The GST provides governmental revenue, which if I read the news correctly, the government – state and federal – are spending in a bid to stimulate the economy, as a whole, create jobs, etc.

It doesn’t provide government revenue (pretending that something was created out of thin air). Like any tax, it diverts private money into government hands. Private people also spend money (after all, money is no use if you never spend it) and stimulate the economy, creating jobs etc. Do you have any evidence that government spending would be more efficient at achieving this result?

More importantly there will be a whole lot of things that are not price sensitive – food being a classic example, you cant stock up on fresh food, for instance – you buy it as you go.

People with more money tend to buy better quality food so their spending can increase even when their energy intake remains constant. People with a real lot of money pay someone to prepare their food and pay for a nice place to sit and eat.

From the perspective of the programmer of the system that has to pay WST you’re absolutely right. There were (from memory) around 80,000 of them. After that the whole tax goes through the supply chain no questions asked, and therefore with no additional bookkeeping load.

There are now (again from memory) over a million GST payers and GST is paid on tiny invoices and each time something goes through the supply chain, another invoice, another GST to record and reconcile. So all up, the GST involves a lot more admin costs than the WST (but not admittedly for the sorry few who pay it.

A small but far from trivial point is that the GST is administratively intensive. And changing the rate means at the very least reprogramming everyones accounting system, dealing with a bunch of transitional issues and for really small operators like small stores, you might have to wander through the shop sticking new price labels on every item. All this for a price change of 1.5% some of which wont even be passed on.

Compared to the old Wholesale Sales Tax system, GST is a walk in the park. The only thing hard about GST is that some items get it and others don’t so you have to track each item through the system (otherwise you could ignore the items completely and just 10% the total).

When I was doing GST transition programming, I always made sure that the 10% was a global variable somewhere in the system. From talking with other people doing similar work, everyone expected the 10% to change fairly soon. One thing that dealing with government really beats into you is never trust what they say, and expect them to change their mind quickly.

If they really want to increase programmer’s wages, they might consider setting different rates for different line items. Ha Ha Ha!

Yeah. This can actually be far nastier than you might expect. I was in the UK in April 1990 when they increased the rate from 15% to 17.5%.

At that point I discovered two nasty problems with the accounting package I was using (cheap, not a lot above shareware but nonetheless I’d paid for a license)

1. I couldn’t change the VAT(GST) rate myself but had to wait for them to release an upgrade (for which I would have to pay)

2. It was programmed to assume that your financial year was the same as the tax year* and assumed a constant rate throughout the year. Change it mid-way? Welcome to month after month of manual adjustments.

I hacked my way round the first (which took a couple of hours), but the second I couldn’t defeat.

* For those not familiar, it is common in the UK for small companies to have financial years that differ from the tax year – this allows you to start a company in March (say) just before the budget, but only be required to lodge one return rather than two. You still have to comply with the conditions in the different years but the accounting costs are reduced.

In any case, on the GST issue, too many have ruled it out as untouchable. That is dangerous thinking. Instead, we need to actually run the cut through a macro model like the British did and work out whether it is worth it. Then, of course, unlike them, we will have to persuade all of the States to adopt the change. Not an easy task but perhaps with all of the extra money they have been given …

On your point one – I’m afraid I’ll have to know more about your argument. But here is the relevant part of an abstract from the literature.

We use store-level data to document the exact process of changing prices and
to directly measure menu costs at five multistore supermarket chains. We show
that changing prices in these establishments is a complex process, requiring dozens
of steps and a nontrivial amount of resources. The menu costs average
$105,887/year per store, comprising 0.70 percent of revenues, 35.2 percent of net
margins, and $0.52/price change. These menu costs may be forming a barrier to
price changes.

The Magnitude of Menu Costs: Direct Evidence from Large U.S. Supermarket Chains
Daniel Levy, Mark Bergen, Shantanu Dutta, Robert Venable
Quarterly Journal of Economics August 1997, Vol. 112, No. 3: 791825.

