Thursday, July 30, 2020

The at-risk-of-poverty rate for children

The at-risk-of-poverty rate is an important indicator.  The chart below shows the share of children living in households who had an equivalised disposable income that was less than 60 per cent of the national median for the countries of the EU15 from 2004 to 2018.

EU15 SILC AROP Children 0-17 2004-2018

In 2018, the rate for Ireland was 15.8 per cent and puts it at its the lowest recorded level and was the fifth-lowest in the EU15.  Indeed if we look at the changes since 2004/05 we see that Ireland has seen the greatest reduction in the at-risk-of-poverty rate for children within this group of countries.

EU15 SILC AROP Children 0-17 Change 2004 to 2018

As the chart shows, the AROP rate for children aged under 18 has fallen by 7 percentage points in Ireland over the last 15 years (22.8 per cent to 15.8 per cent).  At the other end this compares to an increase of a similar magnitude for Sweden.

It could be that the first chart does not seem to fit with what we might expect.  Yes, there are reductions in periods of strong growth (2006-07 and 2017-18) but where is the crash of 2008 and the subsequent period of austerity?  For the ten years from 2008, Ireland’s at-risk-of-poverty rate for children was pretty much unchanged but this was a time of significant changes.

One of the reasons for this is that the at-risk-of-poverty rate is a relative measure: it is assessed against a moving target – the national median equivalised disposable income.  The threshold against which being at-risk-of-poverty is assessed will change based on how the national median income changes.

EU15 SILC AROP Threshold 2 2 Household 2005-2018

In 2008, for example, the at-risk-of-poverty threshold used by Eurostat for a two adult plus two children under 14 household in Ireland was around €29,000.  In the following years as the impact of the crash and austerity was felt virtually all incomes fell which meant that the median income, or midpoint of the income distribution, also fell. 

By 2011, the AROP threshold for a 2+2 household had fallen below €25,000.  This reduction in the threshold meant the AROP did not reflect the increased difficulties faced by many households.  The median income threshold began to rise in 2014 and by 2018 the threshold for a 2+2 household exceeded €31,000. 

This means deteriorations and improvements can be partly masked and not fully show in the at-risk-of-poverty rate and emphasizes that, while having poverty in it’s name, it is as much a measure of inequality.

We can get closer to a measure that reflects changes in absolute poverty by anchoring the threshold at a fixed point and examining the share of people which are below that fixed threshold (after adjusting for inflation).

EU15 SILC AROP Anchored Children 0-17 2004-2018

The pattern of the anchored at-risk-of-poverty rate for Ireland probably better reflects our prior expectations with a clear deterioration evident from 2009 to 2012.  Relative to its 2005 threshold adjusted for inflation no country has seen a larger reduction in its anchored at-risk-of-poverty rate

EU15 SILC AROP Children 0-17 Anchored Change 2004 to 2018

Greece went “off the scales” in the previous chart here we can see that its anchored AROP rate in 2018 was nearly 27 percentage points higher than it was in 2005.  Most countries show a reduction though for some it is relatively minor while no data on this measure is available for France.

An important consideration when assessing at-risk-of-poverty rates is the role of social transfers.  While Ireland in 2018 had an AROP rate for children that was significantly less than it was in 2005, if we look at the AROP rate before social transfers it has been at pretty the same rate for the past few years that it was from 2005 to 2008, i.e. in or around 40 per cent. 

EU15 SILC AROP before Transfers Children 0-17 2004-2018

Ireland’s AROP before social transfers for children has been close to the highest in what was the EU15.  Although it is at the level it was pre-2008 it has been on a fairly steady downward trend since 2011.  The changes in the previous chart are of a similar magnitude to the changes shown below for the share of children living in households where there is very little paid employment undertaken.

EU15 SILC VLWI Children 0-17 2004-2018

For every year for which data is available Ireland has had the highest share of children living in quasi-jobless households – households where the adults of working age are in paid employment for less than 20 per cent of the available time.

Eurostat also provide a measure of the at-risk-of-poverty rate after deducting housing costs from disposable income.  The threshold is left unchanged but the benchmark is assessed against the adjusted income of the household.

For Ireland this stood at 29.7 per cent in 2018 and has been showing a slight downward trend in recent years.

