Anyway, just a quick note to ask whether you'd seen this?
(There's a better figure of the data here)
Well, he didn't really leave me much room for adding more fisking, since it seems already done. Bunching at all the boundaries, not just at the top one.
I can't believe James Shaw is trying to make out high income earners as the bad guys when tax avoidance at every level is right there, staring you in the face: at $14000 the second marginal tax rate kicks in, $23,000 is the maximum amount for a Family Tax Credit top-up via WFF, $48,000 is where the third marginal income tax rate kicks in, and at $70,000 the top marginal tax rate kicks in.
I can't figure out the spike around $18,000, though. Any ideas? (It's not student allowance, that abates at around $10000 a year, and it's not minimum wage earners, who should be on around $30,000 with full time work).
Anyway, thought you might enjoy fisking this one, talking about incentives, etc - you know, the stuff you do so well :-)
The spike around $18k is almost certainly the $16,698 that each of a married couple both on NZ Super would receive, Bryce Wilkinson kindly tells me.
Some folks getting up to the $70k threshold, particularly second earners, won't bother putting in extra hours because the return to work is lower. More of the action would probably be owner-operators paying themselves a salary just up to the threshold, and building equity in the firm past that. It isn't immediately obvious to me that the proportion of taxable income in the spike is bigger now than in the prior two years either.
I'd be more worried by the bunching at the $23k threshold than at the $70k threshold as it's more likely showing something real rather than something accounting.
If you retain earnings in a company and then buy yourself a car for the company, there's FBT hassles and having to prove that you need it for company use and paperwork for what part of it is privately used and then paying FBT on the private part. But maybe you've found it worth that hassle. You consequently have a slight distortion in consumption choices shifting towards things that are plausibly business expenses and away from things that are not 'dual use'.
If you decide to stop working more hours because WFF abatement rates are too high for part of the range and effective marginal tax rates are consequently really high, that seems a bigger distortion.
James also points to this Saez piece that failed to find much kink-point bunching in US tax data.