Some thoughts on benefit rates

Follow up to the 2024 Economic Inclusion Committee Report

Following along with the upcoming Australian Budget annoucements I’ve noticed that one of my favourite topics has come up. Should we increase the level of the unemployment benefit?

Here I’d like to explain how I think about this question, and the type of evidence I look at. Note: The evidence is in a separate post here.

Lets get some points out of the way. It looks like the government will increase payment rates. It looks like the vast majority of commentators and organisations agree that it should be increased (note the Economic Inclusion Advisory Commission Report). And my personal value system would involve higher benefits and taxes.

But this blog isn’t about making a pitch for a specific policy. We want to figure out what information and arguments are relevant for you to form your own views - views that could also push against an increase.

Let’s do this. But first a tl;dr as this is very long - including a separate graph post.

tl;dr the determination of the right benefit level is tricky - and does depend in part on personal values. But we can establish a base set of facts. These are that at current benefit rates: the income replacement is moderate to low relative to history and overseas, work incentive changes to an increase appear modest, and the fiscal cost is likely to be similar to the moon-shot industrial policy being discussed.

However, the relative payment is reaching a level where we should start to think more carefully about trade-offs. This also doesn’t suggest that substantive increases should occur each year - instead we need to reevaluate these facts as we go, and make a clear case to the public regarding why changes are appropriate.

---

How do we think about the benefit level?

Every year the benefit rate increases with inflation. After COVID it was increased by more ($50 above CPI), last year it was increased by more ($40), and this year it will be increased by more as well. But what do these increases mean, and what would we imagine to be the limit of benefit increases?

Lets start with the benefit level. What does the unemployment benefit reflect?

I’d argue that we all agree it represents a minimum income that, as a society, we agree every family should have access to. I say family, as we pay this to a family (and the payment changes by family size) and we consider this with reference to the sharing of resources and means of the family.

So the concept here seems pretty simple! Let’s define what that minimum standard is and just pay it!

There are two ways that people do this:

1. Define a basket of goods and services and set the benefit to cover it. Then increase the payment with that basket.

2. Define a proportion of the average persons income that we view as a standard. Set the benefit to match that.

Across Australia and New Zealand this first conception has been the driving force for the benefit level since the 1930s. It also informs the work by UNSW and ACOSS talking about potential benefit insufficiency, and the Social Indicators provided by Melbourne Institute.

However, a tricky thing about this is determining how - or why - the basket might change through time. Or even what is in the basket! The Mother of all Budgets in New Zealand slashed benefits in the early 1990s, but based these changes on an updated basket of needs.

Another way to think about it is that the argument about indexation reflects how we expect the basket to change.

1. Indexation by CPI (or at least a value of CPI that represents the cost of living change at that point in the income distribution) is a measure of absolute poverty. Doing so involves saying that the basket is a fixed set of goods and services, irrespective of the means of others.

2. Indexation by WPI (or similar labour earnings measure) is a measure of relative poverty. This tells us that the basket of goods and services moves with the basket of goods and services that could be received by the average worker.

Conceptually, there are relative elements (a cell phone has become a necessary cost of living through time) and absolute elements (the cost of housing services is the cost of housing services). As a result, indexing by CPI and then introducing discretionary changes in the payment rate based on a new basket appears the most accurate.

However, in the real world of political economy - and given subjective debates about what is a need vs a want - keeping the minimum income standard at some proportion of wages implies that these debates won’t be needed. And ensures that the gains from growing productivity are shared with those unable to enter the labour market at present.

Why not just ramp this benefit level up close to average wages?

If we could improve peoples financial status without any cost we would do it. But that isn’t the case. As noted in the past, there are two key things to keep in mind as benefit levels increase.

1. It influences labour market choices: They increase the value of not working - with potentially postive effects (the ability to spend time finding a good job match, power to bargain against a bad employer) and negative effects (the decision to not work when reasonable options are available).

2. There is an opportunity cost for, and behaviour by, the taxpayer: The payment is funded by taxes which implies income earning individuals have to contribute to this rather than other things (infrastructure spending, personal consumption, saving) and that their decision to earn may also be negatively influenced.

A lot of arguments boil down to views around this - however these are things we can measure somewhat!

So how does the data look

Now that we know the ideas, we can look at the data with clear eyes. And we will do this with a post that is released about 15mins later - as Substack says I have too many pictures. See you all over there!