On comprehensive income taxes

How I learned to stop worrying and love taxing everything

Hello hello, long time no see. My apologises I got a bit sick around the time I last posted, and have been playing catchup with family and work things ever since. I will try to slowly build back up from here.

However, I thought it would be fun to discuss a recent micro note that came out from e61 and the interpretation it is receiving - this note on effective income tax rates by Greg Kaplan, Matthew Maltman, and myself. An AFR piece did a good job of positioning this, and Greg’s speech, in the more general discussion around tax reform.

Note: The tax reform roundtable was fun - e61 CEO Michael Brennan also shared a lot of insights based on areas e61 has been thinking about around bracket creep and the State-Federal nexus. When it is available I’ll share links to all the speeches here.

Now I’m going to give you my own view - so if you disagree with me don’t blame my amazing co-authors, as I’ve probably said something dumb or purposefully inflamatory. So lets do this.

tl;dr Taxing income consistently is good and fair. We all need to step back and view government finances - taxes and spending - holistically. Ad hoc hyperfocused interest group based policy making might be good politics, but it is bad for Australians.

Why do you guys want to tax capital gains at full rates?

Why are people saying this? The note, and Greg’s comments at Allegra’s roundtable were quite clear - we should be treating all sources of income consistently irrespective of their source. And we don’t.

This doesn’t necessarily mean taxing capital gains more - it could also mean taxing other capital income less. The main thing is reminding everyone about principles of vertical and horizontal equity, and taking about the myriad of ways (deductions for charitable giving, capital gains discounts, subsidising superannuation) we have moved away from that.

Varela, P., Breunig, R., & Sobeck, K. (2020). The taxation of savings in Australia does a great job of showing the inconsistency of how this works for savings - our charts merely reflect how this then shows up across the income distribution.

So what did you do?

The exercise we showed is part of a deeper project where we are trying to understand effective tax rates in Australia. Although I could cite a number of big authors with publications in great journals to motivate this I want to mention less publicised work that I believe gives a better and clearer path to practically doing this - New Zealand Treasury work on Effective Tax Rates.

When digging into the weeds there are things I disagree with on their treatment of income and tax, as the author will be able to tell you. But the broad principle and structure is strong and pulls us to the question of asking what income is.

So what is income? Income is what you can consume now. Sounds easy. It is not.

If your super balance goes up is that income?

If the government commits to pay a higher age pension in the future for you funded by others is that income?

If the government gives you a benefit and taxes it, is that whole amount income?

If you sacrifice a dollar to save and inflation occurs, does this subtract from income?

How we answer these questions influences how we measure income.

We take small steps towards a more comprehensive measure of income than “what is taxed”. To do this we:

• Add in the discounted component of taxable realised capital gains.

• Disallow deductions that aren’t associated with generating income - viewing them as consumption.

• Add back in voluntary superannuation contributions.

Each of these changes is consistent with the way the current tax system operates - so we aren’t going to some extreme associated with “economic income”, or saying that your leisure time gives a value that should be taxed. Just adding in some of the low hanging fruit that is “of the nature” of the way other income is taxed in the tax system.

And that gives the following:

A progressive tax system, where there is growing differences in the spread of tax rates in later percentiles - akin to the discussion of horizontal inequity in Herault and Jenkins (2024).

But whether something is really “horizontal inequity” depends on whether we are really comparing people in similar situations.

When we look at this spread a lot of it is driven by family structure (tax offsets, medicare levy, non-taxable benefits) - while at the top the spread in such taxes is due to the differential treatment of capital gains and deductions for charitable giving and voluntary super contributions.

The former suggests that individual income is a poor reflection of needs based targeting in the tax system - the later is closer to true horizontal inequity (or treating people in similar circumstances differently).

No but inflation, and we want to encourage charaties, and …

Great stuff. You’ve identified that the tax system is inconsistent in how certain things are treated - which means the ability to organise around this inconsistency can lower someones tax rate without changing the substance of what they are doing.

Wait, that wasn’t your point - ok let me run through this.

I don’t disagree that inflation isn’t income. In fact I agree!

But we tax the inflationary component, or more broadly the normal rate of return on all capital earnings. As a result, the tax system biases us to consume now rather than to save and accumulate earnings.

The taxation of capital gains at full rates is equivalent to the taxation of accumulating interest income at full rates.

In both cases, capital income is overtaxed. The issue is that we are choosing to (excessively) correct for it for one capital income item - and not for others.

Asking for consistency isn’t saying overtax capital income more - it is saying make sure we are taxing such income consistently. If we decide we should remove the inflation component for one type, we should do it for all types.

