Friday 29 October 2021

Afternoon roundup

The day's worthies:

Protecting the Aristocracy of Pull

Andrew Geddis, at Spinoff, on how the government has been thwarting attempts to end MIQ's Aristocracy of Pull. He couches it in all the usual lawyer stuff like "may or may not have been entirely coincidental", but it's obvious what's going on here.

The entire MIQ allocation system is a corrupt mess. Not bribery corruption. The more typical New Zealand corruption where the rules get written so that the bureaucrats just automatically do what's necessary. Carve out a national interest exception, define it the right way, and the politically favoured just find spaces in MIQ where others can't. 

It's been obvious since the thing started. Remember how, somehow, spaces for horse track workers happened to be in the national interest because, somehow, horse race tracks got picked as shovel-ready projects when the government was scrambling to find things to spend money on? Nobody needed to bribe MBIE to do that. MBIE wouldn't take bribes. But whatever arrangements led to tax breaks for pretty horses and piles of spending on horse race tracks in a pandemic, which may or may not have been entirely coincidental with one of the coalition partners' funding arrangements, that was above the pay-grade of MBIE. Somebody up the chain deemed horses to be in the national interest, so spaces for horse track workers would be found. 

And here's Geddis on how the Government has been insulating the system against legal challenge. Bottom line: if you can't get a spot, and you haven't got political pull, suing the government will be pretty likely to have them just give you a spot rather than let the courts weigh in on the mess that the government is running. 

The primary method of MIQ allocation – first a “fastest finger” approach, then an always resetting random lottery – might be considered unfair at best, discriminatory at worst.

To take but one example, I still cannot for the life of me see how the overall allocation process could regard the English netball team as being more deserving of a MIQ place than Bergen Graham, a pregnant New Zealand citizen whose request for an emergency spot in MIQ was turned down six times. A small part of me wonders if permitting decisions like this represented a form of extra reward for those of us resident in lifeboat Aotearoa. Our hard, collective effort had won “us” a normal life, and so our life would be made as normal as possible – netball games against international competition included. And if that meant our citizens and residents outside of the country suffered, well, maybe “they” just shouldn’t have left in the first place? Out of sight, out of mind, and all that.

Suspicions regarding just how fair and justifiable MIQ’s operation has been in practice are compounded by the way the government has avoided having its allocation decisions scrutinised. On the two occasions that people who were refused emergency MIQ spots took their claims before the courts, previous decisions that they failed to meet the criteria were reversed mid-trial. It may or may not be entirely coincidental that doing so had the consequence of putting the legal action to an end before the High Court had a chance to rule on the government’s actions. And now the government is busy moving the legal basis for making MIQ decisions from secondary legislation into a primary enactment. That this parliamentary move will have the effect of largely insulating the overall MIQ process from being judicially overturned at a time when it is being challenged in the High Court again may or may not be coincidental.

I have sometimes wondered what would happen if someone with the resources to do so just got a private jet, put up double-vax and testing certificates, announced credible plans to self-isolate, announced when his flight would be landing, and came home.  

Thursday 28 October 2021

Who pays sin taxes?

Freshly out in the NBER working papers.

Who pays sin taxes? You probably can make a pretty good guess:

We find that sin good purchases are highly concentrated with 10% of households paying more than 80% of taxes on alcohol and cigarettes. Total sin tax burdens are poorly explained by demographics (including income), but are well explained by eight household clusters defined by purchasing patterns. The two most taxed clusters comprise 8% of households, pay 68% of sin taxes, are older, less educated, and lower income. Taxes on sugary beverages broaden the tax base but add to the burdens of heavily taxed households. Efforts to increase sin taxes should consider the heavy burdens borne by few households.

Results will apply a fortiori in New Zealand. 

They note that, in New York, a 1.75L bottle of vodka might be as cheap as $11.99, including $7.97 in tax - all USD. If that bottle is 40% alcohol, that's 700mL of alcohol. 700mL of alcohol, in NZ, draws $39.64 in excise. And of course GST on top of excise. 

They note that House Democrats proposed doubling federal excise to $2.00 per pack, on top of state taxes with a median $1.78 burden per pack. New Zealand's excise is $1.045 - per cigarette. So $20.91 per pack.

The most heavily-taxed clusters in our sample are the Everything cluster, which comprises only 2.5% of the population yet pays 27% of all existing sin taxes (around $288 per adult per year); and Smokers, who comprise 5.5% of the population and pay 41% of all existing sin taxes (around $211 per adult per year). The groups are quite similar to one another, except that the Everything group purchases a large amount of alcoholic beverages (primarily beer and spirits for 9.4L of ethanol per adult per year or 10.2 standard drinks per week), while the Smokers purchase almost none. Both groups purchase more sugary beverages than any other cluster (157-170L per household per year or about 1.5L per person per week). Both groups also purchase about a half pack of cigarettes per day, but with substantial dispersion across households (the top 20% of households in these groups purchase at least one pack per day). Demographically, these two groups look similar to one another: they are older, lower-income, lower-education, less likely to have children or belong to racial or ethnic minority groups. They are most likely to be between the ages of 55-64 and least likely to be below age 35.21 Because many of these households are in the lowest-income bin (< $25,000 per year), their overall sin-tax burden as a share of income can be significant. It averages 2% of income for the Everything group and 1.5% of income for the Smokers group. The hypothetical SSB tax would add another 0.2%-0.3% of income on top of that.

