Archive for January 29, 2020

Taxing Figures

Posted in Bad Statistics, Politics with tags , , , on January 29, 2020 by telescoper

Following the campaign for the forthcoming General Election in Ireland has confirmed (not entirely unexpectedly) that politicians over here are not averse to peddling demonstrable untruths.

One particular example came up in recent televised debate during which Fine Gael leader Leo Varadkar talked about his party’s plans for tax cuts achieved by raising the salary at which workers start paying the higher rate of income tax. Here’s a summary of the proposal from the Irish Times:

Fine Gael wants to increase the threshold at which people hit the higher rate of income tax from €35,300 to €50,000, which it says will be worth €3,000 to the average earner if the policy is fully implemented.

Three thousand (per year) to the average earner! Sounds great!

But let’s look at the figures. There are two tax rates in Ireland. The first part of your income up to a certain amount is taxed at 20% – this is known as the Standard Rate. The remainder of your income is taxed at 40% which is known as the Higher Rate. The cut-off point for the standard rate depends on circumstances, but for a single person it is currently €35,300.

According to official statistics the average salary is €38,893 per year, as has been widely reported. Let’s call that €38,900 for round figures. Note that figure includes overtime and other earnings, not just basic wages.

It’s worth pointing out that in Ireland (as practically everywhere else) the distribution of earnings is very skewed. here is an example showing weekly earnings in Ireland a few years ago to demonstrate the point.

 

This means that there are more people earning less than the average salary (also known as the mean)  than above it. In Ireland over 60% of people earn less than the average.  Using the mean in examples like this* is rather misleading – the median would be less influenced by a few very high salaries –  but let’s continue with it for the sake of argument.

So how much will a person earning €38,900 actually benefit from raising the higher rate tax threshold to €50,000? For clarity I’ll consider this question in isolation from other proposed changes.

Currently such a person pays tax at 40% on the portion of their salary exceeding the threshold which is €38,900 – €35,300 = €3600. Forty per cent of that figure is €1440. If the higher rate threshold is raised above their earnings level this €3600 would instead be taxed at the Standard rate of 20%, which means that €720 would be paid instead of €1440. The net saving is therefore €720 per annum. This is a saving, but it’s nowhere near €3000. Fine Gael’s claim is therefore demonstrably false.

If you look at the way the tax bands work it is clear that a person earning over €50,000 would save an amount which is equivalent to 20% of the difference between €35,300 and €50,000 which is a sum close to €3000, but that only applies to people earning well over the average salary. For anyone earning less than €50,000 the saving is much less.

The untruth lies therefore in the misleading use of the term `average salary’.

Notice furthermore that anyone earning less than the higher rate tax threshold will not benefit in any way from the proposed change, so it favours the better off. That’s not unexpected for Fine Gael. A fairer change (in my view) would involve increasing the higher rate threshold and also the higher rate itself.

All this presupposes of course that you think cutting tax is a good idea at this time. Personally I don’t. Ireland is crying out for greater investment in public services and infrastructure so I think it’s inadvisable to make less money available for these purposes, which is what cutting tax would do.

 

*Another example is provided by the citation numbers for papers in the Open Journal of Astrophysics. The average number of citations for the 12 papers published in 2019 was around 34 but eleven of the twelve had fewer citations than this: the average is dragged up by one paper with >300 citations.