On 2 November BusinessDesk published my article on the politics of pandemic traffic lights:
At the time I thought that the traffic lights system had been initiated by the Ministry of Health (experts outside the Ministry were not supportive). Subsequently, however, according to senior Health Ministry officials privately, it came from the Prime Minister’s department.
This helps explain the working it out as you go along approach that is causing confusion among many. Jacinda Ardern’s claim of the system being world leading is overcooked.
The traffic lights system, commencing on 3 December, was described by me as a form of branding for a new narrative to justify the Government’s surprise abandoning of the elimination strategy towards the delta variant of Covid-19.
I attributed the abandonment decision to government over-confidence over the effectiveness of the progress towards elimination of delta. Despite contrary advice from its modellers, this led to the earlier than expected lowering of Auckland’s alert level from 4 to 3. This was followed by political panic when, surprising to government but unsurprisingly to experts, daily infections quickly surged upwards.
My analysis was based on the data and the absence of evidence to confirm an alternative credible explanation. Now Newsroom investigative journalist Marc Daalder has published an excellent article (29 November) based on official documents about the decision-making process that reinforce but provide more context to my my conclusion:
20 September and 4 October – what went wrong
In his words, Daalder describes “what went wrong” between 20 September and 4 October when New Zealand went from being on the verge of eliminating Delta to admitting defeat.
On 19 September Director-General of Health Ashley Bloomfield optimistically advised Covid-19 Minister Chris Hipkins that infection numbers in Auckland “were falling steadily” and almost all of the clusters had been ringfenced. This was just over a month after the whole country had moved to Level 4.
Bloomfield was confident that the outbreak in Auckland was contained. Peaking on 28 August, “…daily numbers [were] generally decreasing or remaining static at low levels as the expected tail of the outbreak manifests.” Out of the 19 known clusters, all but two of were considered dormant or contained.
The two clusters of concern were “complex and need to continue to be monitored closely, particularly with respect to welfare and social compliance issues”. However, this did not deter Bloomfield from comfortably recommending a move to Level 3 for Auckland (and from Level 3 to 2 for the rest of New Zealand).
At this point the Director-General and Cabinet were expecting a long delta tail but was confident of returning to zero cases. Further, they expected the elimination strategy to continue after border reopening in 2022 (remembering that elimination is zero tolerance, not zero cases).
Two weeks later “Cabinet threw in the towel on elimination.”
Bloomfield’s triumphant paper was taken by Hipkins to Cabinet the same day. The Minister did concede to Cabinet it was “theoretically possible” that elimination might not be achieved. However, elimination would continue to be maintained into the future including in a new protection framework that was being considered by officials.
So the objective was to eliminate the Auckland outbreak and then move to the new framework once vaccinations were sufficiently high.
As an afterthought it was noted that officials were also working on a transition scenario where the current outbreak was not eliminated and there was a gradual move into the new framework while vaccination rates increased.
But everything pointed to a diminishing outbreak that was fully under control. Cabinet believed delta elimination was at its “fingertips”. Consequently it was confident about lowering Auckland to Level 3. After all, this is what had happened in Australia outside New South Wales and Victoria.
Ignoring red flags
But Daalder points out that there red flags were raised. The Ministry’s Director of Pacific Health warned a parliamentary select committee on 28 September that the virus was settling into “a gang environment and the homeless” which were less likely to be trusting of the health system.
Daalder correctly observes that if, as it appeared, the albeit declining daily infections were in transitional housing and gangs, it should have suggested possible more widespread and undetected transmission.
Separately there was also a warning from an independent advisory group chaired by Sir Brian Roche in a 23 September letter and report to government. While focused more on safely reopening the country’s borders, Roche also highlighted were shortcomings in the outbreak response. These included a shortfall in proper engagement of Māori and Pacific providers.
Cabinet and the Director-General had forgotten their previous powerful and evidence-based message to the public that people movement is critical to eliminating, containing or spreading Covid-19. Level 4 was about minimising people movement as much as practical; Level 3 allowed for a reasonable amount of increased people movement.
From optimism to pessimism: a new narrative
Unfortunately the Government was slow to realise what was happening. On 29 September, five days before the Prime Minister ambiguously abandoned it for Auckland, her Director-General told Newsroom that zero cases were still possible. That same day the unexpected shuddering spike of 45 infections occurred.
In the subsequent days before 4 October, Bloomfield’s narrative changed to a pessimistic tone. He referred to the risk of South Auckland sub-clusters not being contained, delta’s circulation in communities that face complex socio-economic issues, and possible slowing down of contact tracing.
The rest as they say is history. On 4 October Ardern was ambiguous about the future of elimination in Auckland. Presumably her ambiguity was because she and her cabinet colleagues were still trying to work out how best to publicly explain the sudden reversal. Its abandonment was confirmed seven days later.
Falseness of public non-compliance
Daalder makes several other pertinent observations. These include increasing pressure by some businesses to lower to Level 2 (the Government’s response was to gradually reduce restrictions under Level 3), increased daily infections were greater than Government expected, increased sub-clusters and unlinked cases, public health teams were “run into the ground”, and a workforce that was “stretched, tired and fatigued.”
He then addresses the claim now being made for the first time by Hipkins and others that public sentiment may have been turning against the lockdown. He concludes that the evidence for this claim was “thin”. Polling from the Government at the time found common emotions had gone from “neutral” and “joy” in July to “neutral” and “sad” in September, though “neutral” was still the most common.
Further, actual compliance had increased within Auckland. Aucklanders were more compliant than the rest of the country when it came to staying home when sick and to using the tracer app. There was also broad support for masks.
Following its abandonment of elimination in Auckland, the Government claimed that lockdown breaches were behind the decision to lower its alert level and the subsequent delta upsurge. But, according to Daalder, the evidence says the opposite. Instead, lowering Auckland’s alert level thereby increasing people movement was largely responsible for the upsurge.
If the Government had stuck to their modellers advice Auckland would have been in Level for two weeks or so longer but most likely out of Level lockdown by the end of October.
As a result much fewer Aucklanders would have been infected and hospitalised (to a lesser extent fewer deaths). There would have been much less stress and pressure on hospitals and the health workforce.
Do I feel joy in this vindication of my analysis published by BusinessDesk? Absolutely not! Am I disappointed by the Government’s duplicity in obscuring its error? Very much so!