On compensating for a change to GST

When discussing the upcoming changes to the New Zealand tax system the National party has made it clear that they want any change in GST to be “compensated”, so that those on low income aren’t “worse off”.

Now this is actually a wildly complicated question.  Any change will create winners and losers, that is undeniable.  If we think of compensation in welfare terms there is no way to perfectly compensate everyone without making any change viciously complicated – or not making any changes at all.

My presumption is that the goal isn’t “welfare compensation” per see.  My guess is that any compensation will be such that people (outside the property sector and those in the top income bracket – where a rejig “might” take place) will pay the same proportion of their lifetime income in tax (although any progressivity from any change in incomes will also be accepted).

Why do I think this?

Ultimately, I think that the government believes that the fair burden of tax should depend on someones lifetime income.  They seem uninterested in changing the progressivity of the system per see (although this may change on budget day) and as a result it appears that the goal of the tax shift is to move from taxing when the income appears to taxing when the consumption from that income occurs.

As a result, all the government needs to do is treat the increase in GST as a one-off increase in the price level.  Increase nominal transfers (the dole, WFF) and their abatement thresholds on the basis of this “price shock” and increase income tax thresholds based on this price shock as well.  Then cut all income tax rates by the same amount to compensate for the increase in GST (such that the implied rate from GST and income tax is the same as before).

Now there is a common claim that people on low incomes borrow – so they pay more in GST then they do in income tax (this is implicit in this discussion here).  This is true.

However, we have to ask why people borrow.  People borrow because:

  1. They expect their future income to be higher,
  2. They want to consume now instead of in the future (tilted consumption),
  3. Consumption in durable goods is lumpy (you use a dishwasher for a while – but the cost of buying one occurs all upfront).

This all implies that people are borrowing out of future income.  Given that an individual will consume their income over their lifetime, for a given level of income the mix of GST and income tax doesn’t matter (in proportional terms).

In fact, it is fair to say that people with “low incomes” probably have low incomes relative to their lifetime wealth level (think students), if this is the case then we would expect more borrowing from that group – and the change in tax mix just means they are paying a greater proportion of their lifetime tax now instead of later.

So when we include the idea of borrowing and lending, we have to shift to looking at “lifetime wealth” instead of just static wealth when we are talking about a compensated tax system.

So, this is where I think they are going with “compensation”.  Now there are welfare costs (and benefits) that this does ignore – and I will discuss one of the major ones a little later on.

3 replies
  1. rauparaha
    rauparaha says:

    I agree that it will be impossible for them to perfectly compensate for welfare, but I interpreted Key differently. I thought he was saying that ‘compensation’ was to ensure fiscal neutrality, and then they’d biff some more money at low income people to partially offset the GST cost. I’m not convinced that National’s analysis is as nuanced as you’re inferring!

    Substantively, I think it’s also important to remember that the lifetime income hypothesis doesn’t have a lot of empirical support. People’s consumption tends to track income far more closely than Friedman predicted. So, if the government IS thinking about it like that, then they should try consulting the data more closely.

  2. Matt Nolan
    Matt Nolan says:


    “I’m not convinced that National’s analysis is as nuanced as you’re inferring!”

    If they are thinking along the lines of the tax working group, then I think this was the road they were moving down (that was my impression from reading papers).

    On the welfare idea, I intend to right about credit constraints later – shifting to GST has a welfare cost for people facing credit constraints. Given that credit constraints are even worse now, it isn’t exactly the best time to introduce the shift 😉

    “Substantively, I think it’s also important to remember that the lifetime income hypothesis doesn’t have a lot of empirical support”

    Consumption smoothing doesn’t – but the assumption required in the above post is weaker, it is that lifetime income does provide a binding constraint.

Trackbacks & Pingbacks

  1. […] tax might, at the margin, reduce immigration.  Now when the income tax cut is measured with an increase in consumption taxes, this argument becomes a lot […]

Comments are closed.