And on charitable giving lets keep something in mind - we might want to subsidise activities we view as “good”. But these should be treated like any other government expenditure, and evaluted through the same process. Not slid into the tax system to lower “compliance"/oversight.

Otherwise it stops being a reflection of the true social will regarding what is good, and just a means for interest groups to poke holes in the tax system in order to protect their own interests.

So what is income exactly?

Yeah good question. As we said above, income in a period of time is what you can consume. The idea here is that you have an ability to pay that occurs when you have the option to consume the income - and that in turn reflects when you should face the liability for the tax. Taxing this income at a different time is then similar to taxing the income as a concessionary rate - as you receive the time value of deferral.

For actual measurement there is a nice paper from 2001 that discusses concepts of income with respect to household surveys - concepts that are super important for people working with national accounts and income distribution figures.

Lets look at their table:

The income concept used in the note is someway to Gross Cash Income - which is the hypothetical tax base which is adjusted in comprehensive systems, like those in the US, Australia, and New Zealand. This is different to the taxable income as we retain amounts that we deem to be income even though they are adjusted out of the income tax base.

So what exactly do you want us to do!

Well before we say what to do I want to make sure we agree on the principles.

The habit for policy circles in Australia to demand “practical solutions to the problems we know exist” - without having done the framing work to understand the problems, principles, and trade-offs - makes me extremely angry.

Demanding practical solutions immediately doesn’t make you smart, it makes you do the type of dumb stuff that breaks policy and makes things harder for future generations. If this statement upsets anyone please highlight this in the comments or via email, so I can repeat myself directly to you.

But assuming you’ve now bought into the principle of equal treatment we’ve discussed what are the policy levers to look at:

1. Broaden and increase GST - this does not tax that rate of return, and treats consumption through time the same way. (Fair enough if you are concerned about regressivity here - I’ll come back to this in a future post. But first I’ll do a post about Chris’s discussion of excess burdens later this week - neat paper here that he is discussing)

2. Adjustments to capital income that only taxes the real rate of return (i.e. dual income tax system, ACE).

   1. Relatedly, clarifying the demarcation between capital and labour earnings.

3. Then remove the random capital income subsides around the place.

Things like trusts I find tricky. My understanding if that the benefit of trusts stems only for the ability to “spread” income among multiple people - so it is a function of the progressive scale. This income splitting is definitely tax minimisation, but it also raises the issue of whether it is the individual - or the family - that reflects the right unit of analysis for our tax system. Would be interested in more reading on this to understand the scale of the concern!

Why aren’t we worried about inflation and labour earnings? That is also an interesting question. The idea is as follows - when you sacrifice the use of a dollar, inflation reduces the real value of that dollar and you require compensation to maintain its value. Therefore there is a stream of nominal income for capital that isn’t about consumption.

When you earn labour income you are sacrificing the use of time. Time is a real not nominal economic variable - so the value of time is invariant to inflation.

Now a very fun implication of this is that there are capital gains generating activities that are “more like labour income” than bank interest - i.e. renovating a house, building up a small business and not paying yourself wages. These do not need compensation for inflation … and so there is a bias towards doing these instead of spending your time as a wage worker.

Wait, you want to increase taxation on wage earners?

Lets start from the start. If we want to maintain government spending as a higher percent of GDP we need to raise the revenue - if we don’t we could cut spending, but such cuts involve deciding where to reduce services. Calling the government a leviathan and pretending public servant numbers are to blame doesn’t help.

If we as a society don’t want to start cutting government services in health, aged care, and the NDIS, then the revenue to fund these programs needs to be raised in the most efficient way possible.

This likely involves more taxation of wage earners irrespective of how we do it. We have choices about where we put this burden (i.e. whether it falls more on high income earners), and there may be ways we will tax “economic rents” efficiently that can lower the burden a little (i.e. land taxes, wealth taxes, supernormal profit taxes). Furthermore, since some capital items are oversubsidised the case that these items are favoured while both labour and other capital items are not may still hold.

These are debates people can have.

But having a principle that anything that taxes capital more is the solution - when the inflation component of capital income is not income - is bad policy and bad economics.

So if we want an efficient tax system, and we want a larger government, we’ll all have to pay a bit more ultimate tax on our labour earnings - either through income tax or through GST. And when we do this we want the burden to be fair - people who struggle need to be given a fair go of it.

There is no money tree of ultra wealthy people that can pay for everything we desire. And if you think that way you are delusional. I’m not here to feed your delusions - and hope other Australian economists wake up and tell the policy entrepreneur community saying this to live in reality.

The Australian people are adults, and we should be honest with them about these trade-offs. If we were, we’d likely find that they are a lot more comfortable with reform than a lot of the political junkies that dominate the airwaves.