Act’s “Honesty for Taxpayers” policy sounds nice, but if the objective is clarity, it will be unhelpful or worse.
In America, every government form has a small “OMB Approval Number” in the corner. In line with the Paperwork Reduction Act, all government departments must have all forms approved by the Office of Management and Budget every three years, and show an “estimated burden time” alongside the approval number. There’s probably some benefit to this: perhaps agencies would otherwise have a habit of asking for more than they need. And the direct costs aren’t too bad—less than $10 million (Shapiro 2013). But the irony of creating a bureaucracy to fight bureaucracy seems lost on people. Approval, required to collect any information from more than ten people, takes 60 days including a public comment period, often longer in practice. If the problem is inefficient governments, slowing down the government seems like a counter-intuitive move.
Supporters of Act’s new “Honesty for Taxpayers” policy would do well to keep this in mind. This doesn’t mean that the policy is a net harm, of course, but the equation is not nearly as simple as its leader, Jamie Whyte, makes it out to be.
Why so much spending?
Some of Dr Whyte’s diagnosis is woefully misattributed. He blames the absence of checks and balances for making New Zealand “the fastest spenders in the West.” Perhaps so, but fast isn’t the same as wasteful or opaque, and more dire cases of wasteful spending are found in America, where checks and balances run galore. America’s legislators are notorious for sneaking unrelated clauses into bills that push federal spending towards pet projects in their constituencies. Each line is small in the context of the whole federal budget, but they add up. The second chamber and entrenched constitution don’t seem to help.
Also, Dr Whyte forgets that speed runs both ways. Just as governments can increase spending easily, they can cut programmes—as the current government did with student allowances and the R&D tax credit, much to the chagrin of left-wingers and the tech industry.
Similarly, California’s administrations may have been overspending, but only because they were bound to by direct democracy initiatives. Act might like how, in California, tax increases require a two-thirds supermajority of both houses. But Californian voters also had a habit of approving spending for new programmes in voter initiatives, which means that their legislators get little discretion over the government budget—and hence, the trade-offs they should make when spending starts to run away.
Perhaps Act believes that California would not have voted that way if Act-style income tax warnings had been included in the official guides, rather than just the total costs (though that’s not what they said). Maybe that’s the case. But if they want “honest”, useful information, Act’s proposal is an odd way of going about it.
This won’t mean anything, either
A back-of-the-envelope calculation to derive Act-style income tax figures is relatively trivial, but like the total cost, it’s useless information. In no world without Working for Families would “the 17.5% income tax rate be 12.5%”, because no responsible consequential tax adjustment would change just one tax bracket. Assuming we wanted to keep the tax system equally progressive and non-distortionary, all tax brackets would be adjusted, along with the company and trust rates to follow the top income tax rate. It might be sensible to adjust bracket boundaries as well as rates, and maybe GST too.
In fact, the warnings Act proposes could be dangerously misleading. Someone who understands income tax would realise that if the 10.5% rate drops to 3.5%, that’s (mostly) just another way of saying $980 per earner per year. But one would be forgiven for thinking that WfF comprises close to a majority of the government budget, or that they’d have 7% more of their income. This isn’t stupidity. It’s the natural at-a-glance impression of anyone who, unlike me, hasn’t spent hours musing about taxation. Perhaps the extremity of that example would bring people to their wits, but something like “the 28% company tax would be 25%” would not.
If we applied this analysis to superannuation, which Act strangely forgot about, you could wipe the 10.5% rate completely and drop the 17.5% rate to probably about 4%.  There’s a reason Act neglected this: canning superannuation would, unlike canning WfF, be universally unpopular. What’s more realistic is peeling back superannuation: raising the retirement age or means-testing it, for example. Act’s policy doesn’t allow for transparency in the nuances that matter.
Far from being meaningful revelations, Act’s policy would open a new can of worms. How do we determine where the tax burden of a policy lies? Do we assume it’s equally distributed by person, or proportional to the tax they pay now? Some difficulties are by design: an Act-style income tax statement would make no sense for a national highway funded by an earmarked road tax. Some require thought about the counterfactual: would tax cuts for welfare cuts be aimed at the poor, or would welfare be funded by taxes on the rich? (This is a dichotomy: think about it marginally.) And would estimates take into account the impacts on the behaviour of consumers and companies? Or savings elsewhere: say, for a corrections policy, savings resulting from a reduction in reoffending?
Act for honesty, or Act for small government?
There is a more basic tension in Act’s proposal. Act opposes government spending beyond the basics, and said as much in Dr Whyte’s speech. In most cases, the misleading effect of their statements would probably make spending seem more significant than it really is. Supporters of small government are unlikely to complain about this if it means people turn off government programmes that they think are wasteful.
Yet it is difficult to reconcile this with the policy’s stated objective. Dr Whyte says that people “should have a clear idea of the price of [an] agency in their taxes or rates”, that “good decision-making depends on good information.” You would think, then, that the policy is about providing high-quality information. The sort of back-of-the-envelope calculations Act proposes are the precise opposite. And they give themselves away at the end:
Politicians from the big spending parties will oppose this policy. That shows what a good idea it is. The bureaucracy will also resist it, because voters will be surprised to realise that much new spending is generated by bureaucrats.
Their language doesn’t really contemplate the possibility that oftentimes, the cost will be worth it. The information they provide focuses only on income tax rates, and not on all the other factors a policymaker would (and should) take into account when making a decision. Act dresses this up as being about informed citizens, but they are really only interested in certain information that will help achieve the objective they seek.
If you’ve read any of my previous posts, you might know where I’m heading with this. Demanding the disclosure of information to help voters is a value-agnostic procedural policy. Act is curiously selective about where they apply this principle. You should bear this in mind when reading their rhetoric about “honesty”, “information” and “accountability”. It’s really about something else.
 This is an extremely rough estimate based on the Treasury numbers on the effect of changes to tax rates. Crucially, it assumes that the effect of a two-percentage-point change is twice the effect of a one-percentage-point one and so on, which is patently wrong, but should still give a ballpark figure.