The usual claim is that while excise is regressive, the health benefits are progressive. Of course, the ones paying a ton in taxes aren't the ones enjoying substantial health benefits.  

Thanks to Paul Walker for the pointer. 

Wednesday 27 October 2021

Afternoon roundup

The closing of the browser tabs brings the following worthies:

What sucks least - ICU edition

Every option is terrible. Which is least bad?

It seems impossible that the hospital system will not wind up having to ration care as Covid cases rise. 

Under standard operating procedures, elective surgery is first to go when there are pressing cases coming through the door whose lives are at risk. Electives are scheduled, and can be delayed: at cost to those whose surgeries are delayed. Their lives are worsened, or shortened, or both. 

Every rationing option is bad. Which is least bad? I put up one at Newsroom, now ungated. 

Lots of folks on twitter seem to agree it is terrible. I agree with them that it is terrible. I haven't yet seen a feasible option from them that looks less terrible though. That's the terrible thing about spots where all options are terrible. Just saying one of them is terrible isn't good enough. 

The default track has one distinct advantage. Nobody has to pull a lever to change course. Nobody has to take on that responsibility, or bear that horrible burden. The next bed always goes to the person in most desperate need, and that has a logic to it. But it does come at a horrible cost to those whose conditions are debilitating and painful, but not immediately life-threatening.

There is another option.

It is also terrible. But is it less terrible than what happens if we choose not to decide?

I'd really love to see an actually less-terrible option. Because I really don't like this one. 

But note that a lot of options are "Yes, and" things. 

Get better treatments? Yes! And if there are still shortages, you'll still have to ration. 

Increase ICU capacity? Well, it's a bit late starting in on this one, but sure, absolutely agree. Agreed about it before you suggested it. And if there are still shortages, you'll still have to ration. 

Push vaccines out to places with low vax rates? Absolutely! Yes! And note that I'd said my option should only be invoked where there's been realistic access to vaccines. But when all have had access, if ICU hits capacity, you'll still have to ration.

Here's an option even more awful than mine: wait until the crunch hits the hospitals, realise that the unvaccinated have worse recovery prospects than the vaccinated, and default into the prioritization scheme I've put up but without warning people this train is coming so they don't get off the track. 

Addendum: Here's one hospital's plan for an outbreak. Note how care will be rationed. This is the default path. The plan says who will no longer be able to access care, when all capacity is taken up by Covid. 

Response level two would be triggered immediately after a single Covid-19 positive patient is admitted to the hospital, and would have the patient isolated in a single emergency department room.

The entrance to the hospital would be manned, and anyone needing emergency treatment would be required to phone before attendance.

Stringent PPE would also become more commonplace.

Some elective surgeries and radiology services would be cut, with only the highest-priority appointments undertaken, such as for trauma injuries or cancer procedures.

Response level three would kick in once the hospital reached more than five cases or two cases in intensive care.

At this point, the hospital ramps up its response. A dedicated Covid-19 ward would be established, and the hospital's intensive care capacity would be designated solely to Covid-19 patients.

All non-Covid patients would first be processed via Whanganui Accident and Medical, except for those requiring serious hospital-level treatment.

Hospital staff would also begin repurposing other areas of the hospital in preparation for an explosion in cases.

Response level 4 would be initiated once the number of cases in the hospital topped either 15 in total, or three cases in ICU.

Level four is the DHB's highest level of response.

At this level, the emergency department would be restricted to just Covid-positive patients and non-infected patients at status level 1, such as trauma patients or those requiring resuscitation.

A triage area would be set up outdoors, behind the ED where patients could be assessed based on their needs.

The DHB would also defer cancer treatment at the hospital.

Vax for kids

The FDA's decision on vaccines for 5-12 year olds should be up today. The workings from their meeting of 26 October are here.

I expect that kids in the US will be able to be vaccinated in November.

Last week, I emailed MedSafe to ask where things were in the process here. I'd hoped that they'd have been asking Pfizer to get its data to them at the same time, so that MedSafe might consider things in parallel with the US. We'd then be able to hit the ground running.

Ever the optimist. 

Their reply:

Here is our response attributable to Chris James - Group Manager, Medsafe:

In order for Medsafe to consider granting provisional approval for the Pfizer vaccine for 5 to 11-year-olds, Pfizer must submit an application to Medsafe, which has not yet been received.

We understand Pfizer has submitted an application to the US FDA for Emergency Use Authorisation of their COVID-19 vaccine for use for children aged 5 to 11.

Pfizer does not carve up its data for different countries. Regulators have harmonised requirements on the format and supply of data for applications. It is Pfizer’s decision who and when to submit applications to. We recommend you direct any further questions in this area to Pfizer.

If, and when Pfizer does submit an application, Medsafe will prioritise the assessment of the data for this age group. If Medsafe provisionally approves this, further clinical and scientific advice will be sought from the COVID-19 Vaccine Technical Advisory Group before it goes to Cabinet for a “decision to use”.

