A Progressive Tax System? The Poor Pay as Much Tax as High Income Groups?

, , 1 Comment

0 Flares Twitter 0 Facebook 0 0 Flares ×
Print pagePDF pageEmail page

Remember all those observations?  About how the highest income groups pay almost all the tax and how terrible it is that begrudging lefties want to tax them more?   About how Ireland has the most progressive tax system in this quadrant of the Milky Way?  The Government has lead the chorus making this claim but in truth it is not based on comparative measurement of tax progressivity (see Note at the end of this post for a discussion of the Government’s claim).

So along comes a study that blows those arguments away.  Dr. Micheal Collins and Dara Turnbull investigated the issue in a working paper published by the Nevin Economic Research Institute, based on the CSO’s Household Budget Survey 2009/10.  They found that, contrary to the received wisdom, the poorest 10 percent income group pays as much tax as the top 10 percent tax and that our tax system is far less progressive than some have claimed.

Here’s the bottom line chart.


Oh, my.  The poorest 10 percent income group pays a tax rate of 28 percent – that is, their tax payments make up 28 percent of total income (which includes income from work and social transfers).  The top 10 percent pays a tax rate of 29 percent.  Doesn’t look that progressive to me.

How could this be?  Micheal and Dara estimated the impact of all taxation – income tax, USC, PRSI, and (and this is the key innovation of this study) indirect tax such as VAT and Excise, and levies such as TV licenses and vehicle taxes.  Previously, claims about the tax contribution of high income groups narrowly focused on income tax and, sometimes, PRSI.  But these make up only part of the tax system.  Over 40 percent of tax revenue comes from indirect taxation.  The following shows the extent to which indirect taxation undermines the progressivity of the tax system.


Unsurprisingly, the lowest income groups pay substantially more of their income on VAT, excise and levies than higher income groups.  So when this is combined with direct taxation – income tax, USC and PRSI – we get only an overall marginally progressive effect.

Another perspective is to assess how much tax each decile pays in relation to the proportion of total income received.


We find that the top 10 percent income group receives 29 percent of all income in the state and pays 35 percent of all tax.  The lowest income group receives only 1.8 percent of all income but pays 2.1 percent of all tax.  In all other deciles, the amount of total income received and total tax paid are close together.  That certainly isn’t very progressive.

Nationally, the average tax rate is 24 percent.  For the top 10 percent it is 29 percent even though the top income group receives three times the income as the national average.  And that, too, doesn’t sound very progressive.

This is a significant study and is worth reading in full.  The authors state that this is only the first step in deconstructing the progressivity or otherwise of the Irish tax system.  Further studies will analyse data based household size, composition (number of children, etc.) and direct / indirect tax per adult rather than household. They also intend to incorporate adjustments in subsequent budgets.  When this project is finished it will be a considerable addition to our understanding of the distributional impact of our tax system.

However, I would urge one note of caution.  Taxation is only one side of the redistributive coin. The other side is the range of benefits and public services that taxation finances.  The gross income in this study does include social transfers (Child Benefit, pensions, unemployment payments and other payments) as per the CSO survey.

However, it doesn’t include the income benefit of public services.  Free GP care and heavily subsidised prescription medicine would have a more progressive impact on average income groups than high income earners.   Ditto for subsidised childcare and truly free education (without the charges for school transports, schoolbooks, uniforms, and ‘voluntary fees’).

For instance – what would you think of a tax system that has the following:

  • 25 percent VAT rate, including VAT on food
  • Above average excise taxes
  • A national top rate of tax at 25 percent (on top of a standard tax rate of 20 percent)
  • A flat-rate tax of 31 percent at local level

This doesn’t sound very progressive at all.  High VAT on food, a two-tier tax rate of 20 and 25 percent, and a flat-rate tax which never progressive.  Indeed, this tax system looks to be extremely regressive.

Who applies this tax system?  Sweden.  And Sweden has one of the lowest levels poverty and deprivation and one of the more egalitarian income structures, supported with strong public services (Ireland would have to spend €13 billion more on public services to match Swedish levels).

It is difficult and somewhat of an abstraction to put a monetary value on indirect benefits such as public services in order to assess the redistribution of tax revenue.  This means that much redistributive analysis, based on income deciles, necessarily focuses on cash payments whether tax or social transfers.  This doesn’t tell the full story, though.

This is not a criticism of Micheal’s and Dara’s study which is just intended to  analysis the distributional impact of taxation – and which blows away some myths in the process.  We must work step-by-step to get a firm grasp on aspects of these questions and the two authors have made an invaluable contribution to that end.

Is Ireland’s tax system progressive?  Not very.  Now let the debate start – this time from actual evidence.

* * *

NOTE:  The government’s claim that we have one of the most progressive tax systems in the industrialised world is based on OECD data from Taxing Wages.  It’s not that this data is flawed, it’s just that, in the first instance, the data doesn’t attempt to measure overall tax progressivity.

  1. The data measures the ‘tax wedge’.  This measures difference between labour costs to the employer (wages, PRSI) and the corresponding net take-home pay of the employee including some child income cash benefit (Child Benefit).  The average tax wedge measures identify that part of total labour costs which is taken in tax and social security contributions net of cash benefits.
  2. This data focuses on statutory tax rates and basic allowances.  It does not include tax reliefs, allowances and exemptions which would be difficult to do across all OECD countries.  Nor does it include other cash income that households may receive (social transfers, capital income), nor does it include self-employment income.
  3. It includes employers’ PRSI which, in the first instance, is not a tax paid by workers.
  4. It only assesses three income levels.  In Ireland’s case, it assess the tax wedge for incomes of €21,894 €32,841 and €54,736.  This is can hardly be comprehensive.
  5. It doesn’t include the monetary benefit that workers receive for, in particular, social insurance payments.  For instance, a worker in country A may pay a higher level of social insurance (PRSI) than a worker in country B.  Therefore, country A, according to the OECD table, is deemed more ‘regressive’.  However, the worker in country A receives free GP care, subsidised prescription medicine, high maternity/paternity payments, pay-related social insurance pensions, pay-related sick pay, etc.  Worker in country B doesn’t – s/he has to pay for these in the private market or out of personal savings; and these payments are higher.  Who is better off?   The OECD data doesn’t and can’t tell us.
  6. The OECD data can produce perverse results.  For instance, let’s say the Government decides to double employers PRSI to pay for those increased benefits mentioned above – free GP care, prescription medicine, pay-related pensions, maternity/paternity benefit, etc.  This would be a major boost to the living standards of low/average income earners and would not entail one extra cent in tax.  However, under the OECD measurement, the ‘tax wedge’ would become more ‘regressive’.  This is perverse.

It’s not that there is anything wrong in the OECD’s methodology.  The ‘tax wedge’ is a useful measurement but like all such measurements, it tells us what it tells us.  The key is to understand that message in the context of other measurements.  The problems begins when someone claims that it represents more than what it is intended to tell us.  That’s the problem with the Government’s claim that Ireland has one of the most progressive tax systems in the OECD.


The following two tabs change content below.

Latest posts by Michael Taft (see all)


One Response

  1. Robert Sweeney

    December 6, 2013 3:19 am

    How is it then that we go from having the most unequal ‘market’ distribution of income to around the mean/median post-tax Gini?