Levies as end-runs around the Generic Tax Policy Process

A few years ago, Willie Jackson proposed levying tech platforms to fund news outlets. 

I'd warned that this kind of thing amounts to a dangerous end-run around IRD's generic tax policy process.

Levies can make sense in some contexts. If a producer group agrees to be levied to fund research or marketing that has industry-wide benefits, that's fine. Agreement tests whether those benefitted actually benefit. 

Or if it amounts to a user-charge that can't easily be collected in other ways. It requires a tight link between what's being funded and who's being levied to pay for it.

But Jackson's proposal was nothing like that. There's no link between tech platforms and news outlets that would warrant a levy. 

The levy instead tried to use force to recreate a relationship that had been superseded by technological change. In olden-times, newspapers were the best place for advertisers to reach customers. Google and Facebook became better ways of linking advertisers with eyeballs. Since platforms 'stole' that link, it must be reforged through levies. It's a terrible approach to tax and tech policy. 

Paul Goldsmith was initially enthusiastic about continuing with that approach when he was made Minister, but it's since been shelved. 

Now Goldsmith's back with a new levy proposal. This time, Disney and Netflix will be levied to fund NZ content creation. 

Same problem as last time. There is no link that justifies a levy here. 

NZ taxpayers subsidise local content creation; it gets broadcast on by anyone willing to pay for the rights to distribute it. It's a generally decent approach because what gets created still faces a market test. NZ content creators are perfectly free to license to Netflix or Disney or anyone else who's willing to pay, and those outlets will be willing to pay if they expect the additional offerings to get or keep subscribers they otherwise would have missed. It's fine.

Irene Gardiner, president of NZ Screen Producer's Guild Spada, has views:

“The big international streaming companies operate here without any regulation. They don’t pay company tax here, they use our broadband infrastructure that the taxpayers paid for, and they have no requirement to commission any local content or contribute to the New Zealand screen sector in any way.

“We’ve been lobbying the Government for some form of legislation in this area for over two years.”

Gardiner worries that after legislation was introduced in Australia last year, New Zealand is getting “left behind” – particularly amid the “devastating” impact of streaming services and Big Tech on local media.

“If any of the big streaming companies, Netflix or Apple or Amazon, had taken a genuine interest in commissioning in New Zealand and done some significant commissions in the long time that they’ve been operating, I think we’d feel differently. 

“But the reality is that they haven’t, and so if they’re not going to do it voluntarily then here we are.”

Households pay for broadband. Their broadband subscriptions help cover the cost of the broadband network. They can choose to stream whatever over that fibre, including Netflix or Disney or Amazon or whatever else. 

If Kiwi subscribers put value on seeing NZ content on any of those platforms, those platforms would have incentive to offer it. As it stands a lot of NZ content is only available on really crappy NZ services where you can't pay to avoid ads. If Kiwi viewers hate ads more than they like seeing NZ content, then they won't watch there. You could maybe make a case that international streaming platforms, by offering a far better product, wind up meaning declining viewship for stuff only available on TVNZ+ - but that would be a case for TVNZ+ to start offering a no-ads subscription version. 

And presumably any platform seeing potential gain in it could outbid TVNZ+ or whoever else for the streaming rights. If they aren't, then the benefits they see in increased global subscriptions aren't worth the cost - even though the product's creation and consequent cost was likely heavily subsidised through existing content subsidy schemes. 

If there were a principled tax-policy basis for taxing international digital platforms, that case should be evaluated through IRD's generic tax policy process. 

This levy-based approach will prove increasingly tempting to a government that does not want to reduce spending to meet its tax revenue, doesn't want to increase taxes transparently, and wants to provide services through other funding mechanisms. 

Coming up with new tied levies is a way of short-circuiting all of that. "It's not a tax, it's a levy" to keep the Taxpayers Union from yelling at them (probably won't work) but also to keep it away from IRD analysis on whether the proposed tax is coherent with the rest of tax policy. 

Appropriation first, policy afterwards

The government has not yet announced what it wants to do in the online child-protection space. There's a member's bill endorsed by Luxon that tries to follow Australia's social media age limits. But the education select committee wound up with much broader recommendations and Stanford's tasked with responding to them. 

What that'll all turn into is anybody's guess. Australian age-gating for social media? Ofcom-style 'let's make everyone do an ID check to look at darn near anything on the internet while sending angry demand letters to American platforms that don't want to comply with UK regs'? Something else?

Whatever it is, the government seems to figure the regulatory regime will cost $8.5 million per year when it's up and running.

The budget includes this new initiative, which gets $6m in the first year rising to $8.5m in each of the last two forecast years: "This initiative provides funding to develop policy and possible regulatory options to improve children’s online safety, subject to future policy and funding decisions."

None of that makes sense if it's an appropriation for developing policy and possible regulation. It's too much money, and it rises over time rather than declining in the out-years when the policy development work is largely done.