And if there is a decision to use, please rest assured, we will have enough vaccines for everyone. We will provide updates on progress.
So when Pfizer gets around to getting an application in to a small country at the far end of the world, Medsafe will start thinking about it. It may go over summer break. Then the Technical Advisory Group has to weigh in. Then Cabinet has to decide. It's a political decision, after all, about whether I should continue to be banned from having my daughter vaccinated.

It would be nice if drugs approved elsewhere were also just here available, without having to jump through Medsafe, then the Technical Advisory Group, then Cabinet. 

In the absence of being able to have my 11 year old vaccinated, it would be great to have a stock of rapid antigen tests around the house. If any of us develop any symptoms, we'd be off for a Covid test. But it can take a while to get results back. A rapid test could help while waiting. 

Alas, those are also still banned. 

Tuesday 26 October 2021

Afternoon roundup

The worthies, on the closing of the browser tabs:

Core business for a central bank

Inflation hit 2.2% in the most recent quarter. Not the annual rate - the rate for the quarter.

Arguably, Reserve Banks shouldn't be making statements about inflation outside of the scheduled monetary policy schedule. 

But it seemed a bit odd that the Bank put out a release on establishing the Māori Bankers Rōpū to coincide with the release of the inflation statistics. 

They came out within minutes of each other, as though the Bank were saying "Yes, the CPI numbers are out today, but here's what we're really interested in." 

New Zealand has an Emissions Trading Scheme that caps net emissions.

The ETS is nowhere mentioned in the report. 

The Executive Summary tells us that "Climate Change is part of our core business". 

It simultaneously tells us that the Reserve Bank has forgotten what its core business is. 

Its core business is keeping inflation between 1 and 3 percent over the medium term while maximising sustainable employment outcomes (which really means just maintaining stable inflation between 1 and 3 percent because the rest is out of the Bank's hands). 

If climate change is core Bank business, because it affects financial stability, is there *anything* that is not part of the Reserve Bank's remit? Housing policy would be the obvious starting point, but why stop there? Freshwater regulation may have financial stability implications. 

The question is whether the ECB, other central banks, and international institutions such as the IMF, BIS, and OECD should appoint themselves to take on climate policy, or other important social, environmental or political causes, without a clear mandate to do so from politically accountable leaders.

Moreover, the ECB and others are not just embarking on climate policy in general. They are embarking on the enforcement of one particular set of climate policies — policies to force banks and private companies to de-fund fossil fuel industries, even while alternatives are not available at scale, and to provide subsidized funding to an ill-defined set of “green” projects.

To be concrete, I quote from Executive Board Member Isabel Schnabel’s recent speech. I don’t mean to pick on her, but she expresses the climate agenda very well, and her speech bears the ECB imprimatur. She recommends
First, as prudential supervisor, we have an obligation to protect the safety and soundness of the banking sector. This includes making sure that banks properly assess the risks from carbon-intensive exposures…
Let me speak out loud the unclothed emperor fact: Climate change does not pose any financial risk, at the 1, 5 or even 10 year horizon at which one can conceivably assess the risk to bank assets.

“Risk” means variance, unforeseen events. We know exactly where the climate is going in the next 5 to 10 years. Hurricanes and floods, though influenced by climate change, are well modeled for the next 5 to 10 years. Advanced economies and financial systems are remarkably impervious to weather. Relative market demand for fossil vs. alternative energy is as easy or hard to forecast as anything else in the economy. Exxon bonds are factually safer, financially, than Tesla bonds, and easier to value. The main risk to fossil fuel companies is that regulators will destroy them, as the ECB proposes to do, a risk regulators themselves control. And political risk is a standard part of bond valuation.

That banks are risky because of exposure to carbon-emitting companies, that carbon-emitting company debt is financially risky because of unexpected changes in climate, in ways that conventional risk measures do not capture, that banks need to be regulated away from that exposure because of risk to the financial system is nonsense. (And if it were not nonsense, >regulating bank liabilities away from short term debt and towards more equity would be a more effective solution to the financial problem.)

...Now you may say, “climate is a crisis. Central banks must pitch in and help the cause. They should just tell banks to stop lending to the evil fossil fuel companies, and print money and hand it out to worthy green projects.”

But central banks are not allowed to do this, and for very good reasons. A central bank in a democracy is not an all-purpose do-good agency, with authority to subsidize what it decides to be worthy, de-fund what it dislikes, and to force banks and companies to do the same. A central bank, whose leaders do not regularly face voters, lives by an iron contract: freedom and independence so long as it stays within its limited and mandated powers.

The ECB in particular lives by a particularly delineated and limited mandate. For very good reasons the ECB was not set up to decide what industries or regions need subsidizing and which should be scaled back, accordingly to direct bank investment across Europe, to set the price of bonds, and and to print money to subsidize direct lending. These are intensely political acts. In a democracy only elected representatives can take or commission such intensely political activities. If I take out the words “green,” you, EU member states, and EU voters would properly react with shock and outrage at this proposal.

