BACK IN BUSINESS BURSTING THE BUBBLE?
Fretting about the trans-Tasman bubble? Keith Lynch explains why the data suggests there may be more danger closer to home.
The trans-Tasman bubble is officially open, bringing an unprecedented period of national solitude to an end. From today, New Zealanders can travel to and from Australia without the need for isolation. The thing is, people can now avoid our first line of defence – the managed isolation and quarantine facilities (MIQ).
This may well have some New Zealanders worried. While the world has been ravaged by Covid-19, New Zealand has, for the most part, continued as normal. We’ve avoided lengthy lockdowns. Shops, bars and restaurants have mostly stayed open. Kids have stayed in school.
We’ve been an island of normality in a choppy sea of chaos. Some may well be asking: is the bubble worth the risk? Yes, there are obvious economic benefits, but what if Covid19 creeps in, carried by a traveller who doesn’t need to undergo a twoweek stint in a hotel room?
To explain the risk of Covid-19 making it across the Tasman, Michael Plank, a professor in the School of Mathematics and Statistics at the University of Canterbury, and principal investigator at research centre Te Pu¯ naha Matatini, ran some numbers.
Over the last year, Plank has created a variety of models to help understand the spread of Covid-19 in New Zealand.
To be clear, the numbers we’ll show below are estimates based on a number of assumptions. This is how mathematical modelling works. You have to plug in some numbers to get a result.
For example, one number needed is: how many people will travel from Australia to New Zealand? We don’t know that, and have to make a best guess.
There is another underlying assumption: the most likely cause of a community outbreak would be a leakage from an isolation facility. This form of Covid-19 spread has been evident in both countries.
How many people will travel here?
It’s really hard to answer this question. According to Stats NZ data, over the course of 2019 about 1.3 million New Zealand residents arrived from Australia. About 1.5m Australian residents travelled here – so a total of 2.8m.
The Ministry of Transport has estimates for the first three months of the bubble, but these have significant caveats.
The numbers are:
■ 203,000 arrivals in May
■ 208,000 arrivals in June
■ 237,000 arrivals in July
The numbers are based, in part, on how many flights have been put on, and historical seat capacities.
The ministry says, ‘‘until the bubble begins, we don’t know what actual demand will be – there are anecdotal arguments for both directions around pent-up demand v caution and travel confidence’’.
Tourism New Zealand, meanwhile, says visitor levels could return to about 80 per cent by next January.
Covid-19 and Australia
In total, there have been about 1000 cases of Covid-19 in Australia in 2021, the majority in New South Wales and Queensland.
But, similar to New Zealand, community transmission is now very rare.
On Friday, Australia had about 150 active Covid-19 cases (with only one acquired locally in the past seven days). To give you some context, there were about 100 active cases in New Zealand’s MIQ.
We have seen Covid-19 arrive from Australia, but it’s been a while. The last case was in early January.
According to an analysis of Ministry of Health data by Wigram Capital Advisors and Stuff, only three cases have originated in Australia since May last year.
How likely is a border leak?
To work out the risk to New Zealand, it’s necessary to determine how likely it is for the virus to leak out of a quarantine or isolation facility.
Earlier this year, academics from New Zealand and Australia produced a study seeking to determine just that. They found, to the end of January 2021, there had been 16 failures in total.
There were seven in Australia (one in Victoria caused 19,800 cases and over 800 deaths) with six lockdowns, and nine leaks and one lockdown in New Zealand.
At the time, the researchers also determined the likelihood of failures in the future. The data suggested 1.8 failures or leaks per month in New Zealand and 0.7 in Australia. If, however, new arrivals had a variant that is more infectious – similar to the UK or B.1.1.7 variant – those numbers could become 2.8 and 1.
Making assumptions
OK, to estimate the risk, we need to make some assumptions.
In this case, they are:
■ We’ll see 80 per cent of the 2019 travel total – about 2.24m arrivals – and those will be equally distributed throughout the year. This might not happen in real life, but it’s a reasonable assumption to make to outline the risk. And it’s broadly in line with the Ministry of Transport’s estimations.
■ There’ll be an average of one border-related outbreak every month in Australia. For each outbreak, there’ll be, on average, four cases who could potentially travel to New Zealand before the spread is detected and restrictions introduced. Remember, this is an average. Sometimes only one person may be infected. Sometimes there might be more.
■ Border workers are regularly tested so that any outbreaks are identified early and travel is quickly suspended or subject to quarantine.
■ These cases are distributed randomly among the Australian population of 25m. Again, this is not likely in real life. But if we don’t make this assumption, the modelling becomes too complex.
■ Nobody is vaccinated against Covid-19. Yes, we know people are vaccinated but, again, if we start adding in another variable, the modelling gets out of hand. The average incubation period, or how long people take to show symptoms, is one week.
What the numbers say
Now we have our assumptions, let’s run some numbers.
In a given week, the average number of Australian community cases (before the outbreak is identified) would be 1. Remember, we’re assuming an average of 0.25 border-related outbreaks per week. Multiply that number by 4 cases per outbreak, and you get 1.
So that means the probability of a random person travelling to New Zealand while incubating the virus in a given week is 1 in 25m (the population of Australia).
From there the maths get a touch more complicated, but when you extrapolate over a year and factor in 2.24m travellers or trips, the probability of one infectious person
travelling to New Zealand ‘‘is roughly equivalent to one infectious arrival every 11 years on average’’, Plank says.
‘‘So this shows that the risk from the trans-Tasman bubble is smaller than the risk from our own MIQ.’’
Remember, things change
There are, of course, things that could change that risk estimate.
Firstly, the number of journeys could be much bigger or smaller than 2.24m a year. Right now, it’s reasonable to assume fewer, and that would bring the risk down.
If there is a major border-related outbreak that is not detected early (such as the August Auckland outbreak), the risk goes up. It’s really hard to predict if that could happen.
There are other factors that increase or decrease the risk of travellers being exposed to Covid-19. For example, if people make short trips or visit areas a long way from Australian isolation hotels, the risk of infection is reduced. But if people travel to and from major cities where the facilities are located, that would theoretically increase the risk.
Finally, there’s the vaccination rollouts. This is important – if more people are covered by the vaccine, it reduces the risk of leakage. But even that won’t be perfect.
Bottom line: the bubble is open. We don’t know exactly how it’ll play out. But it’s good news.