And if you look into the Vote Internal Affairs categorisation, well, the thing's classed as regulatory services - not as policy and related services. 

It's far more plausibly an operational allocation for running a new regulatory regime.

One that, as yet, not only has no supporting legislation, but also no hint of what it's meant to be doing. 

My column in Newsroom this week, now ungated, goes through it.

All of this is pretty dumb.

Nobody has yet figured out a way of age-gating social media or potentially sensitive stuff online that doesn't suck. 

New Zealand is unlikely to be the first place to find a way of doing this that doesn't suck. 

The potential harms are real, but often overstated and highly heterogenous. 

There are existing controls that parents can use to gate access for their kids. Some of those controls are undermined by school accounts that parents cannot control. 

The government could be very helpful in providing resource to schools to help parents understand the tools that are available to them, and in helping schools to not undermine their families' choices by setting school accounts whose controls don't mirror those set by parents (or otherwise provide circumvention options on time limits or app limits by logging into the school account on their device or on a school-provided device). 

Anything beyond that, and just enforcing existing law on other bits around grooming etc, should be a watching brief. If somewhere else *does* find a way of my trilemma, great! We could piggyback on their version if voters wanted to do that. We wouldn't have bespoke compliance costs that platforms would be quick to ignore - or to use as basis for just blocking countries that are too small to be this stupid. 

And I just despair when I hear people from industries that have suffered enormous costs from legislation set to 'send a message' regardless of any cost-benefit assessment claiming to support social media bans because they 'send a message'. Crooked timber...

A weird way of slicing the stats

Ages ago I supervised a superb Honours thesis, which turned into a Masters, looking at the lesbian wage premium. It showed up regularly in the US data: homosexual women earned more than heterosexual women - the opposite of the pattern that obtains for men. 

I was curious whether the difference could in part be due to employers' expectations about the costs of accommodating maternity leave. And it looked like that mattered. Hayden Skilling did superb work on it, helped in part by Ron Oaxaca's visit to Canterbury while Hayden was writing. 

I'd wanted to use New Zealand data but that seemed to be impossible. We could, with a few clicks, get US data from the ACS without any bother. If we wanted to use NZ data, it would have been impossible. Stats New Zealand just makes it too hard to access NZ microdata. So we wind up with NZ researchers using American data and helping advance global understanding of what's going on in the US. 

Hayden's thesis was out in 2014. 

It's 2026. Stats NZ just put out a couple of releases looking at earnings among LGBT+ populations. 

I wanted to check whether the lesbian wage premium held up in NZ data. 

So I went to have a look. 

And it's just a big mess. 

First, we get a press release highlighting substantial differences in age-adjusted average annual personal disposable income between the (lumped together) LGBT+ population and the non-LGBT+ population, with the transgender and non-binary population having the lowest age-adjusted annual disposable income. 

That sounds like discrimination right? 

But then you check the second page. The one listed as a "related page" that folks might not click on. That one notes that the LGBT+ population reports disability rates of 25.6%, compared to 15.8% for others. 37.9% of the transgender and non-binary population were identified as disabled.

Disability will mean lower earnings. Stats NZ's big headline press release figures adjusted for age but didn't adjust for disability. That seems like an important omission. People might chalk differences up to identity that are at least partially differences by ability. A cross-tab so you could compare earnings by disability status across the categories could help, but this is one of those 'if we didn't pre-supply the cross-tab, it is unknowable' things. 

I downloaded the excel sheet, hoping that I might be able to check for differences in earnings between heterosexual and homosexual men and between heterosexual and homosexual women. 

But that appears to be impossible. 

In their gender splits by row in the cross-tabs, male and female include the transgendered identifying with each category. Heterosexual males (including the cisgendered and transgendered) earn more than males (including the cisgendered and transgendered) who identify with a sexual minority. That latter category lumps together people identifying as gay, those identifying as asexual, and many others. There could be substantial differences within that bundling. 

So I can't have a clean read on earnings differences between non-trans straight men and non-trans homosexual men. 

And among females (including cisgendered and transgendered), mean personal disposable income for heterosexuals is suppressed. Stats NZ does this when reported numbers are too low. But they do report earnings among females reporting as sexual minorities. Which is difficult to understand. The sample size tends to be smaller for minority groups. But even if it were not suppressed, it wouldn't be helpful. Because I wouldn't be able to get difference between cisgendered heterosexual women and cisgendered homosexual women. 

I don't know if US data has gotten any worse over the period, but NZ data sure hasn't gotten any better.

I've emailed SNZ asking whether any of this is knowable.

There could be good reasons for lumping groups together as they have; they have a lot of potential categories, and splitting out each one would just mean everything would wind up being suppressed. But splitting out the main obvious categories would seem pretty possible.  