That’s why this movement goes through the convolutions of pretending that defunding fossil fuels and subsidizing green projects — however desirable — has something to do with systemic risk, which it patently does not.

That’s why one must pretend to diagnose “market failures” to justify buying bonds at too high prices. By what objective measure are green bonds “mispriced” and markets “failing?” Why only green bonds? The ECB does not scan all asset markets for “mispriced” securities to buy and sell after determining the “right” prices.
But the Bank makes it even more obvious that they have no clue what they are doing when they talk about the need to calculate and reduce the Bank's emissions profile. 

The only emissions from the Bank that are not covered by the ETS will be those coming from international travel, and maybe if they're getting bank notes shipped in from Canada. The latter will be trivial. Both could easily be offset if the Bank were so inclined. 

The only thing that happens if the Bank decides to take costly effort to reduce its emissions within the cap, efforts above and beyond what would be consistent with basic optimisation that accounts for an expected rising carbon price path, is that ETS credits are freed up for someone else to use. It doesn't make sense to spend $100 to avoid buying a carbon credit that costs $60. Someone else will just buy it instead. But it totally makes sense to spend up to that $60 to avoid buying the credit. 

The Bank calculates its revised carbon footprint at 10,014 tonnes. At $60/tonne, that's just over $600k in carbon charges. I wonder whether they spent $600,000 on reports calculating their carbon footprint. 

A future government that cares about having a Reserve Bank that knows its core business, and sticks to it, will have a bit of a job ahead of it.

The Frank Tay Scholarship

Frank was retiring as I joined the Economics Department at Canterbury in 2003. He'd still visit the Department for seminars and to catch up with colleagues. 

His article in the inaugural issue of NZ Economics Papers set the tone for the next several decades of excellence at Canterbury: professional training in economics requires at least an Honours qualification, with appropriate rigor all the way through. 

Alfred Guender had an excellent interview with Frank a few years back. He's since launched a memorial scholarship, to support post-graduate study. 

Details below, from the email Alfred sent through to me. 

A few months ago I sent a letter to friends, former colleagues, and former students of Frank Tay to inform them about a fund-raising effort to establish a scholarship in his name for a post-graduate student in economics at the University of Canterbury.

This is an opportune time to provide you with an update on the Tay Scholarship Fund. Thanks to donations received, the fund has grown to about NZ$24,000. A well-known economics graduate of the University of Canterbury, Prof. David Teece (University of California at Berkeley), has now indicated that he would match dollar for dollar the contributions to the Tay Scholarship fund, up to a limit of NZ$50,000.

Prof. Teece’s pledge provides all of us who are involved in this joint effort with an incentive to raise additional funds, Could I ask you to consider donating to the scholarship fund that bears Frank’s name? It would also be immensely helpful if you used your personal connections to former classmates at UC and other UC alumni to spread the word about Prof. Teece’s generous offer to match contributions. This may just provide the extra kick that is needed to establish a continuing scholarship that pays a higher stipend than what is currently possible.

Here are the details needed to make a contribution to the Tay Scholarship Fund, which is administered by UC Foundation. The BNZ account number is 020874 0015430 00.

Once again I thank you for your support of this initiative.

Update: Alfred says the UC Foundation now has a separate page for the fund.

Tuesday 19 October 2021

Held to ransom by the unvaccinated

It is not a pleasant calculus.

New Zealand apparently really only has 186 fully staffed ICU beds. The rest is potential surge capacity with staff who've had only short bootcamp training. 

Vaccinated people can still catch Covid, but they're very unlikely to wind up in ICU. Alberta's population is a bit smaller than New Zealand's: 4.4m to our 5.1m. They put up all their vaccine outcome stats, from their very large ongoing outbreak.

Just look at the numbers here. Let's imagine everyone were fully vaccinated. 

First off, if that happened, the number of vaccinated people winding up in ICU would drop considerably, because fewer vaccinated people would catch Covid, because there would be fewer cases around transmitting it. 

But even if you held cases constant, you'd have ICU admission rates, even among the elderly, of around 10 per hundred thousand population over a 120-day window. If an ICU stay is 20 days, divide that 10 per hundred thousand by 6. And much lower rates for younger cohorts.

It isn't good. Inflating upwards to a population of 5.1 million, and assuming that the basic age distribution is similar between here and Alberta, it's 103 people in ICU over a 120-day period, or under 18 in ICU at the same time if a stay is 20 days. 18 is about 10% of fully staffed ICU capacity. A big outbreak, if everyone is fully vaccinated, doesn't wreck everything.

But the numbers on the unvaccinated side are truly awful. 

Remember that 76.7% of Alberta's 12+ population is fully vaccinated. They took up just under 9% of ICU spaces over the past 120 days. 

An additional 9.1% of Alberta's 12+ population is partially vaccinated. They took up just under 4% of ICU spaces. 

The completely unvaccinated 14.2% of the 12+ population account for 87% of the 12+ ICU demand over the 120-day period. 

If everyone were fully vaccinated, ICU demand would have been less than 12% of actual demand. More than 7 out of every 8 people who did wind up in ICU wouldn't have wound up needing ICU. The system wouldn't be busting. 

