hats off to all the service providers, not just the Wisps. They are a credit to the industry and to the nation. Spark, Chorus, Vodafone, Enable, Northpower and UFF have all reported on their performance. The smaller players may make less noise but their efforts are also appreciated.

In some cases the performance has been astounding.

On Friday Chorus reported the average fibre customer now gets through almost 500GB of data a month. The actual figure is 495GB. That’s up 30 percent of pre-lockdown consumption in February.

When you add in the customers on the copper network, the nationwide average drops to 406GB, an increase of 36 percent on pre-lockdown use. Clearly the old copper network is enjoying a new lease of life, even if in the long term it is on the way out.

It says the average speed on its network is also up to 150Mbps. That’s because more people have switched to higher speed fibre plans. Gigabit broadband is especially popular today and hyper fibre is on the way.

Fibre is king

At the time of writing around 80 percent of the country can connect to the fibre network. That will rise to 85 percent by the end of 2022.

The fibre uptake level before lockdown began was somewhere between 50 and 60 percent. So in very round numbers, almost half of New Zealand now has fibre.

All fibre companies report an increase in demand and there is a post lockdown level 4 backlog of orders to work through.

That’s huge vote of confidence for the decade-old UFB programme. It was originally started with around $1.5 billion of government money in the form of a soft loan3.

Late last year I interviewed Sir John Key and Steven Joyce who were the two ministers responsible for the original plan. Both said then that, in hindsight, the UFB was one of the best investments a government has made in recent times. That was before the pandemic. Today, that investment looks even better.


  1. Either that, or we accept that more remote fibre customers pay a premium to cover the higher cost of getting a service. This goes against the grain of the one-price-for-everyone approach of UFB, but it would speed things up. ↩︎
  2. There are people who would prefer a return to state-owned and operated telecommunications. That would require a political earthquake. If the Covid–19 pandemic doesn’t change this, there’s little prospect of it happening. ↩︎
  3. A small fraction was set aside for schools broadband, leaving about $1.35 billion for fibre. ↩︎

It’s time to consider whether New Zealand’s fibre network should be extended beyond the reach of UFB and UFB2. There is a sound business case for doing so. Yet considerations need to go beyond the balance sheet.

New Zealand’s fibre network has more than proved its worth since the nation went into lockdown. It helps hundreds of thousand to work or study from home. It keeps people entertained when other forms of fun are restricted.

There’s more than enough network capacity to cope with increased demand.

While there is still a digital divide that needs to be addressed, the UFB has made fibre affordable for most people. Unlimited data plans with gigabit speeds start at around $80 a month. That’s all the broadband an everyday user could wish for at a knock-down price.

UFB1 and UFB2

The first phase of the UFB programme finished late last year. It gave three quarters or 75 percent of the country the opportunity to install a fibre connection. The second phase, UFB2, scheduled to complete at the end of 2022 extends that footprint to reach 85 percent of the country.

This leaves 15 percent of New Zealand having to rely on either fixed wireless broadband, copper networks, or, in more remote cases, satellite broadband.

All these technologies have improved in recent years, but they can’t match fibre in terms of price performance.

All the way?

There are people who argue fibre can go all the way. After all, the copper network reaches around 99 percent of the nation. We built that at a time when there was less money than today. It was seen as a nation building exercise. There was also an element of investing in infrastructure to create jobs.

We could go down that path. There is a case. It would be expensive and take a long time.

Yet it isn’t necessary. Wireless technologies are better suited than fibre when it comes to connecting more remote homes and businesses. And for the most remote places, satellite will remain the smartest choice.

Let’s agree now that the last one percent is only going to have satellite as an option. This includes places like the Chatham Islands. Building fibre to the remote-windswept-farmhouse there would be economic madness.

Cut off point

This leaves us with the question of the best cut off point between fibre and fixed wireless.

We can leave it at the 85 percent mark, which is where it will be at the end of 2022. But there’s a strong case for connecting those people who live just beyond the fringe of that network, and there will be again if we extend it further.

On the other hand, pushing to 99 percent doesn’t make sense.

The actual cut-off point is more about economics than technology. There’s nothing technical to stop us building more and more fibre.

The barrier here is economic. By the time we get to the 85 percent network level in 2022, government and private investors will have tipped something like $6 billion into fibre.

