sign up log in
Want to go ad-free? Find out how, here.

Matt Nolan looks at how taxes on consumption and capital income work, and how they differ from income taxes

Matt Nolan looks at how taxes on consumption and capital income work, and how they differ from income taxes
Part five of a six part series on tax.

By Matthew Nolan*

As a note, so far in this series on tax the following issues have been touched on:  Why do we tax?  What are the distortions of tax?  What are poll taxes, and how does fairness matter?  What are factor/income taxes?

In the first article in this series it was stated that tax and spending was how government redistributed goods and services within the economy.

Last time involved going through the tax part of this process by looking at taxes on labour, capital, and land and how they change the incentives for labour, capital, and land owners regarding what to produce.

But this all seems a bit indirect when, in essence, our interest is in terms of the underlying goods and services that are produced with these factors of production – not the factors of production themselves.

As a result, why not just tax the goods and services (consumption) directly?

This makes thinking in terms of consumption taxes logical.

Consumption is the “final purchase of goods and services” at which point we either eat them up then, or eat them gradually over time without resale.

It is the consumption that we value as individuals, we earn income by producing goods and services specifically because we want to consume.

And so it is the ultimate “ability to consume” that we are really taxing, and more broadly it is consumption we are considering redistributing in the name of fairness.

Consumption taxes

As a starting point, let us treat consumption as one gigantic block that everyone is involved in producing and consuming, although consumption and production by individuals occur at different points in time.

When it comes to taxation we reach another potential view of fairness:  we would like to avoid taxing consumption at different points in time differently.  This is really just an example of horizontal equity which we mentioned here.

Although we can talk about this in terms of fairness, this principle is largely talked about with regards to efficiency – by treating consumption over time the same way, we are not distorting the incentives about when to consume.  We will get into this more below, but first let’s expand the amount of commodities we could tax.

Frank Ramsey worked out that if, instead of taxing all commodities in the same way, we had relatively higher taxes for goods whose quantity supplied and demanded are relatively less responsive to tax, we can lower the deadweight loss/inefficiency from taxation. 

However, there are two issues with this. Firstly, the information required and the compliance costs involved are likely to be prohibitive (for example, here is recent work solely on income elasticities in New Zealand). 

Secondly, by viewing things solely in terms of efficiency we have again strayed down the same path as a poll tax or a land tax – we need to think about how this fits into our earlier principals of vertical and horizontal equity based on ability.

To help us start to think about these issues, it is useful to compare income taxes and consumption taxes.

Consuming, saving, and capital markets

When we look at our taxing the inputs to producing things we want to consume, we are in a sense taxing consumption – just in an indirect manner.  As a result, we can use these thoughts on consumption taxes to understand a bit more about income taxes. 

To do this lets use the idea that consumption at a point in time is equal to income minus additional saving.

Capital income is the return on capital investment, and the value of the stock of capital someone has will bear a relationship with the stream of expected returns from that capital through time.

These “returns” have value because we can consume with them, and as a result the value of the stock of capital can be thought of as a stock of savings for future consumption. This also implies that investment is really a flow of savings – an addition to the stock of capital/savings.

The return on capital, capital income, is then the return on savings – it is a form of interest rate.  In fact, the return that you get from savings with a bank can be thought of in exactly this sort of way.

Interest rates, or the expected rate of return on capital, are a price in capital markets – which in turn sets the quantity of capital supplied and demanded equal to each other. This price represents a variety of things, including people’s inherent willingness and ability to “shift their consumption through time” (eg borrowing when their income is low relative to future income, and saving when their income is high). 

Taxing capital income creates a wedge between the cost to the borrower and the return to the saver.

However if we think in terms of consumption we can say the following.  By taxing capital income, we place the burden of taxation on future consumption as the tax can be “avoided” by consuming income now instead of investing it.

Essentially, capital income taxes involve taxing people more when they want to spend current income in the future than if they use that income now.  As a result, we are also violating the idea of “treating consumption at different points in time the same way”.

Of course, our distinction between capital and labour is a little bit loose – after all we are not born with a clump of skills and knowledge, we have to accumulate them.

In this way, a tax on labour income also reduces the incentive to invest in human capital!  [Note:  This “wedge” is unlikely to be as large as for capital income, given that the upfront cost of human capital accumulation is lower, and the opportunity cost (sacrificed labour income) is also taxed at the income tax rate!]

The key difference

Here we’ve noted that the key difference between “output taxes” and “factor taxes” is that output taxes treat consumption through time equally (outside of credit constraints) while taxes on the labour and capital income tax future consumption of income more – and thereby reduce the incentive to accumulate capital in its broadest sense.

Now this is not to say that the level of the consumption tax doesn’t reduce the incentive to accumulate capital – it does due to the fact that it reduces the “goods and services” you can buy from the additional income you earn from any investment you make.  But as we are inherently taxing current consumption at the same rate as future consumption we are not “biasing against investment” in the same way income taxes do!

This is not a slam dunk for replacing income taxes with GST, but it is one of the driving ideas behind why there should be a shift from income tax to GST. 

For perspective, one of the costs associated with a consumption tax are that they tend to hurt those with more “volatile” income more (eg students), due to the fact that credit markets are not perfect and future income is genuinely uncertain.  Furthermore, if different countries have different rates of consumption-income taxes, the decisions around where to work, buy consumer goods, and where to retire will be distorted (Note recent online discussions on this and solutions here, here, here). 

Looking ahead

Here we have run through the idea of consumption taxes, and inadvertently come up with the main reason why economists tend to state, that in terms of efficiency, we should have lower capital taxes relative to income/consumption taxes.