It isn't much different in Singapore. There, 15% of the population is unvaccinated and that 15% takes up half of the ICU spaces. A lot of that 15% will be kids too young to be vaccinated and unlikely to wind up in ICU if they do catch it: a tiny proportion of adult unvaccinated people are taking up half of the ICU capacity. 

Policy restricting everyone's liberty is being set, in part, to prevent the hospitals from being overwhelmed by the small proportion of people who do not want to be vaccinated. 

Is it any wonder that Auckland Mayor Phil Goff says:

"There has to be a price to be paid for not getting vaccinated and that may be access to hospitality areas, it will be access to events and it may be access to a whole lot of jobs that are public facing." Exceptions will have to be made for people who can't get vaccinated, he said.

If every person who could be vaccinated were vaccinated, the hospital system could hold up even in a big outbreak. And a big outbreak would be less likely. But if even 15% of the population remains unvaccinated, as in Alberta, that 15% impose so much cost on the health system that you get horrible outcomes that deny hospital services to everyone, vaccinated or unvaccinated, for all the other things that send people to hospital. 

Care has to be rationed, one way or another. Surge capacity is better than no surge capacity, but the short bootcamp training will not provide care that is up to the normal standard.

Every option is terrible. Delaying and deferring cancer and other surgeries to save ICU space for the unvaccinated Covid cases is horrible too. Imagine though a rule that, if triage decisions have to be made because of those terrible capacity issues, the willfully unvaccinated are the lowest priority. It sounds horrible. Every option is horrible. But doing it this way might encourage more people to be vaccinated, and in doing so massively reduce the problem that we'll be facing. 

What other option might you suggest that you think would be more fair in allocating those scarce spaces while also reducing the burden that ICU will be facing? Don't just tell me my option sucks. I know it sucks. Give me something that is feasible and sucks less

Monday 18 October 2021

Morning roundup

 The morning's worthies:

Thursday 14 October 2021

No accountability in Covid-land

Dileepa Fonseca has more on the latest installment of the Rako saga.

I know I've been going on about this since last December. 

But it really matters. 

If you can't get accurate test results quickly back to people in a pandemic, you're always going to be well behind the curve. Contact tracing is a mess of too-little capacity. Pinning down cases quickly makes that job easier. Fast results, and especially if collection is unobtrusive, reduces test hesitancy. If you have to self-isolate for days on waiting for a result, and getting a test is inconvenient and a bit painful, you'll be less likely to go do it for a sniffle than if you just have to drool in a spoon and get results same day (or overnight if it was later in the day). 

And the government has absolutely messed this up. 

Rako had offered to scale up its testing to help in the public health effort. The Ministry told them to piss off. 

Now the Ministry's threatening to expropriate Rako via requisitioned testing. 

This is fifty different kinds of stupid, but here's one of those flavors of awfulness. 

Suppose you had a testing company, and you knew you could make an investment that would 8-fold increase your capacity. 

Would you make that investment early, knowing it would wind up being needed for the public response once the Director General of Health pulled his head out of his arse, so you'd be ready to run?

How would that calculus change if the government threatened to just steal all your stuff?

If you make the investment early, increasing capacity, the government can just take it at the price the government sets, and you have to fight them in the courts about it.

If you know that they are giving themselves that power, well, it would be stupid to invest in more capacity. They can't steal what you don't have. But you might be able to work it into future negotiations, so you don't get wiped out for having made the investment before the bandit comes calling. So it's then better to wait. Even in a big outbreak where a pile of testing is needed. 

Someone on Twitter said that NZ right now is like the first people showing up at Fyre Festival. Would be nice if those who'd been warning about this lack of preparedness all year weren't being dragged along for the ride. 

Morning roundup

 Ok. It's down to one Chrome sheet. The worthies!

Sellotape IT systems and covid testing

I'd struggled to understand why the government would be blocking Rako's access to the Covid test tracking system. Surely, even if they hated Rako, they'd want to get results reported in for their own tracking purposes. 

Dileepa Fonseka has a potential explanation. The IT systems, like most government IT systems, are a disaster.

The Government is struggling to meet its existing border worker testing obligations, allegedly thanks to IT systems “held together by sellotape”.

There have been growing calls for border workers to be tested more frequently than once a week in response to the more infectious Delta strain of Covid-19. Yet a recent move to saliva testing has highlighted a significant IT issue standing in the way of more frequent testing.


The problem was exposed during the saliva testing roll-out, because workers were required to be tested twice a week.

The mish-mash of IT systems, and other delays within the testing regime, meant the deadline for a border worker’s next saliva test could pass before they were asked to take their next one.

One source close to the problem said the whole system was “held together by Sellotape, and it should have been replaced 15 years ago”.

Wednesday 13 October 2021

Morning roundup

There are 5 different Chrome windows, each chocka. I can't do every bit justice. So, a roundup while I try closing maybe 2 of them.
The last month has been utterly mad. There was an old Roger Douglas line about the reforms in the 1980s moving too fast for opposition to them to mount. Labour's basically doing that again, but mostly in the wrong directions, in the middle of a pandemic, while not being appropriately on top of the pandemic stuff. 

