Monday, 17 December 2012

The effect of marginal tax rates

It has sometimes been suggested that marginal tax rates in Ireland are too high, and that raising them would harm growth. They may affect the incentive to work. It is true that they are 10th highest, of the 34 countries in the OECD. Also the marginal rate hits in relatively low (affecting average earners). Helpfully, the OECD provide the data here.


However, what is the link between marginal tax rates and the economy?
 
As can be seen in the above graph there is no negative relationship between GDP per capita and marginal tax rates. In fact it is slightly positive. Of course this is not conclusive evidence of a positive link, but the evidence certainly does not support the hypothesis that high marginal rates harms GDP.

But does it affect the incentive to work?

No relationship is found between unemployment rates and marginal tax rates in the OECD.

High tax rates are not the problem. In other countries, such as the Nordic countries, high tax rates are used to fund social services such as child care, which make it easier for people to work in the market economy.

7 comments:

Anonymous said...

Really, Rory, you compare marginal tax rates and unemployment levels and conclude there are no dis-incentive effects?

Is there any kind of peer review before stuff gets posted on this site?

High marginal tax rates only impact on above-average to high earners. And the decision point they face is not between working and unemployment, it's between working more and working less!

(Or even coming here to set up an outpost of some MNC as opposed to staying back at HQ. Or staying here paying tax versus joining Denis O`Brien in sunnier climes.)

If you want to study the disincentive effects of welfare rates versus minimal wage levels, then the unemployment rate would obviously be relevant.

But not so much in the current case. Certainly not enough for you to "disprove" the disincentive argument.

Rory O'Farrell said...

@ anonymous

Just to be clear, you first object to my saying there is no link between marginal tax rates and an incentive to work, and then you give some reasons for why there might be no link?

If you are trying to make a point it isn't very clear what it is.

As can be read from the title the post is about marginal tax rates, not minimum wages and welfare.

Anonymous said...

As can be read from the title the post is about marginal tax rates, not minimum wages and welfare.

Exactly!

The point is that unemployment levels are barely relevant to the disincentive effects created by high marginal rates of tax.

(Whereas unemployment levels are very relevant when studying the disincentive effects created by the welfare code).

And why are unemployment levels not relevant?

Because high earners do not react to rising marginal tax rates by choosing to go on the dole.

Instead they modify their economic behaviour in a range of other ways that do not show up in the unemployment data.

Rory O'Farrell said...

" Instead they modify their economic behaviour in a range of other ways that do not show up in the unemployment data."

I know, that's why I put up the graph.

It doesn't show up in the GDP per capita data either.

Anonymous said...

"I know, that's why I put up the graph."

Honestly Rory, you might as well have put up graphs demonstrating no dependence between marginal tax rates and the sun-spot cycle.

If you want to claim that rising marginal tax rates don't hurt the economy and ultimately the exchequer, you're going to have to do a lot better than that.

Seamus Coffey said...

@ Rory,

You conclude:

In other countries, such as the Nordic countries, high tax rates are used to fund social services such as child care, which make it easier for people to work in the market economy.

The marginal income tax rates for earned income in the applicable Nordic countries are:

Sweden: 56.5%
Finland: 48.2%
Denmark: 48.1%

Ireland: 48.0%

And in Ireland the rate applies at a lower income threshold than the first three. It seems we are already in the class of high marginal income tax countries.

Rory O'Farrell said...

@Seamus

The data I used in the graph is the "Personal income tax & employee social security contributions (All-in rate)"

As I pointed out in the post we are 10th of the 34 OECD countries. Actually we are joint 10th with the UK. If we just look at the 'personal income tax' we are 6th, at 48%. I prefer to include PRSI.

It's true that in Ireland the top tax rate hits in at a low level. However, of the 9 countries that have a marginal rate above ours (when including PRSI), how many have a rate of at least 52% on average earners? I don't know. That's one problem with the OECD data set. E.g. if Ireland had a maximum rate of 52.1% affecting those on €175,000 then the data would show Ireland's top rate affecting those on over 5 times the average wage; while those on the average wage still have a marginal rate of 52%, but this would not be reported in the dataset.

@ Anonymous

"If you want to claim that rising marginal tax rates don't hurt the economy..."

I note that you have nothing to say on the GDP per capita graph. There no evidence from this graph of any negative relationship between GDP per capita and marginal tax rates. Of course absence of evidence does not necessarily mean evidence of absence. It's just a bivariate correlation. I'm not expecting to win the Nobel prize for it.

Regarding unemployment, just because an anonymous poster somewhere expects no relationship between two variables is no reason for not checking it out. Some logical arguments have been made for why marginal tax rates could affect unemployment. Some arguments have been given for why high marginal rates may affect job creation. For example see here: http://www.irishtimes.com/newspaper/opinion/2012/1120/1224326838841.html
Given the data on GDP per capita and unemployment, that argument is less convincing.