07 November 2013

Spiky EMTRs can also be prickly - how to really stick it to students

A little over 18 months ago I wrote a post about the way in which the Austudy income test seemed to me to be only half-built, causing substantial financial losses for some couples.  I finished it with a comment that the problem would only get worse over time.  There will be a few changes to Austudy payment from 1 January 2014, so with that earlier comment in mind I thought it might be interesting to see how the problem looks once these are in place.  And here’s a spoiler – it isn’t pretty.

First up, these are the 1 January 2014 changes I’m referring to:

·         Austudy payments and the associated student startup scholarship will increase as a result of the standard annual indexation of rates, based on movements in the consumer price index.

·         Austudy rates will also increase due to the implementation of the clean energy supplement (CES).  There's some background on this here.

·        those with private income may also benefit from the indexation of the Austudy income test.

Once these three changes are in place we’ll end up with the combination of Austudy payment, student startup scholarship and income testing shown in Chart 1.  In it I’ve covered three household types – single student; couple, both students, with a 50:50 private income split; single income student couple (100:0 private income split).

Chart 1

Each of these traces starts out as would be expected for an income tested payment package: as private income increases there’s initially no change in the payment amount, then once the income test free area (the amount of income allowed before payment reduces) is exceeded then the lines trend downward.  However, all three traces have an abrupt reduction at the tail end of the payment range, reflecting the “sudden death” removal (ie, non-tapered loss of complete entitlement) of both the student scholarship and the income support bonus.  (For the single income couple - 100:0 income split - there’s a short income range on the end where an entitlement to the low income supplement may be available.  It’s part of the household compensation arrangements introduced with carbon pricing.)

Note though, that the single income couple also has a noticeable “notch reduction” (ie, non-tapered removal of part of an overall entitlement) at a private income of a little under $30,000 a year.  This is a result of three things:

·         The loss of partner 1’s (P1) student scholarship

·         The loss of P1’s income support bonus

·         The operation of the Austudy income test on partner 2 (P2) at the point where P1 exits Austudy payment

Of these, the latter element accounts for well over half the total loss at that point.

These abrupt reductions, be they sudden death or notch style, can lead to significant financial losses for households whose private income crosses the relevant threshold.  In turn, this introduces disincentives to increase private income.  They are also problematic from an equity perspective – identically constituted households with almost the same levels of private income can end up with wildly different disposable incomes.  Yet despite these issues, around one-third of the entitlements of single income couples are withdrawn in this fashion.

One oft-used way of looking at possible financial disincentives in the tax-transfer system is to use a chart showing the effective marginal tax rates (EMTR) encountered as private income rises.   Chart 2 shows the EMTR for the single income Austudy student couple we’ve been looking at.

Chart 2

If we ignore the spikes in the chart for now, the EMTR is not that different to that of other single income couples getting some kind of non-pension income support payments (eg, Newstart allowance).  It has the characteristic series of ever-higher plateaus, reaching nearly 100% at the end of the income support payment range, with a marked reduction once income support has been fully withdrawn (in this case, once the Austudy to P2 has been removed).

The real story is in the spikes.

Spikes show up in EMTR charts where payments are withdrawn in an abrupt or non-tapered way - at the point where the withdrawal occurs the EMTR goes off the scale.  This is one of the (many) weaknesses in EMTR charts because a spike like this conveys very little in the way of useful information about the impact of the withdrawal.  The fact that the spike is concentrated on a single income point rather than a range of income can lead to commentators and analysts being somewhat dismissive of EMTR spikes.  There’s an idea that once private income has crossed the EMTR spike, things settle down again and so it’s not a problem for cases with income not actually at the spike point.

This can be true.  The off-scale aspect of EMTR spikes occurs even where the abrupt loss of income is small.  For example, some income support recipients receive an increase in their payments known as pharmaceutical allowance.  It’s only $3.10 a week for a single person ($1.55 for a member of a couple) and is withdrawn in a sudden-death fashion when income exceeds the relevant limit.  The loss of this fairly trivial $1.55 will still show up as an off-scale spike on a typical EMTR chart.  Understandably, a spike due to losses like this can be dismissed as illustrating no more than a possibly clunky design element in the system.  But there’s a risk of throwing out the proverbial baby with the bathwater if all EMTR spikes are dismissed in this way, and in the Austudy case they are significant indeed.