It's not ideal. 

Google and the provision of public goods

Anything that makes the Android operating system more valuable to users is good for Google. More people taking up the Android platform means more people in the Google universe, using their apps, seeing the ads that they sell. 

They have rather decent inventive to do awesome things for the platform.

Yesterday, about 2.55pm, I got a push notification on my phone. It made a sound I'd never heard before. I looked over and it noted that there'd been an earthquake, estimated magnitude 5.6, about 100 kilometers away off shore. I'm going from memory here as I didn't screenshot it. 

By the time I remembered that Google had added earthquake warning functionality into the Android OS, and remembered David Hood's skepticism about it, the room started shaking. Somewhere between 3 and 10 seconds. I should have started counting. 

Had the number on the phone been bigger than 5.6, I'd have gotten under the desk. And I've had had time to do it too. 

I kinda hate earthquakes. But now, I'm kinda looking forward to the next minor one. I want to count it down, like the gap between lightning and the thunder. 

Thank you Google!

Contact details

It seemed really obvious, really early, that robust privacy protections were needed around Covid sign-ins. If even some people expect, entirely unreasonably, that there's any chance that police could subpoena that data, then it needs to be bright-line illegal for anyone to hand that data over to police. 

The government can insist all it wants that the police wouldn't do that, but people who don't inherently trust police to do the right thing would rightly insist on stronger protection than that, and especially when Australian police have been up to dodgy stuff.

The people who you most need to be checking in for contact tracing are the ones who are nervous about the thing, because they may be in the cohorts you're going to have a hard time tracking down otherwise.

Newsroom's Sam Sachdeva says Labour is the only party opposing a rule change here. 

A push to provide stronger privacy protections for contact tracing data now has the support of four political parties in Parliament, leaving Labour as the sole holdout for legislative change.

The issue of how contact tracing data is used, and could be misused, has been bubbling away for some time and came into the spotlight again last week in relation to the Covid-positive woman who was refusing to share her movements around Northland with health officials.

When asked by Newsroom last week whether the Government would offer the woman immunity from any criminal charges in exchange for being candid about her whereabouts, Hipkins said public health officials did not use information gathered in interviews for any purpose other than eliminating Covid-19.

“The public health teams are not collecting information that’s then passed on and used in other law enforcement procedures, because to do so would mean that people wouldn’t provide the information in the first place.”

However, the standard of rules governing the use of contact tracing data has been called into question in the past, yet the Government has taken no new action to address those concerns.

In September, an open letter signed by over 100 lawyers, experts and academics said the public health order establishing mandatory recordkeeping had “insufficient" safeguards to protect against the misuse of data collected through either the Covid Tracer app or paper-based records.

Some are concerned that police or other law enforcement agencies could compel the release of the information through a search warrant or production order, while others have said the private sector could also misuse data collected through paper-based records.

The Government is legislating themselves the right to expropriate private covid testing labs, while insisting they won't use that right. 

At the same time, they're refusing to legislate against the use of covid check-in data, because they say that there's no need to because they wouldn't abuse the data and everyone should just trust them. 

Would it be crazy for someone who doesn't trust police or the government to conclude that the government's refusal to legislate here is because they actually kinda want to be able to use check-in data for dodgy purposes?

Tuesday 12 October 2021

Further, on taking tests

On Monday, I managed to get a submission in on the Government's awful proposal to give itself the power to requisition tests and testing materials. Was a bit of a rush job, but it's a bit of a rushed legislative process. I hadn't known about it much before a hasty blog post on it on Friday

The submission alludes to a few problems in giving the Ministry of Health authority to set quality standards for testing, which the legislation also does, in addition to letting the Ministry take Rako's tests. 

My column at Newsroom, written after the submission, also goes through the problems. Currently gated. [Update: ungated]

A snippet:

If the Government wanted more testing capacity, it could have contracted for that capacity. And it still could. But over the past year, the ministry has repeatedly rebuffed Rako’s offers to provide testing. The Ministry of Health also told Rako not to maintain any extra testing capacity for the Government as its services would not be needed.

Shifting from a view that Rako’s testing was not needed, to that they are so necessary that they could warrant requisitioning, seems somewhat incoherent.

Once the Government has the ability not only to requisition materials away from prudent labs, paying only some deemed market value for the effectively stolen supplies, but also the ability to force those labs to ignore existing contracts and to provide testing for the State instead, bad things happen.

Existing clients would see the obvious threat to the security of their continued ability to undertake testing. That makes contracts less valuable. Labs can purchase sufficient supplies to tide them through shortages and international shipping delays. They cannot purchase sufficient supplies to ensure that they have enough left in the event that the Government decides to requisition all of the testing materials or all of the testing capacity.

And if the biggest thief in Wellington, whose offers are impossible to refuse, sees your prudential stockpile of testing materials as tempting, what incentive do labs have to maintain those stockpiles in the first place? After all, the Government could well instead decide to requisition someone else’s supplies on your behalf, come the crisis.

