Compliance Costs

No Right Turn has changed his mind, and now considers anyone who opposes GST on food to be a facist bigot.

Ok, maybe not quite in those terms. But I’m sure that’s only a matter of time, now that he’s made his mind up.

So, low compliance costs (if we manage the change right), and positive health benefits. Against that, we have the simplicity of the tax system i.e. the intellectual purity of some bean-counter’s ideological schema. One of these things is worth more than the other, and it isn’t religious economics. I’ll take solid real-world benefits over pure thinking any day.

Now, part of his case is a study that shows people will purchase more food if GST is taken off. I’m not going to question that here, other than point to the Hawthorne effect – it’s entirely possible that part of the change was due to the study (and the nature of the study) itself. Whether or not the benefits would continue after the change becomes routine is quite another question.

No, my biggest problem is the compliance costs arguement. Gordon Campbell sets these out thus:

Any introduction of a GST food exemption would create two potential sources of cost to business : namely, the one-off costs of the equipment to manage the exemption efficiently, and the ongoing costs of identifying and managing the distinctions between items that qualify, and those that don’t.

Gordon appears to be arguing that in Australia, the bar-coding system can solve these problems, as it’s standard across all retailers. The system identifies the exempt items to the retailer. Brilliant!

Um, no. That takes care of part of the issue. You can identify all you like, but it’s useless information unless someone listens.

What that means is that the barcode system can identify exempt items until it’s blue in the face, but unless the retailer’s systems are modified to accept that information and change the price calculation for each exempt item to exclude GST, nothing is going to change.

And doing that is a cost on each business. Worse, it takes a simple, easily replicated calculation and makes it horribly complicated.

Before: sum, then calculate GST

After: cycle through each item, test if GST applies, if not place in “exempt” total, if so, place in “not exempted” total, calculate GST on “not exempted” total.

Sure, it’s just making one small part of the system much more complicated. Except that software systems already are quite complicated. Discounts, stock, fly-buys, deli/butchery items… each of these creates relationships with each other item.

Adding one doesn’t make the system 10% more complex, it makes it 10 times more complex in the parts it affects. It means someone has to work out which of the other elements it interacts with, and how, and what the circumstances (which themselves will be complicated) of the exceptions are to those relationships. Then, it must be programmed, and tested. That could easily cost tens of thousands of dollars or more.

I’ve seen bugs in software, simply due, not to the complexity of any part, but due to the fact someone had (to stay within the supermarket example) forgotten to account for GST exceptions when someone on an account purchases something from a butchery and swipes their flybys.

Then there’s reporting & checking. Previously, with GST on everything an auditor could simply check the totals and a handful of existing exceptions – something that might take a few seconds. With this change, an auditor (or manager checking his reports) would have to go into a whole new level of complexity to check his reports, and take many times longer to check each item by hand for exemptions before doing his checking calculations.  Sooner or later, someone is just going to put aside that difficult task, assuming that someone else understood and did everything correctly. Then, when the problem is finally identified deep within the system, the company has a massive liability issue.

Then, tomorrow we have another crazy idea (say, increasing GST on “unhealthy” products) that adds in another level of complexity again. And systems suddenly become another order of magnitude more complex.

My point is that this isn’t “the intellectual purity of some bean-counter’s ideological schema”. It’s real-world difficulties created by increasing complexities of systems, caused by bright ideas from people who’ve never worked a real day in their life.

Sure, much of this is hidden. It seems trivial to the outside observer, who never sees the actual work done. But it contributes to company overheads, which leads to increased prices in order to maintain margins.

There’s no such thing as a free lunch.


  1. Fully agree. and here’s the obvious rub for Idiot/Savant: Compliance costs ultimately increase the cost of the product.

    What would be interesting is to get a good assessment on what the cost overheads are. When looking at the Fairtax proposal (end all PAYE and company tax and just have a GST set at 21%) they produced some good documentation to show that ending company and income tax would simplify accounting and compliance costs so much, we could expect a 20% average decrease in the cost of goods. This was in America, where the tax code is into millions of pages and is simply unreadable in its current state.

    Even in NZ, wiping out so much accounting (because a single rate GST is actually easy to administer, but PAYE, FBT and depreciation becomes far more challenging) I would expect we could see the cost of goods drop by 10%, which would balance out the GST costs given also everyone gets more money in the hand.

    1. Good point – my preference has been to eliminate GST, but you make a sound case to go the other way and make it the only tax.

      1. Yes, it’s a fairer tax than the wages and income tax. I think simplicity and less tax streams also keep things fairer. I’m against GST going up at the moment, because it’s in exchange for introducing new taxes, and PAYE going down. However, that was the exact basis GST was introduced in the first place, in 1986. A few years later the 10% rate went to 12.5% and in 1999 Labour increased the top PAYE rates again, putting us back where we started AND having a 12.5% GST to live with.

  2. This also assumes that a retailer has a complex bar code system. How does this apply to small retailers who do not always have bar code systems. How would it apply to a small cafe that makes its own food. How do they decide if an item is healthy or not healthy and then bar code it in, if they even have a bar code system.

  3. Anyone who watched the introduction of GST in Australia (I wrote about it for the Australian Financial Review) will remember the fuss about Chickens and GST.

    A frozen chicken in a supermarket is zero rated because it’s a basic food, Kentucky Fried Chicken has GST because its cooked food and therefore in the same category as restaurant food – a luxury good.

    But what about a cooked supermarket chicken? Or Chicken roll from the deli counter. Or smoked chicken, or chicken nuggets.

    The list goes on and it will drive you – and retailers crazy. The Australian Tax Office had to draw up complex guidelines about what was in and what was out.

    There were (I don’t know if it ever got fixed, as I came back to NZ) other weird things – a bottle of Coke in a supermarket was tax free (it counted as food!) but in a take-away it had GST. I remember a discussion about GST if it comes from a fridge.

    I think you can see where this is going.

Comments are closed.

%d bloggers like this: