Income tax exemption limit

In my post on income tax of a few days ago, the sample structure that I presented went part of the way back to the rates and bands in effect in 1996, a good year.

One feature of the mid-1990s income tax structure was the much lower exemption levels effectively achieved nowadays through tax credits. Should this trend be reversed?

Two comments on my earlier post point to problems in lowering the effective threshold. Colm McCarthy worries about incentive effects given the interaction with social welfare. Aedin Doris finds it difficult to justify taxing a low income single mother with two kids.

I have sympathy with both views, and this is not a make or break issue for revenue (though there is some revenue potential even at the low end).

I also note counterpoints. The more revenue we seek from the system as a whole, the more a high exemption threshold/general tax credit pushes other workers into higher marginal tax rates; bad for incentives. And, depending on their family/household circumstances, not all low income part-timers have low consumption.

Then there is the political/ideological view that as wide a range of citizens should feel involved in the national housekeeping by paying some income tax (though all pay expenditure taxes anyway).

The impact of a lower exemption threshold on low income tax payers could be considerably eased, as has been suggested, by re-introduction of a third low income tax rate.

My guess is that, for a Minister of Finance, lowering the effective threshold (by lowering personal tax credits) offers too big a hostage to fortune to be worth the revenue it would raise.

What do others think?

(Update: the first version of this posting used misleading language about thresholds, I have modified it without changing the intended sense).

13 replies on “Income tax exemption limit”

I think the issue of comparisons of taxation of people with children leads to a number of complicated issues such as factoring the child benefit payments and costs of childcare. To keep things simpler, consider the figures from 2007 for total tax contributions (PAYE and PRSI) from the OECD’s Taxing Wages report for single earners at various percentages of average earnings:

67% of Ave. Earnings: 5.9% (Ireland), 24.7% (EU15), 22.2% (OECD)
100% of Ave. Earnings: 13.9% (Ireland), 29.9% (EU15), 26.6% (OECD)
167% of Ave. Earnings: 25.9% (Ireland), 36.5% (EU15), 32.2% (OECD)

While all of the Irish rates are low, the ones that really stand out are at the low and medium end and this reflects very generous exemptions and tax credits. And because a lot of people earn in the 67%-100% the government loses out on a serious amount of revenue.

Whether it is done through exemptions or raising the basic rate, it’s pretty clear that we can no longer afford to have such low tax rates on low and medium earners.

Another feature that should be factored in is the possible pro-cyclicality of incomes of the top 3% of Irish earners who pay 33% of income tax revenues. Suppose that the average income of this group declines by 20% since a substantial proportion of this income comes from boom-related profits and bonuses. (Of course, different individuals will occupy the cohort post-boom but a 20% decline for the changing cohort seems reasonable?). That could hit income tax revenues quite hard on its own since it gives a 6.6% decline holding all other income taxes constant (they may also decline somewhat). Do we know how muc pro-cyclicality to expect in top cohort incomes? It may turn out to be another “frothy” source of tax like stamp duty.

Does it make sense to provide universal lump-sum tax credits while taxing the money back for most taxpayers using necessarily higher marginal tax rates? Eliminating the basic tax credit for the almost 0.5 million taxpayers on the higher marginal would save close to 1 billion euro (this ignores the additional credit paid to married single-earner tax payers and also the PAYE credit).

Of course, the difficult part is how to phase out the tax credit with higher incomes. Phasing it out between 25 and 36.4K would add roughly 16 percentage points (1.8/11.4) to the marginal tax rate over this range — an ugly number to be sure in what must be a thick part of the income distrbution, but less than the marginal rate increase from extending the higher rate band down to 25K. Of course, additional savings are achieved in the phase-out range.

An even more comprehensive reform of the tax credit system would be to shift to a US-style earned income tax credit (also used by the UK, Canada, etc.) The credit initially rises with earned income, reaches a plateau, and then is phased out. This means that low income workers actually receive a marginal subsidy to earned income, which would help allieviate the disincentive concerns Colm has raised.

Sorry, one question of clarification. I am a bit confused by the discussion of exemptions. Although I’m sure I am just misinterpreting the rules, my reading from the Revenue’s website is that the 20K exemption (and associated rules for marginal relief) only applies to those 65 and over after 2007. I do see that there is an exception from the income tax levy for those earning below 18.3K, but I don’t think that is what Patrick is getting at.

Here is my information source:

The top end of salaries is going to be hit hard, especially as far as the Revenue is concerned. There is just no way there are going to be 1400 people declaring incomes over 1m for 2009 as there were in the past according to today’s paper (which at an average of 30 percent tax and average income of 2m will have yielded around a billion euros in tax, I think). For one thing they just won’t make the money. For another, they will restructure or defer the income to avoid the higher rate of tax.

John, You are right — there is no actual exemption limit these days except for the elderly. A senior moment had me going back to that old language and I have adjusted the text above to fix that. Also I have made some fixes on the earlier post.

Karl Whelan is too young to remember when the top rate of tax was called Surtax. To pay it was the Irish equivalent of a knighthood, and I think there were only about 3000 Surtax payers when it was abolished in 1973 and replaced by the top income tax rate.

Failure to index bands and allowances through the 1970s and 1980s extended the titled minor gentry wonderfully. The late Paul Tansey described the process as ‘Bringing Surtax to the masses’. Is this what Karl has in mind?

To think outside the box for a minute.

I understand the traditional tax-band set-up. But. Does the reason for such a simplified set-up still exist? Tax bands made for easier accounting and easier policing by the revenue in times of abacus and pencil.

Can we not now introduce a tax model where there is no tax free allowance, but low earners pay very little tax (maybe 1% on subsidence income), the rate then rolling up as income increases?

For example, let the first €10,000 be at 1%. Then for every extra €10 earned the rate would increase by 1 basis point. So the €20,000th would be taxed at 11%, the €30,000th at 21%, the €46,450th at 37.45% etc.

Obviously the rate of increase would have to slow down as the wage started to demand a 60%+ tax take on extra euros earned, but the idea is not complex, not would be implementing it. A simple formula would take care of it.

It would mean that all earners would be making a contribution, and dare I say it, could also be levied on social welfare payments.

In my previous comment, I wasn’t complaining about suggestions that we tax single mothers as such. My point was really that just because other countries tax low wage workers, doesn’t mean that we should do it.
The particular number I picked out – a single parent with two kids on two thirds of the average wage paying 20% tax (on average in the EU15) was intended to illustrate this – it’s pretty clear that the incentives here are terrible. If we agree on that, then we should be very wary about doing it. I’m not suggesting that under no circumstances should those currently not paying tax ever have to pay any, I’m just saying that if we do it, it should be because we think it’s the least bad thing to do, not because other countries do it. Benchmarking to other countries is not a great way of making policy.

On the incentives issue that Patrick raises, it seems to me that there are three groups of people whose incentives we need to worry about. First, the low skilled, whose wage rates give them low incentives to participate. Many, but not all, of those not currently paying tax are in this group. Taxing them will certainly reduce incentives.

The second group are the high fliers, the ones who take the risks and create the jobs etc. etc. As I’ve said before, I think that given limited mobility for as long as the recession is global, losing them to other countries is not much of a worry, as long as we make it clear that it’s a temporary situation.

The third group is married women. Their labour supply is the most responsive to wage rates, and they also face higher marginal tax rates because individualization isn’t complete. These are probably the ones most likely to respond to any increases in marginal rates. Or at least they would in normal circumstances. At the moment, though, there is probably a strong income effect from spousal wage cuts/job losses increasing this group’s participation probabilities. Perhaps this would be strong enough to counter any negative effects of increasing marginal tax rates.