When all is done, we have a Ministry of Health that, after spending most of the past year seemingly doing its best to thwart Rako Science’s provision of accurate saliva-based, PCR Covid testing, and after telling Rako that its services would not be required for the public health effort, is now preparing simply to take Rako’s testing capacity through requisitioning powers. While saying it hopes not to need to use such powers.

It is not right. It will have repercussions for security of contracting and investing in New Zealand. And it will hinder the Covid response.

Dileepa Fonseca provides further very helpful detail in a piece today at Stuff. It's truly a must read, spectacular scoop. Leaked ESR test validation details, Ministry insider comments on the debacles. 

Here's how it starts. It goes on at length, and only gets worse for the Ministry.

A leaked study into saliva testing shows the Ministry of Health is still trying to figure out how accurate various types of saliva tests are as the Delta strain starts to take hold in Auckland.

The first stated objective of the leaked Institute of Environmental Science and Research (ESR) study is to compare different PCR saliva testing methods used on border workers by Asia Pacific Healthcare Group (APHG) and its partner laboratories.

However, it is the second objective of the ESR study that has people like Yale school of public health saliva testing expert Anne Wyllie worried.

That objective is “to generate clinical sample testing validation data from Covid-19 volunteers for the testing laboratories as requested by IANZ [International Accreditation New Zealand]”.

Wyllie was leaked the data by concerned scientists, and says there are no circumstances under which data gleamed through the study should be used to validate saliva tests.

“This is a preliminary, exploratory study. If APHG is basing their clinical diagnostic claims off of this – this is not at all robust or reliable, and not true validation.”

A spokesman for APHG wouldn’t comment on whether the data from this study had been used for validation purposes, or as part of their previous public statements on the accuracy of their saliva tests.

“We are not interested in commenting further on clinical validation except to say that our labs have had our work on that recognised by IANZ.”


“Why is there no clarity in the data that was used for validation? Why no clarity on which tests are being used for saliva? Why all the secrecy,” Wyllie says.

“Every lab in the US has to tell all its patients which test it is testing with and all that data is publicly available.”

The Ministry of Health argues testing validation data is the property of private laboratories and there is no requirement such results be published.

The Ministry of Health set the RFP for saliva-based PCR testing. 

The Ministry of Health could easily have set a requirement that any provider have a validated test, or get its test validated, and provide all data from that validation. 

The Ministry of Health chose not to do that. 

And then the Ministry of Health says "Oh, well that data is property of the labs so we can't ask for it." 

The Ministry of Health absolutely should not be able to set standards on Covid testing. 

Dileepa continues:

The authenticity of the documents, which were published just last month, have been confirmed by a senior Government official, who claims the whole saliva testing roll-out has been mismanaged by Ministry of Health officials right from the start. The official has requested anonymity.

All of this is part of a long-running saga with Rako Science, which was the first to achieve diagnostic validation for their saliva test, but was passed over for a $60m border worker saliva testing contract in favour of Asia Pacific Healthcare Group, which hadn’t been similarly validated at the time.

“They’ve got themselves into a situation with Rako [Science] and I don’t think they know how to manage their way out of it,” the official says.

An IANZ spokesman said it would normally assess things like Covid-19 tests against a fixed set of international standards, however these standards had been modified by the Ministry of Health’s clinical advisory teams.

”We do not set the accreditation criteria, we simply assess the labs against the criteria that are provided to us.

“Because of the rapidly changing nature of the Covid-19 testing environment, these criteria have been informed/modified by MOH clinical advisory teams composed of experts.”

And on the Ministry's competence in setting standards:

The Covid response bill before Parliament will allow the Ministry of Health to requisition testing supplies, and unilaterally set standards, something a senior government official says is equally worrying.

An official who spoke to Stuff sees the new law as dangerous because of the lack of understanding about this type of testing within the Ministry of Health itself.

“There just seems to be so many individual commercial interests now ... it’s hard to find somebody who is not conflicted.”

The official believes ministry officials didn’t understand what they were procuring when the saliva testing contract went out to tender. That meant officials focused on more easy-to-understand issues like the number of collection points available.

“It’s unconscious incompetence. It’s the people at the Ministry of Health in charge of a large procurement programme,” the official says. 

Friday 8 October 2021

Taking tests

The least-bad potential explanation for what the government is up to, in its rushed legislation that would give it the power to expropriate Rako Science's testing setup, runs as follows.

Suppose that, as Director General of Health, or the Minister of Health (who knows where this came from), you finally FINALLY figured out that most of your advisors on saliva-based testing are, in fact, utter idiots. They have led you astray for almost a year. All that stuff about saliva PCR testing not being that reliable was wrong, even if DG Health announced it as truth from the podium of truth. It's just that your idiot advisors couldn't figure out how to make it work themselves so spent the past year trash-talking it. And suppose you also figured out that there really only is one game in town if you want to scale up to a ton of testing that meets international ISO standards. And you've finally FINALLY figured out that you actually need a metric shit-ton of testing capacity. Because you only decide to do anything about an outbreak when there is an outbreak. Not before.

When you finally realise that you need to contract for tons and tons of testing capacity, and the only possible reliable supplier of an international-standard validated test in New Zealand is the company you have been utterly screwing over for the past year, well, you might just be a bit worried that prices could be higher than you'd like. 