Superannuation affordability options

Lyric Waiwiri-Smith at The Spinoff asked me what I thought the options might be for dealing with rising superannuation costs. 

Her story's here, along with comment from Max Rashbrooke and Shamubeel Eaqub. 

My most-preferred option is ongoing increases in immigration rates, coupled with shifting to CPI-indexation of super benefits and indexing the age of eligibility to healthy life expectancy.

I did some rough ballparking. If net migration were around 1.8 people per person turning 65, we could maintain current-ish ratios of 'working age' people relative to 65+. That would have to increase over time as life expectancy increases. 

But there is a real and obvious play here. NZ is aging; Europe as aged. Immigration New Zealand could explicitly advertise for young productive migrants in places where those young workers are being predated upon by their country's elderly voters even more heavily that they are here. Getting a small proportion of young and productive workers from large countries could postpone the inevitable here almost indefinitely. Remember that one big reason that NZ Super costs haven't already blown out is higher net migration than Treasury had expected in the 2000s. 

Consider the number of people aged 65+ per population aged 20-64. On that basis, NZ is around 29.5 and Germany is already at 39.8, France at 40.2, Italy at around 42. 

This kind of play is self-sustaining, on the receiving side. The more young Germans who leave for younger shores, the more who'll want to leave. And there'd be advantages to being in the first wave of leavers, because if the flow gets too large their government might start setting exit taxes. 

I asked GPT to come up with some sample Immigration NZ campaigns targeted on this basis: places where the demographics make NZ attractive by comparison. It did a reasonable job!

The top five, if this were a real campaign

I’d put serious money into:

1. Germany. The cleanest combination of size, ageing pressure, worker burden, education, and plausible NZ fit.

 

 

2. UK. Not the purest demographic case, but probably the best cost-per-success market. The sales pitch is less “escape pension collapse” and more “same language, better lifestyle, credible residency pathway, and fewer inherited fiscal messes.”

 

3. France. Strong burden story and strong human capital, especially if targeted at engineering, health, tech, science, agriculture, and public-sector professionals tired of state sclerosis.

4. South Korea. This is the big non-European play. The message writes itself: “Your country is about to experience the steepest demographic cliff in the OECD. Move before your peak earning years become the funding mechanism.”

 

5. Poland/Czechia/Slovakia. A regional Central European campaign could be very productive: educated, mobile, demographically pressured, and plausibly attracted by an English-speaking high-income destination outside the EU rat race.

 

Another non-tariff barrier

I do not see the problem here. I do see a lot of ways of creating a problem though.

Andrew Bevin writes for Newsroom:

Ministers were warned of trade implications relating to mandatory Health Star Ratings before choosing to vote against the development of the scheme.

If Health Star Ratings are made compulsory by Australasian food ministers, and New Zealand manages to opt out, Australia would likely block imports of non-compliant food.

That’s according to advice given to the Cabinet Economic Policy Committee ahead of New Zealand’s vote against developing a mandated Health Star Rating system earlier this year.

If NZ made the FSANZ health-star ratings compulsory, Kiwi firms would have to comply. So would anyone else wanting to sell food in NZ. International outfits would then either eschew our market as not being worth the hassle, run limited production runs meeting the FSANZ standard (at higher cost both because of the smaller run and because they'd lose flexibility to shift products across markets as market conditions change), or make Kiwi retailers put stupid little stickers onto everything manually. 

All of those would limit competition here and push up costs. 

If NZ did not make the health-star ratings compulsory, Kiwi firms wanting to export to Australia would have to comply. And that's fine. They can do that. They could even sell the same version of the pack in NZ. And product from other countries could come in too - if they met our biosecurity standards. 

Why create another non-tariff barrier?  

Assorted budget bits

A few minor bits I noted in looking through the budget - won't bother going through the headline stuff that will have been well covered elsewhere.

• They expect to save $1.97 billion by 2029/2030 through the public sector transformation project reducing staffing numbers. There will be pressure on that figure despite the substantial increase in public sector staffing, both in absolute numbers and as fraction of population, over the past six years. 
• Statistics New Zealand gets a large budget increase, both opex and capex, to modernise the IDI. At the same time, it will be held to baseline savings like the rest of the public sector. The Minister for Statistics might want to watch that SNZ doesn't siphon IDI money out for other activities. 
• They're fixing part of the FIF regime as it faces domestic investors, to match the fixes made for those moving to NZ. Taxing unrealised gains was always dumb; good that they're fixing this.
• The Proposal for Reducing the Risk of Online Harm to Children puts $30.75m over four years "to develop policy and possible regulatory options to improve children's online safety, subject to future policy and funding decisions". This is just for the policy development work. It seems like far more money than necessary for a project that shouldn't be being undertaken in the first place.
• They've set a Defence Technology Accelerator as part of the Defence Capability Plan. Sounds neat; only gets $16.1m over 4 years. Maybe if its first year looks promising, funds from the online harm thing could be shunted over here.
• Customs is getting $15.3m opex and $19.5m capex to respond to "increased smuggling", text says it's aimed at illicit drugs. And $35.9m in third-party levy revenue. Could affect tobacco excise too.
• A whole page of the BEFU Supplementary Materials goes through the weaker outlook for tobacco excise. They've sharply reduced forecast tobacco excise revenue as compared to the HYEFU forecast: $1.58 billion over the forecast period. They note a weakened demand profile - but they don't get into whether it's a drop in smoking or a shift to illicit markets. It's a drop in demand for excised tobacco in either case. But they do note an offsetting minor increase in forecast tobacco revenue over the same period. And this is kinda funny.
  