Chart 3 looks at the disposable income of the Austudy single income couple and includes a breakdown of the tax-transfer elements that go toward it at various private income levels.

Chart 3

Here the EMTR spikes shown in Chart 2 translate into reductions in disposable income.  The first three spikes illustrate the point I was making earlier.  The first, at around $29,000, shows up in Chart 3 as a significant fall in disposable income, one which leaves the couple worse off even if private income further increases by up to $20,000 a year.  Clearly it’s a spike in the EMTR that cannot be brushed aside as simply something that has an effect at a single point.  The second, at around $40,000 is still apparent in its impact on disposable income, whereas the third, at a little under $45,000 (caused by the loss of the low income supplement) is barely discernible on the disposable income chart.

So, what do these various abrupt reductions in assistance do to attempts by our couple to increase their disposable income?  What do the carefully designed tax transfer arrangements do to promote the laudable goal of increased self-sufficiency?  The disposable income chart makes it pretty plain that some increases in private income are associated with falls in disposable income and we could leave it at that.  But I’d like to hammer the point that this structure, this so-called design, is atrocious.  To that end, here’s another way of looking at it.

Chart 4 uses one of my “effort” charts (described in more detail here) to depict the terrain to be traversed as the couple attempts to increase their disposable income.

Chart 4

For a little over the first $10,000 or so the couple keeps every extra dollar they earn – the terrain is flat.   To increase disposable income after that, however, is a different story – there’s an increasingly steep hill in the way.  The climb is pretty tough to increase disposable income by a little over $16,000, but then things go from bad to worse.  A little extra effort sends the couple crashing back to a disposable income gain of around $10,000 due to the government’s clawback (either increased taxes or reduced transfers) suddenly increasing by well over $5,000.  (The red line connecting those two points indicates there are no intermediate values – it’s an abrupt change.)

This combination of a tough climb (reflecting the underlying high EMTR) and a sudden loss continues with the couple never managing to cement itself a bigger gain in disposable income than approximately $16,000 until the government has first extracted another $25,000 from the couple’s increased private income.

It’s the kind of terrain a mountaineer might enjoy, but it’s not pathway that encourages progression to higher income destinations.  In fact, this kind of design (I use the term loosely) may actually encourage reduced workforce participation.  Consider that if the couple needed an extra $10,000 in disposable income they could achieve it by earning either $10,000, almost $30,000 or almost $42,000.  Given they are full-time students, which option would you choose?

Chart 5 picks up this theme of increasing disposable income by working less.  It shows before and after private incomes.  If you look at where these two cross there’ll be a colour – white or blue.  If the intersection of the before and after incomes is in a blue zone then the couple’s disposable income either increases or remains unchanged by reducing their earnings.   To me, it illustrates that this perverse incentive occurs over a ridiculously large combination of incomes.

Chart 5

You’ll note that there a number of tiny (and one not so tiny) “islands” shown in the chart that track along the rising diagonal.  These are a result of the HELP/HECS scheme repayment arrangements – my calculations assume the existence of HELP/HECS debt.

Are there any signs that this design issue is going to be fixed?  There are two things (perhaps three) that could affect the current outcomes.  One is the Government’s stated intention to abolish the income support bonus.  This will reduce the disposable income of low income households, albeit not substantially.  The second is the possibility that the student startup-scholarship will become a repayable debt, to be recovered using the HELP/HECS arrangements.  On its own that doesn’t change the results shown above, but it may discourage some students from taking the scholarship, which would reduce the size of the payment package used in my calculations.  Neither of these is really a fix though.

The third is somewhat tenuous.  As seems to be the norm when there’s a change of Government, there’s probably going to be a review of social security and similar arrangements, possibly headed up by Patrick McClure.  Maybe, just maybe, the issue of how to properly income test couples might get a look in.  But I’m not going to hold my breath.

No comments:

Post a Comment