In that case, having backstop ability to requisition testing from them at a price that you deem fair is a bit like the public works act, but worse. But you hope not to have to use it and that some negotiated contract could work. 

That would be the least-evil explanation for the powers the government is giving itself. 

There are, of course, more evil explanations. But I don't think there are less-evil ones. 

They brought the legislation 29 September, without telling Rako. Rako found out about it on 5 October. Submissions close Monday. The government is giving itself the power to steal all of Rako's kit: requisition their materials, requisition tests. 

I wrote it up as this week's Insights column.

There’s an old saying that makes the rounds now and again, with various attributions. “First they ignore you, then they laugh at you, then they fight you, then you win.”

I wonder how this one is going to end for Rako Science. I desperately hope it ends well. If it does not, it will be an utter travesty of justice.

The Initiative has long been keen on better Covid testing methods. They featured again in my report, released this week.

The University of Illinois’s SHIELD protocol saliva-based PCR testing rolled out very successfully in August 2020.

By late 2020, Rako Science had brought it here, under licence. By January, Rako was providing accurate and rapid PCR testing for private clients under contract, with a test validated to ISO15189 standard.

Rako had offered its testing services to the Ministry of Health in December 2020.

First they ignored Rako.

When Rako’s deployment spread through early 2021, the Director General of Health made erroneous statements, perhaps through simple mistake, at 1pm briefings about the accuracy of saliva-based PCR testing.

Then they laughed at Rako.

When it became increasingly obvious that it was impossible to scale up nasal swab-based testing in any practicable way, the Ministry of Health set a procurement process for saliva-based PCR testing. That process is the subject of a complaint by Winston Peters to the Auditor General.

And, for months, according to Rako, the Ministry ignored or denied requests from Rako to link Rako’s test results into the nationally integrated reporting system.

Then they fought Rako.

Last week, on 29 September, the Government introduced the COVID-19 Public Health Response Amendment Bill (No 2) into Parliament.

Rako found out about it on 5 October. Submissions close Monday 11 October.

Section 11 of the Bill provides the Government with the ability to requisition Rako’s materials and services, compensating the company at a deemed “market price”. Existing contracts with private hospitals for testing might need to be voided, come the requisition order.

On a charitable and optimistic interpretation, the Government has finally realised that it needs to contract for a massive amount of testing and does not want to have to negotiate with the provider of the only test validated to the appropriate ISO standard.

On a more frightening one, the Government is preparing to quasi-nationalise the provider they fought for so long.

Let’s hope Rako wins this one. If they don’t, we all lose.

BusinessDesk also has it, because Pattrick there has been the single best journalist in the country in following this file. Here is a snippet, you should subscribe. 

How else to explain parts of the COVID-19 Public Health Response Amendment Bill (No 2), tabled in Parliament last week, which would give the health ministry remarkable new powers to commandeer the use of private laboratories’ testing capacity? 

The bill proposes that the director general of health have power to make an order – and provide compensation – to a private laboratory to “undertake testing solely for the purposes of the public health response to covid-19 while subject to the order ... whether or not the laboratory is contracted by the Crown for that purpose”.  

It proposes the ministry be able to demand that any and all private test results be available for its national database.  

This sort of draconian drafting doesn't happen by accident and has the potential to produce results that could be both highly ironic and commercially damaging, depending on your perspective. 

Imagine, for example, if the Ministry of Health were effectively to commandeer the saliva testing resources of Rako Science to ramp up the national testing effort, having fended the company off by any means possible ever since its highly accurate saliva test became available last year. 

Rather than contracting for the service, the ministry would simply be obliged to "compensate" such a provider for their trouble. Watch this space.

Jonathan Milne at Newsroom also has it here. He quotes Rako's Leon Grice:

"There is other legislation where the Government can come in and expropriate or requisition private property – that's the Public Works Act. But that has more protections, like a process to determine a market rate that the Government must pay.... They can just insist we give up our stock and our reagents and our premises that we need to do our work."

If you care about the rule of law in New Zealand, and about the Covid response, you might consider submitting

If you haven't time to do anything else, just telling them that it is wrong to propose stealing the testing system that they spend months and months deriding might be nice. 

If you have more time, reminding them that they ought at least be indemnifying Rako against all of the breach of contract messes that will result if the government steals Rako's testing capacity, taking testing capacity away from the private hospitals and others who have paid for this service, causing a breach of contract, after having told Rako repeatedly not to reserve testing capacity for any government need and not to scale up for it - well, that could be a nice addition too. 

On a straight-econ side, even the least-evil explanation above is terrible. 

If investing in testing capacity is high risk, because the government will spend a year crapping all over you, telling people not to contract with you, telling media at the 1 PM Podium of Truth sessions that your testing system isn't reliable, because you're not their preferred supplier - and then the government FINALLY figures out that they might need to get a ton of testing from you? 

Well, at least some return on that risky-as-hell investment is in order if you want anybody to be willing to take similar risks in future. 

None of the international corruption indices understand how New Zealand works.

I'll be submitting on Monday. I'll probably write more on this for next week's Newsroom column. It stinks.