29 March 2013

Newstart allowance and the vanishing cost of living increase

Note: This has been amended from the version I originally posted on 29 March 2013.  There was an error in the original, making a bad story even worse.

In an earlier post I wrote about what I thought was an error in the way the Government’s new Clean Energy Supplement (CES) had been put in place.  My contention was (and still is) that as a result of a wrongly implemented income test change, the promised level of carbon pricing related compensation has not been delivered to some couples.
It subsequently occurred to me see whether the impact of this was so significant it compromised not just the overall level of compensation, but also the basic CPI increases that were implemented on 20 March.  It appears (fortunately) that this isn’t the case, although the implementation does seem to have eaten away much of the 1% cost of living adjustment.
That last sentence probably needs some clarification.

There’s a difference between what is measured by the CPI (or Consumer Price Index) and changes in the “cost of living” for a particular group.  Getting too far into that subject would involve all sorts of discussion about what is meant by inflation as opposed to specific price increases and I don’t want to go there.  Suffice to say, it’s well understood that changing income support payment rates in line with movements in the CPI might well preserve the value of those payments in terms of the overall purchasing power of the dollars involved, but still not keep up with price increases experienced by households on income support.  That’s why pension payments are adjusted by reference to a pensioner and beneficiary living cost index (PBLCI) as well as the CPI.
Treasury calculated that imposing a price on carbon would add around 0.7 percentage points to the CPI.  But they were also aware that the cost of living impact for those on low incomes (eg, on income support payments like NSA) would likely be rather greater than that.  Accordingly the household compensation arrangements for income support and family payments (like family tax benefit) don’t just follow upward movements in the CPI.  To allow for the additional cost of living impact those payments are also increased by an additional 1%.

This 1% is sometimes referred to as a “real” increase by virtue of the fact that it exceeds the CPI.  A similar arrangement was put in place in 2000 when the GST was introduced – it too included a real (ie, above CPI) increase.
But back to the main issue.  As noted above, I wanted to see whether the implementation of the CES had also undermined the standard CPI adjustment.

To determine this I did a comparison between three amounts:
  • the Newstart allowance (NSA) rate that would have been payable if the Government had not introduced a separate CES but instead had just used the standard indexation arrangements.  In doing this I did not include the special 1% real increase;
  • the NSA rate that would have been paid if the Government had increased rates by inflation plus 1%, but in the standard way rather than via a separate CES;
  • the rate actually being delivered, including the CES
The CES arrangement is, among other things, a way of making it clear that an extra 1% has been added to rates on top of the CPI result (details about the creation of this new rate component are in this post) but in theory does not otherwise change the total payment.  Given this, you would expect to see higher rates under the implemented arrangements compared to the standard CPI model – about 1% higher.  You would also expect the delivered rates would equal those under a standard CPI + 1% approach.  At maximum rates of payment (ie, where private income is zero) this is indeed the case for both propositions.  As per my earlier post I was also expecting to see the gain from this 1% real increase somewhat eroded by the income test “mistake”, and it is - but rather than “somewhat eroded” it’s almost completely wiped out.

Here are a couple of charts to show the amounts involved and the income ranges where it occurs.

Chart 1

This chart shows how much the couple's income support (in this case NSA) increased on 20 March under the arrangements actually implemented - a CES with modifications to the CPI increase – along with what would have happened under standard arrangements and also standard plus 1%.  As you can see, there's a significant fall at incomes over around $840 a fortnight, almost down to the straightforward CPI result.  In other words, the extra 1% cost of living or real increase is almost completely lost.

The second chart simply shows the gain or loss relative to the standard CPI model

Chart 2

Here you can see that where the CES model delivers a better result (as was intended), it's around $9 a fortnight more than straight CPI alone.  It’s also (as expected) identical to a straight CPI plus 1% (ie, there's zero gain or loss).  But once that $840-odd threshold is crossed things go badly awry.  It’s a little over $7 less than what would have happened if they had done a standard CPI with a 1% increase.  It’s also only $1.50 more than a straight CPI increase would have delivered, effectively torpedoing the intended 1% real increase.

As I mentioned earlier, this is the second time a cost of living compensation arrangement has been put in place after a major system rejig, the first being the GST in July 2000.  However, back in 2000 they managed to get it right. 
I still puzzle over why a Government that is doing so very badly in the polls and is so sensitive to criticisms about cost of living pressure, particularly carbon-price related, would do this.  The answer, I think, is that they didn't.  At least, not deliberately.  I suspect the problem is simply a cock-up that could be remedied (I was going to add "easily", but things are often not that simple).

I wait in hope for a fix, because I still prefer to think it was not a deliberate act!



  1. Peter Whiteford31 March 2013 at 13:15


    I probably didn't think about this enough earlier, but is/was the problem analogous to the compensation for the GST? Compensation is straightforward when you have no money apart from your basic payment - just up the rate by whatever you think is the relevant measure of the price effect, CPI or pensioner price index. It is also straightforward to compensate people who are paying enough income tax to be above their "effective compensation thresholds" - just raise the tax threshold enough. However people who are on payments but below the effective tax compensation threshold have a problem because there is no direct instrument for compensating them for the money they receive that is in addition to their basic payments but still leaves them below the tax compensation threshold. With the GST, this was got around by lowering withdrawal rates and having new lump sum bonus payments.

    Perhaps because the dollar amounts appear relatively small this was not thought about this time - but perhaps it was deliberate?

  2. Hi Peter,

    This particular problem is more straightforward. It hinges on whether the CES should be included when working out the income point at which one partner's income reduces the income support of the other (ie, the PIFA as per the earlier posts).

    For all indexation increases since 1995 the PIFA has been increased as a flow on effect of a rate rise. With GST change the same happened - the PIFA rose as a flow on from the extra GST compensation amount.

    This time the PIFA has not been adjusted in the "correct" way and so the intended increases are lost (or substantially reduced) as soon as the PIFA is crossed.

    In a very real sense the problem stems from the decision to separately identify the carbon compensation. Had they not done so (as with the GST where for benefits it was just an "unlabelled" increase) there would have been no problem. I've described this sometimes as making the affected couples pay for the advertising that the CES embodies.