EU15 SILC AROP after Housing Costs Children 0-17 2004-2018

A related measure is the housing cost overburden rate.  This is the share of people living in households who face a total housing cost that is greater than 40 per cent of the household’s disposable income.  Here it is for children aged 0 to 17 years.

EU15 SILC Housing Cost Overburden Rate Children 0-17 2005-2018

Once again, Greece goes off the chart and in 2018 had a rate of 48 per cent which was almost three times greater than the second highest (the UK at 17 per cent).  The rate in Ireland in 2018 was 3.2 per cent which was the third-lowest in the EU15. 

It should also be noted that while mortgage interest payments are included by Eurostat in total housing costs for owner-occupiers mortgage principal repayments are excluded.  Of course, it may be that Ireland’s housing problems are preventing some from entering this category so it would be unwise to view Ireland’s position in a universally positive light.

And this may be evidenced when we look at the at-risk-of-poverty rate for young children:

EU15 SILC AROP Children 0-5 2004-2018

In 2018, Ireland had the lowest at-risk-of-poverty rate for children aged under six.  The sample size may be small so there may be a lack of precision in the estimates but it can be seen that the rate dropped from 17 per cent in 2016 to 9 per cent in 2018.

Now, we could interpret that as being the result of a very rapid rise in the incomes of households with young children.  From 2016 to 2018 the AROP threshold rose by 10 per cent so incomes would have to rise by that much just to match the rising threshold and it would take a much faster rise again to almost halve the AROP rate.

But maybe as well as the standard errors of the estimate could there also be a sample selection issue.  Maybe it is not that the incomes of households with young children are rising rapidly but that those households with young children have higher incomes.  Obviously, we can’t tell something like that from this aggregate data but with housing issues possibly delaying household formation and childcare issues resulting in high costs we should again caution against taking a universally positive view of Ireland’s position in the previous chart.

A hint that as is not as it might seem can be found by looking at the material and social deprivation rate.  This is the percentage of people in households experiencing at least five from a list of 13 items.  This is a relatively recent measure so a time series back to the early 2000s in not available.  Here it is for children aged less than six for the countries of the EU15 since 2014.

EU15 SILC Material and Social Deprivation Rate Children 0-5 2014-2019

So, while Ireland might have had lowest AROP for children under six in the EU15 in 2018, Ireland also had the fifth-highest rate of children in this age group living in housing experiencing material and social deprivation.

To conclude here is the persistent at-risk-of-poverty rate for children under under 18.  This is the share of children who are in an at-risk-of-poverty household this year who were also in this position for at least two of the previous three years.

EU15 SILC AROP Persistent Children 0-17 2007-2018

Again, some sample sizes affects precision and it can be seen that the estimates show a good deal of volatility.  In any given year, a maximum of 25 per cent of the sample from four years remain in the survey.  The survey design is such that a household remains in the survey for four years so one quarter of the sample is replaced each year.  Of course, the numbers exiting will be higher for a variety of reasons (household changes, non response etc.)

And for the above it is further limited to household who had children under 18 in the current year.  Still, recent estimates for Ireland are pretty steady and show that around 10 per cent of children are in households whose income is persistently below the relative at-risk-of-poverty threshold.

Thursday, July 23, 2020

The Modified Current Account in 2019

The CSO have published the details of CA*, Ireland’s current account of the balance of payments modified for some of the effects of MNCs.  The reported outturn for 2019 is pretty remarkable – a surplus of €16.5 billion (8.8% of its sister measure of national income, GNI*).  As the estimates of Fitzgerald and Kenny (2017) show the last time Ireland recorded a surplus of this magnitude was during World War II.

Current Account 1937-2019

If we combine the current account with the sector accounts we can assess the reasons for the increases in recent years.  Here it is since 2009.

Modified Current Account 2009-2019 by Sector

As was the case in 2018, all sectors have the economy in 2019 recorded positive outturns for S – I (gross savings minus gross capital formation).  Both the government and the household sectors increased their surpluses which is in line with expectations.  The negative “non-sectorised” item arises  in Ireland’s national accounts due to the differences between the income and expenditure methods of estimating GDP.  The discrepancy between the approaches is not attributed to any sector.