For these reasons, my inclination would be to start at the top of the distribution, figure out how much that can bear, and then work backwards until you’re left with some amount of revenue that needs to be raised from those currently not paying tax. I get the impression that current government thinking is in the opposite direction – how much can we get out of the bottom, and what’s the residual that needs to be taken from the top?

@John, there is an in-work benefit in operation here, called the Family Income Supplement. It works on the same principle as the EITC, although there are lots of differences in the details, but it is paid by the Department of Social Welfare and so isn’t administered through the tax system. But it is a big part of the reason that the earnings of this single mother are increased by 35% in Ireland, rather than being reduced, as in other countries.

The political/ideological desire to have all people with an income paying tax and contributing to the general exchequer coffers has merit. When we apply for services we are entitled to, it is satisfying for us to be able to know and to say that we have paid our taxes when we had an income and what we seek is an entitlement, not a handout.
The method of pulling low income earners into the tax net would have to be gentle because of poverty traps, social welfare entitlements, medical card entiltements and generally the incentive to work. Any reduction in the tax credit would require a new lower rate of tax (say 15%) on the extra income effectively being taxed. In the interests of “fairness” a higher rate should be implemented on incomes in excess of €100k. I would anticipate a small rise in the standard rate and a rise in the higher rate also.
Social Welfare takes up €21bn of the presently projected €34bn total tax revenue. This is not a tenable situation and calls for taxation and/or reduction of some benefits particularly if negative inflation is expected during the year. The alternative is to tax the old reliables – it is easier politically. Both measures have an immediate effect on revenue with the latter affecting inflation and consequently competitiveness.
Is there a table available that sets out the potential savings to be made or revenue to be generated by the implementation of different spending cuts and extra taxation measures?

Aedin, thanks for the reminder on the Family Income Supplement; I’m very slowly getting up to speed.

I can’t disagree with your identification of key groups of taxpayers to worry about. But I’m not as sanguine as you about the efficiency/incentive effects of marginal rate increases on higher earners. While it is true that Ireland’s typically high international factor mobility — indeed almost regional economy-like bearing — is a key driver of supply elasticities, it is also true that more mundane supply responses are relevant as well.

Unfortunately, here I have to look to the international evidence given my ignorance of relevant empirical evidence for Ireland.

The state of the art in the empirical public finance literature is to look at the elasticity of taxable income with respect to the net of tax share. This elasticity allows for many margins of adjustment in addition to participation and hours (e.g. it includes effort and distortions towards non-taxable compensation).

Martin Feldstein’s “conservative” estimate of this elasticity is one-half. (See, e.g., Feldstein, Martin. 2005, “Rethinking Social Insurance, AER, 95(1)). To put this in more intuitive terms, at current US tax rates (not so different to Irish rates), raising an extra dollar has a total cost (including deadweigt loss) of $1.50.

You might reasonably object that Martin Feldstein’s estimates are at the higher end of the range. But what I think is the most authoritative study (Gruber and Saez, 2002, “The Elastticity of Taxable Income: Evidence and Implications,” Journal of Public Economics, 84) finds an average elasticity of 0.4. Morevover, they find substantially higher elasticities at higher incomes. It is hard to escape the conclusion that distortions at upper end of the income scale add significantly to the burden pf the tax take.

Colm references ‘Bringing Surtax to the masses’ and asks “Is this what Karl has in mind?”

A quick response: With the masses paying between 5.9% and 13.9% in combined PAYE and PRSI taxes, we’re pretty far away from the Surtax for the masses era. If Colm honestly believes that we can get through this period without raising income taxes on average earners, then he must have a lot more Snipping up his leave than we had been lead to believe!

Karl, we can increase the take from income tax without extending the tax take down into the lower reaches of the earned income distribution, as you well know. There are clear risks, as Brendan Walsh and others have noted, in raising marginal tax rates on low earned incomes at a time when private sector wages are falling and unemployment income is untaxed. Am I going too fast?

Comments are closed.