Something like the 80:20 rule comes into play with real world network building1. In very round numbers, if it cost $X billion to connect the first 20 percent or so of premises, it would cost 2 x $X billion to connect the next 20 percent and 4 x $X billion to connect the next 20 percent. The UFB2 programme kicked in with the next 10 percent at a network cost of around $400 million. That’s on top of the $1500 to $1700 cost of each additional connection.

Costs

Or, to put this another way, adding easy to connect homes in dense inner city areas costs a few hundred dollars per connection. Adding the last home that would bring the connection level up to 99 percent would cost many millions.

Which all means there is a pay-off curve.

At this point the discussion gets a little more complex.

When the UFB was originally planned, the politicians and experts hoped take up would reach 20 percent. Connecting 75 percent of the population when the expected uptake was 20 percent made a kind of economic sense.

Uptake better than plan

As UFB developed and uptake rates outpaced the original expectation, it made economic sense to increase the footprint. The economic viability of the more marginal connections changes when the uptake rate is 40 percent compared with a 20 percent uptake rate. That was when the decision was made to build UFB2.

Today’s uptake rate is a little higher than 50 percent. By all accounts there is a huge backlog of UFB connection orders. I expect to see overall uptake reach at least 60 percent within a year.

At an informed guess, the uptake will carry on climbing from today’s level, albeit at a slower pace than in the past. Over time it will definitely reach 70 percent without any external intervention.

We’ll come back to that in a moment.

So, if uptake is 70 percent, it makes economic sense to connect more of the homes in the last 15 percent of the nation. That is the people who won’t be on the network at the end of UFB 2.

This applies even though the per-connection cost of extending the network is higher than for the first 85 percent of the country.

The copper question

All this talk of uptake rates assumes people can choose to keep copper broadband connections in areas where fibre has been installed.

We need to challenge that assumption. Apart from anything else, it costs money to run two networks. Fibre is easier, less expensive to maintain than copper. So, over time, the cost of maintaining a single copper telecommunications line will be higher than its economic worth.

It makes sense to mandate a move to fibre and rip out copper before we get to this point. No-one can claim this would be anticompetitive. After all, those fixed wireless broadband towers aren’t going anywhere. And soon satellite will be a more competitive proposition2.

When the copper network is pulled, we can expect fibre uptake rates of 90 percent or higher. This makes connecting those otherwise more marginal premises in the last 15 percent an even better economic proposition.

The existing rules allow network company Chorus to withdraw copper lines from fibre areas. As things stand, there isn’t much incentive to do this.

Not only economic

Most of this discussion has been about the economics of extending fibre. That’s to underscore that the idea has a sound commercial basis. Yet, as it says at the top of this post, there are reasons to extend fibre beyond the mere economic.

To use an old fashioned cliche, more fibre would be a nation-building exercise. Those people who now live in areas beyond the planned fibre footprints won’t feel like second class citizens3.

These are people who need remote working, they can’t easily catch a bus to their nearest co-working space. Likewise school age children might not always be able to reach classrooms in poor weather. Communications like video-conferencing is more important when you can’t catch over coffee on Ponsonby Road.

When the government first announced its plan for a fibre network, Conor English, then at Federated Farmers, pointed out that rural businesses drive the New Zealand economy, yet were left out of the best telecommunications options.

Fibre to the bush, taking UFB2 further

Which brings us back to the cut-off point. Just how far should we extend UFB into the bush? Can we go beyond UFB2 with a UFB3?

Weighing up all the economic arguments, it seems there’s a case for going well past 85 percent. The last five percent doesn’t make economic sense, at least not if customers pay the same prices as other UFB users. We can talk about subsiding their connections, that’s another debate, maybe another blog post.

The slider runs from 85 percent to 95 percent. You can move it up and down, the higher it goes, the more costly things are. Keep it at 85 and there is no more expense, just a lot of people the wrong side of a digital divide.

We should move the slider closer to 95 percent than 85 percent and, as a nation, we absorb the cost. We’ll get all that money back in the long term.

In the past this would have been a nice-to-have, a luxury. Suddenly there is a lot less money in the economy. But in our new world, the idea of extending fibre is far from a luxury, it’s a necessity. We need to find creative ways to make it pay, but that’s something we’ve always been good at.

Let’s extend the fibre network beyond UFB1 and UFB2.


  1. Or at least it does in New Zealand. Extending the network in, say, Singapore, would be a snap by comparison ↩︎
  2. One of these days I’ll get around to explaining this in a separate post ↩︎
  3. This was particularly acute during the Rugby World Cup where fans in the rugby heartlands had second rate coverage. ↩︎