On your second point – yes, I’m talking about targeting a fiscal stimulus where it can do the most good. When you’re taxing you’re generally trying to influence behaviour as little as possible. When you’re changing taxes or pursuing other measures to stimulate the economy, you’re trying to do the opposite – you’re trying to pull forward activity (consumption or investment) as much as you can. You with infrastructure – some of which I would presume would be privately financed – for instance in telco. Anyway, it seems we differ here, so that’s OK.

On your third point – I don’t think you’re right. In tax policy one is almost invariably trading off one non-neutrality for another. One then has to decide how to play it based on the releveant tradeoffs – one can’t rely on ‘in principle’ arguments’. I’ve argued that despite its rationale in improving neutrality between debt and equity dividend imputation increases other non-neutralities – most obviously the level of company taxation on domestic and foreign equity. Improving neutrality between domestic equity and debt results in more than $20 billion in revenue foregone. I think that revenue foregone is better spent reducing other non-neutralities, for instance by lowering company tax to perhaps as low as 19%.

As always, I’m happy to be corrected. I may well be wrong, but so far at least the way I see it, things are holding up.

First, you need to read the menu costs literature more closely: the point of that literature is that such costs are not incurred. So how much will be lost in menu costs? My guess would be 0.

Second, not targeting? I read housing and cars in your post.

Third, two words: “dividend imputation.”

I much prefer infrastructure spending as an economic stimulus and I think that the Federal Government’s decision to hand money directly to local Councils will give good results.

The answer to your question is in the post. (not in your post-box, but in my post!)

Joahua,

OK, you score a point for my stupid use of food as an example. It was just an example – that is it was used to illustrate a point, that a 1.5% effective subsidy of something that is a typical recurrent expenditure and unlikely to be much affected. So I should have used groceries (and dare I say it, petrol) for the example.

But it looks to me that most of the rest of your response is point scoring (ie for the sake of it).

You dismiss menu costs because the petrol retailing industry is set up to make changes to its prices daily. Other industries are not. Anyway there’s a substantial literature on menu costs, and there are plenty of people who think they’re of some significance. If they are of some significance, I wonder how much of the 1.5% will be consumed in transactions (ie menu) costs.

I have no desire to target particular industries. Nor am I here defending every part of the government’s package – like giving the money to pensioners or increasing the first home owners grant to those buying existing homes.

The criterion of the efficiency of a fiscal stimulus is bang for your buck – how much stimulus you get for a dollar spent. I expect a lot of a GST cut would not change behaviour and so would be less efficient as a fiscal stimulus than temporary measures to pull forward expenditure on a range of things of the kind I mentioned.

I’m intrigued by your comment “You are rejecting that in virtually everything you are recommending on the tax side these days.” I’d be happy to discuss that perhaps in another post, but I don’t know what you’re referring to – could you please elaborate?

It is completely unclear that it is more expensive or less efficient than alternative measures. For instance, the cash payment to pensioners is clearly neither.

On the accounting nonsense, a simple software patch will allow for a different GST rate. That is a hardly a big cost. And what is more, there is no requirement that your prices have to follow GST tax rate cuts so there is no issue there. After all, petrol prices are fluctuating and in my reading stores are having no trouble at all changing their prices.

So you want to target different industries for tax breaks? Give me a break. Those are distortionary and do not have broad implications. To do this under the guise of macroeconomic policy would be an outrage.

The first big principle of taxation is minimise distortion. You are rejecting that in virtually everything you are recommending on the tax side these days. The problem with temporary targeted tax breaks is that they tend to stick. So they are not a stimulus at all.

That said, I can see the case for using direct government expenditures thoughtfully, such as on infrastructure for the environment or telecommunications, to bring forward stimulus.

Also, on a different note, what is the argument for reducing GST rather than income tax, the second of which is easy to stick up again via stealth later on?