Of course, the most notable sector in recent years is the non-financial corporate.  In this instance it is the overall NFC sector with all the “star” adjustments applied to it.  Later in the year we will get a more formal split of the NFC sector into domestic, foreign and re-domiciled.

As we pointed out when these figures were released last year there is something going on with domestic non-financial corporates, and particularly the retained earnings of their foreign FDI and maybe some link to their exploding balance sheet.  When the 2019 figures are released we will see if this continued to have an effect in 2019.  The figures above suggest it did.

All-in-all though the figures for the modified current account are a positive indicator for the Irish economy.  It suggests we do have capacity to respond to the crisis.  And even absent a pandemic there might be reason to look for increased spending (more house building perhaps) to reduce the CA* surplus.

Monday, July 13, 2020

The 2018 Aggregate Corporation Tax Calculation

The Revenue Commissioners have published the 2018 uptake of their aggregate summary of corporation tax returns.  The last of these would have been filed last September and the aggregate data is now available.  As previously, we will look at the outcomes in two stages:

  1. The determination of Taxable Income from Gross Trading Profits, and
  2. The determination of Tax Due from Gross Tax Due

So, lets look at the determination of Taxable Income for the five years from 2014 to 2018.

Aggregate CT Calculation for Taxable Income 2014-2018

In 2015, Irish GDP jumped by around 30 per cent in nominal terms and there’s a lot of action in the first annual change shown in the table.  The unusual GDP increase was primarily the result of the near €50 billion increase in Gross Trading Profits that happened that year.

If we jump down to the bottom of the table we see that that €50 billion increase in Gross Trading Profits only translated into increase in Taxable Income of less than €15 billion.  The main reason for this was the jump in the use of Capital Allowances.  In 2014, €18.6 billion of capital allowances was used against gross trading profit; in 2015 this was €46.2 billion. 

As is well understood this increase was due to increased claims for capital allowances for intangible assets.  Irish resident firms have been making significant capital outlays to buy intangible assets and this expenditure is offset against trading profit as a capital allowance in the determination of Taxable Income.

After 2015, Taxable Income continued to rise recording increases of €6 billion, €8 billion and €16 billion over the next three years.  The 2018 rise was the largest on record.

For the past couple of years we note a couple of things from further up the table:

  • The ongoing rise in the use of capital allowances;
  • The continued level of losses carried forward being used;
  • The increase in foreign income included in Irish CT returns;
  • The decrease in deductions for trade charges; and
  • The level of relief claimed for interest under Section 247.

We may take a closer look at some of these in due course.  For now, we note that after all the adding and subtracting we end up with a figure of €96 billion for Taxable Income in 2018. 

Of this, €87 billion arises from income that is subject to tax at 12.5 per cent (including capital gains which are regrossed to reflect the higher rate of CGT) and €9 billion is income taxed at 25 per cent (mainly non-trading income). 

Although not explicit from the figures it seems likely that the increase in income taxed at 25 per cent relates for foreign income of Irish resident companies.  Net foreign dividend income fell from €8.7 billion in 2017 to €4.5 billion in 2018, while total foreign income rose from €11.4 billion in 2017 to €12.2 billion in 2018.  In most cases, as a result of tax paid abroad, little additional tax is due in Ireland on this foreign-sourced income. 

We can get some insight into this if we look at the second part of the aggregate CT calculation: the transition from Gross Tax Due to Tax Due:

Aggregate CT Calculation for Tax Due 2014-2018

Multiplying the taxable figures by the applicable tax rate gives the starting point of Gross Tax Due (€87 billion x 12.5% plus €9 billion x 25%). 

Ireland had a limited number of reliefs from Corporation Tax.  By far the most significant is the relief for foreign tax paid on foreign income.  Ireland has a worldwide regime for the taxation of resident companies.  They must include all their income in the Irish CT return wherever earned. 

To avoid double taxation, relief is granted for foreign tax paid.  If the amount of foreign tax paid is less than the amount that would be due in Ireland (at the 12.5% or 25% rate) the company makes an Irish tax payment to bring the total tax paid up to the required amount.  In practice very little additional tax is due in Ireland.

We can see that Double Taxation Relief and the Additional Foreign Tax Credit were almost €2.2 billion in 2018.  At 12.5 per cent this amount of tax relief would correspond to an income of €18.4 billion.  As we say above, foreign income in 2018 was €12.2 billion so it must be that a large share of the foreign income is income that would be subject to tax at 25 per cent, and has double taxation relief applied accordingly.