  "The excise rates for heated tobacco products (HTPs) were reduced by 50% on 1 July 2024. Recent data show that the decline in duty from the lower HTP duty rates has not been as large as was expected. Furthermore, subsequent research suggests that future HTP take-up will not be as large as was previously assumed."

  Remember the giant beat-up on Casey Costello in 2024 for the "Tax break for big tobacco"? It was all based on that very stupid estimate that Treasury stuck in the forecasts. Nobody should have believed the figure at the time - it was ludicrous. I don't know whether Hon Verrell actually believed it, or whether it just gave her a convenient line to beat up on the government for its changes to tobacco policy. Neither's great. My column on the stupidity of that figure is here.

• I think Treasury is likely overestimating alcohol excise returns. Recall that SNZ reports very sharp reductions in per capita alcohol available for consumption - those figures are based on excise returns. Treasury has applied a one-off drop to current levels, but then a reversion to prior pre-Covid trend growth in excise. I think Covid and GLP-1 inhibitors and general trends in youth risk-aversion have caused a structural break. Total alcohol available for consumption, on the SNZ figures, peaked in 2021 at 36.3 million litres and have declined since despite population growth - 2025 was only 31.3 million litres.  

• Overall, it'll take a heroic effort to stick to the plan they have, and that'll only get us to structural surplus by 2029. 

SEZs as policy trial areas

A decade ago, I coauthored a report looking at how greater localism and subsidiarity could be achieved in a very centralised country where local councils have variable capabilities. 

We settled on policy trial areas. 

The basic gist was as follows. 

First, a community would pitch a policy trial area - a special economic zone - with different policy or regulatory settings more suitable for local conditions. The idea would come from the local community. Some national-level policies are really unsuitable to some local conditions. 

That community would work with Treasury to come up with indicators ahead of time. How could we tell if the trial were working? What side-effects might we need to watch as well? 

Successful trials would often mean higher tax revenue for central government, lower dependence, or both. share the gains with the originating community as a 'policy discovery' payment. Then let it extend to other communities asking to take it up. Failed trials would fail at small-scale. 

Central government would rule out any proposals that could not, in principle, be extended to other similar communities if the trial were successful. So tax concession areas would be right out. Different consenting processes could be fine; a successful trial could extend to similar consents in other places. 

I have not read the proposal for Marsden Point. 

But I do not recognise our proposal in Bryce Edwards' critique of what's been proposed at Marden Point. 

Edwards writes:

A lobbyist’s paradise

The economist Michael Reddell saw the obvious problem the moment Jones first floated the idea. If there were any substance to the SEZ concept, Reddell wrote, the policy seemed “likely to be a lobbyist’s paradise, and perhaps that of political party donors & recipients”. He recalled that the New Zealand Initiative had pitched something very similar a decade ago.

Reddell’s lobbyist point is the one that NZ First does not want to engage with. The moment you start designating discretionary zones with bespoke tax treatment and accelerated consenting, you create exactly the kind of high-value, low-transparency politics in which the lines between commercial interest and political access become blurred.

Who decides which company qualifies as being inside the zone? Who decides what activities “achieve the aims of the zone”? Who appoints the panel? On what criteria? Under what review process? These are not pedantic questions. They are the central governance questions, and they are conspicuously absent from anything Peters or Jones have said in public. NZ First, of all parties, used to have something to say about that sort of arrangement.

To answer Edwards' questions within the framework that my shop proposed:

1. Nobody decides which companies qualify as being inside the zone. The zone applies to activities within the zone's boundaries. If the company's activities are inside the zone, then those activities would qualify. 

2. Nobody would be deciding on activities, except when the zone is struck. A proposal to, for example, trial a different version of the minimum wage for piece-rate employers in the zone would apply to all piece-rate employers in the zone. There'd be no assessment of aims.

3. We didn't have panels, so appointments and criteria weren't questions. 

4. We did have review - against the indicators that the community had set with Treasury. Central govt doing the assessment. 

Maybe Edwards' column is better in the half that's on the other side of the paywall.