August 2, 2015

Some retail arithmetic

From a Retail New Zealand Media Fact Kit

The amount New Zealanders spend on goods from foreign websites is approaching $1.5 billion and this number is growing all the time.

GST is 15%, so the total GST payable on an amount approaching $1.5 billion would be approaching $225 million.

Since $1.5 billion would be approaching $900 per household (1.68 million households according to StatsNZ), I assume it includes quite a few business purchases as well. For these, any increase in GST on the purchase would later be deducted from the business’s GST liability, leaving a net zero.

For non-business purchases, some of that GST is already payable under current law (if the total value of a single package is over $400). Some would still not be payable if the threshold was lowered to $25.  Also, some might not be payable and definitely would not be easily collectable at the border because the purchase is an electronic download — e-books or music, for example.

Putting all these together, the potential increase in revenue has to be less than 15% of $1.5 billion, though it’s hard to say how much less.

The Media Fact Kit says

The Government misses out on at least $200 million in tax (maybe as much as $500 million) every year

Based on their expenditure numbers, that doesn’t look plausible. Still, maybe their numbers are wrong and it really is more than $200 million, and maybe even as much as $500 million.  GST would still cost something to collect, and more than it does (per dollar) within New Zealand, so the net gain to the Government would be noticeably smaller than this. They don’t even pretend to take the costs into account, they just talk about how many hip replacements could be funded with all the free money.

One of the differences between a Treasury Regulatory Impact Statement and a “Fact Kit” from a lobby group is that the numbers in the RIS have to add up and the document needs to give sources. If, as Radio New Zealand reports, there will be a proposal for Cabinet this month, we might be able to get some real numbers about the likely revenue and costs, and perhaps even how the economic impact would compare to other ways of raising taxes.

avatar

Thomas Lumley (@tslumley) is Professor of Biostatistics at the University of Auckland. His research interests include semiparametric models, survey sampling, statistical computing, foundations of statistics, and whatever methodological problems his medical collaborators come up with. He also blogs at Biased and Inefficient See all posts by Thomas Lumley »

Comments

  • avatar

    It seems to me that this estimate is likely to suffer some of the same issues we’ve faced before: The 8 Billion Dollar iPod

    https://www.statschat.org.nz/2012/03/16/the-8-billion-ipod/

    also worth remembering at this time of TPP (no A yet).

    9 years ago

    • avatar
      Thomas Lumley

      We’ve also previously seen Retail NZ providing numbers that looked a bit ambitious — that time, on shoplifting.

      9 years ago

  • avatar
    Thomas Lumley

    I should also mention that BooksellersNZ has a serious effort at making a case. Eric Crampton writes about it here. He doesn’t agree with it all, but it’s worth engaging with.

    9 years ago

    • avatar
      Megan Pledger

      I bought something on ebay for roughly $15 from China with free shipping. When it got here the cost of the item on the declaration was $1 and it had a $14 postal charge. So my GST cost would have been on the $1.

      A GST tax collected by the people who deliver it means that people who can transfer the cost to other non-taxable parts of the deal will do so e.g. a $20 item becomes a $2 item with an $18 charge for “wrapping for international post” or some such.

      (I suspect that the cost of storage to the delivery people for people who can’t pay the GST fee (yet) and the infrastructure needed to keep track of it all will blow the model out of the water.)

      My idea was to collect the GST at the point of the credit card transaction with a NZ bank. It would get physical and non-physical goods as well. But the banks won’t want to know about it – the development cost would eat into their profits.

      9 years ago

  • avatar
    Jordan Watson

    Interesting points here. Thanks for sharing your insight here!

    9 years ago