After tax reliefs, we see that tax payable in 2018 was €10.9 billion.  As per reliefs, Ireland has a relatively small number of tax credits that can be used to reduce CT payments.  The most significant of these is the R&D tax credit though the cost of claims has fallen in recent years.

In 2016, the R&D credit and the repayment of excess R&D credits when the credit exceeded the starting liability summed summed to €670 million.  For 2018 this had reduced to €355 million.

The item Gross Withholding Tax on Fees is simply the granting of credit for tax that has already been paid to the Revenue Commissioners.  The tax code includes a number of instances where the payment of an invoice requires some of the fee to be withheld and paid to the Revenue Commissioners.  Typically this is 20 per cent of the amount due (reflecting the standard rate of Income Tax). 

The person/company paying the invoice withholds the required amount from the provider of the services and transfers the money to the Revenue Commissioners along with the intended recipient’s tax details.  The person/company from home the fee is withheld can then claim credit for the tax paid on their behalf.  The line for Gross Withholding Tax on Fees reflects this tax that has already been paid.

For all CT returns filed for periods ending in 2018, companies had a tax due amount of €10.2 billion.  As indicated in the recent Revenue paper this equates to an effective tax rate of 10.6 per cent ((€10,211 million / €96,049 million) x 100).

However, as we have seen, while there was a €3 billion difference between Gross Tax Due and Tax Due in 2018, €2.5 billion of this was because of tax already paid, either tax paid in other jurisdictions or amounts withheld from fees. 

If there had been no foreign tax paid or fees withheld for tax then the Tax Due on the €96 billion of Taxable Income reported in 2018 CT returns would have been €12.7 billion, an effective rate of 13.2 per cent.

As it was companies did pay tax to other jurisdictions on their foreign source income and did have fees withheld and transferred to the Revenue Commissioners on their behalf.  This means that the amount of Tax Due was reduced but the tax burden imposed by the Irish regime should take into account amounts already paid.

Anyway, whatever way it’s dressed up, Corporation Tax receipts was increased significantly in recent years.  In subsequent posts we will come back to some of the “big ticket” items in the top first part of the CT calculation.

Wednesday, July 8, 2020

Ireland in the OECD’s aggregate Country-By-Country Reporting (CbCR) data

The OECD have released anonymized aggregate data by jurisdiction of ultimate parent and partner country for MNE groups.  The data come with a significant disclaimer and only cover a single year (2016).  Still, they show what can be shown with the data.

Here will be focus on the figures for Ireland.  A reliability check was carried out by the Revenue on the figures.  These are the aggregate figures (note: financial values are in millions of US dollars). Apart from the ‘total’ row all the figures are as reported on the OECD database.  The ‘total’ row is calculated from the constituent figures.Click to enlarge.

OECD CbCR data for Ireland 2016

The Irish figures are unusual (see the chart on page 40 of the OECD report) but are not unexpected.  The dominance of US firms in the Irish data is clearly evident.

Of the totals, three-quarters of revenue and two-thirds of profit arise in US companies included in the data.  Of the $5.5 billion of cash payments for corporate income taxes in the table $4.3 billion (78 percent) came from US companies.

The role of Ireland in the tax strategies of MNEs is a topic of frequent discussion.  What this, and other, data highlight is that this is almost exclusively linked to US companies.  This points to the issue being something that is specific to US tax laws rather than general to Irish tax laws.

We can use the above data to get some effective tax rates.

OECD CbCR data for Ireland 2016 ETRs

In overall terms, the companies in the data had an effective rate rate of 12 per cent on a cash basis (and 11.4 per cent on an accrued basis).  The profit figures for some of the countries are small (or negative) and the ETR for any single year may not be that informative.

Focusing on the largest countries in the data we can see that the ETR for US companies on a cash basis was 13.6 per cent.  The other significant parent jurisdiction in the data are domestic MNEs headquartered in Ireland.  The Irish MNE groups in this data had an ETR on a cash basis of 6.3 per cent.

As noted above this data only covers 2016 so is already somewhat dated.  The OECD plan on publishing data for subsequent years which will only add to usefulness of this data by allowing changes to be identified.