Having now run through the basics of income and consumption taxes, and shown the similarities between them, we are now prepared to start discussing the structure of these taxes.

Next time we will discuss what “regressive”, “flat”, and “progressive” tax systems are – and the efficiency and fairness challenges that are associated with them.


Matt Nolan is a senior economist at Infometrics. You can contact him here »

We welcome your comments below. If you are not already registered, please register to comment.

Remember we welcome robust, respectful and insightful debate. We don't welcome abusive or defamatory comments and will de-register those repeatedly making such comments. Our current comment policy is here.


How does Matt Nolan recognise taxes like a Land Tax which has no relationship to either income or on consumption?

Also how does he class a tax like the FIF tax on incme that is substantially notional given hat the actual income does not reach the taxable point for other incomes?


Land taxes were mentioned in the previous article, while poll taxes were discussed in the article before that - they are both linked at the top.

We are only building up the principles involved in different "broad" forms of taxation in these articles - so that we can discuss ideas of allocative efficiency and fairness, and thinking about "trade-offs" that can exist whenever we look at changing the tax system.  I'm avoiding going into extremely specific forms of tax, or making judgments about the types of tax and allocative choices I have a personal preference for.



In future articles will you discuss what I would call ''offset'' taxes?

By this I include 'notional' income such as imposing a tax on assets but allowing that tax to be offset against the tax on income actually earned. This implies that assets held for ultimate gain but having no attached current income, would incur tax at a rate that would demand a cash payment each and every tax year. An alternative to CGT?


A bad dream...banks swapping 'old' notes for new whenever the 'old' notes are notes with new colours and new images each side....and charging a fee to make the change...5% no you like that for a bad dream?


Will you be discussing financial transaction (ie. Tobin)  taxes at some point?


Gareth's is a tempting argument, but the next logical step would be to ask the question "Would we be better off having experts or politicians deciding how to spend government money?" At some point along that continuum we would likely end up being like Singapore. And here I'm not trying to jump on one side of the fence or the other. Singapore is a very well run place, and economically very successful.

But I would probably take our system, warts and all, over Singapore's.

A small step I would take is to have the Reserve bank, or separate at least notionally independent body, determine how to fund any government deficit, with an overall economy monetary objective rather than a narrow balance the government's books objective. At present, as I understand it, this is done by Treasury, who seem to take steps such as foreign borrowing that act directly counter to the RB's view that the NZD is overvalued, in a self defeating process that is very costly to the overall economy. Yet it is not something that the citizenry seem to wish to understand or get excited about. So no need for political input. 


Have posted this on the wrong thread; apologies.



Have tried to fully understand your paper; and may have missed some logic.

You say

Here we have run through the idea of consumption taxes, and inadvertently come up with the main reason why economists tend to state, that in terms of efficiency, we should have lower capital taxes relative to income/consumption taxes.

I can understand the taxing of returns on savings or investments may seem to be double taxationin that the original income to create the investment was taxed. This may or may not seem unfair. That taxing the returns on capital is also likely to discourage an investment all else being equal, is a reasonable conclusion. But all else is not equal if consumption taxes go up instead. So I'm not convinced capital taxes are less efficient for an economy than consumption taxes. In many cases that would seem to depend on the relative elasticities of consuming over investing.

In a real world example, suppose I have invested in a restaurant, which would pay tax on its (probably relatively meagre) income. Now the tax on that income is to go, which would encourage me to make such an investment. Except that the customers of the restaurant now will have say 30% GST added to their bill, rather than 15%. Which is likely to discourage a good number of customers, just possibly taking the business from profit to loss- meaning any tax savings are meaningless to me. If the restaurant caters primarily to wealthy investors who now have a bit of spare cash from their tax savings, then maybe it would balance out because they would come more often. But if its a broader clientele, then probably not.

So such an investment may or may not be more likely to happen; and the economy may or may not be more efficient. It would depend on elasticities. (Your linked Treasury paper was a  difficult read, by the by. Good if they added a simple summary of what it all means).

Separate distortions seem possible in terms of encouraging consumption outside NZ- to avoid the consumption taxes. Which may discourage investment here. But enough for now.




I believed we discussed it in terms of relative elasticities in the last article when comparing types of income taxes - but here we are adding this time dimension and relating it back to the final goods and services like I've been promising, consumption!  Here we added the fact that, by taxing factors of production rather than consumption itself we were effectively placing more of the tax burden on "future consumption" rather than on current consumption - there doesn't seem to be any real reason for that, so it is just a straight loss of efficiency.

You are correct that a GST creates a "wedge" between the price paid and the price received by the sellers.  However, this is independent of time - and does not change the point of time when people want to do certain things.  Income taxes, and the way they influence forms of "capital accumulation", do change when people want to do things.  The idea is that, if you earn a bunch of income now and are deciding what to do with it, then in you tax consumption it doesn't change your incentives to invest or spend from this income - but if you tax things such as the rate of return, or the level of savings/capital you have, you convince people to consume more now and less - which is a "distortion"/"inefficiency".

This idea, and a bunch of estimates about long-run elasticities, form the basis of why economists often bang against capital income taxes.  But this isn't a fact in of itself, just an argument, and my goal was just to make the factors behind it clear - especially when we start trying to broaden out our equity arguments a bit further :)

The Treasury paper was not meant to be an easy read - it was an example of the sort of work required just to look at labour supply elasticities - let alone the elasticity of supply and demand every single good and product in an entire economy!  It was an example of the difficulties associated with compliance ;)