Mr. Conor O'Brien:

I thank the Chairman for inviting me to the joint committee today. A paper by KPMG completed in February 2014 was circulated and I may refer to the key points in it. The basis for some of the calculations in that paper is the Revenue Commissioners statistical paper, Corporation Tax Distribution Statistics, and I have copies of it for members.
There has been a great deal of discussion about the effective corporation tax rate in Ireland. The recent papers and articles published on the topic would suggest that there is a 2% effective tax rate in Ireland. That prompted us in KPMG to write the paper in February 2014. I will now go through some of its key points. Before members get into a discussion on what the effective tax rate is, there are a number of key concepts that are important in terms of the territorial scope of taxation.
The first question that one must ask is where a company is tax resident. That is a matter that has been determined in Ireland as a result of case law going back about 150 years. The leading case was the De Beers case in 1905. Our paper quotes part of the judgment which states:

In applying the conception of residence to a Company, we ought, I think, to proceed as nearly as we can upon the analogy of an individual. A company cannot eat or sleep, but it can keep house and do business. We ought, therefore, to see whether (sic) [recte where] it really keeps house and does business. An individual may be of foreign nationality, and yet reside in the United Kingdom. So may a Company. Otherwise, it might have its chief seat of management and its centre of trading in England, under the protection of English law, and yet escape the appropriate taxation by the simple expedient of being registered abroad and distributing its dividends abroad.

It then refers to a case some 30 years earlier, which I presume was in 1875, which held that a company resides where its real business is carried on: "Those decisions have been acted upon ever since. I regard that as the true rule; and the real business is carried on where the central management and control actually abides."
Ireland inherited that rule of residence in 1922 when it became an independent State and inherited the then UK tax system. That has been a central point of our residence rule ever since. The principle of what exactly central management of control means has been evolved by the courts in case law over the years. Irish law follows those cases. I do not think the Irish rules on that are regarded as unusual.
The OECD in its model tax treaties over the years has laid down as the key test for residence of a company the centre of effective management, which is essentially the same test. We can see that as long ago as 1905, or perhaps 1875, the UK courts rejected the concept of incorporation as being the concept by which one should determine residence of a company and they went instead for the concept of management and control. One can see in their judgments that the increased possibilities of tax avoidance that would arise if one went with an incorporation test as distinct from a management and control test were one of the reasons that they went with a management and control test. We have inherited a residence test which is sensible and is based on management and control. Much of the published articles and papers over the past six or seven months have indicated that Irish companies have an effective rate of corporation tax of 2% by looking at the taxes paid by Irish incorporated companies that are not resident here. I do not think that makes sense.
The territorial scope of Irish corporate taxation, which is common to many countries in the world, is that if a company is resident here it pays tax on its worldwide profits, and if it suffers tax in other countries, it may get a credit for the tax suffered in another country. If a company is not resident here, and it could be incorporated in Ireland and not resident here, it will still pay tax in Ireland if it has a branch here. One may come across the phrase "permanent establishment". That is the phrase used in the OECD model treaty and it is essentially the same concept as a branch. Branch is the term used in Irish domestic law, while permanent establishment, or PE, is used in the OECD model tax treaties. If one has an office, a factory or personnel in Ireland, one will pay tax in Ireland notwithstanding the fact that one is not an Irish resident company. Equally, non-Irish incorporated companies that are managed and controlled in Ireland would pay full tax here on their worldwide profits. For example, companies that are incorporated in the Cayman Islands but are managed and controlled in Ireland pay full corporation tax here on their worldwide profits. An analysis that looks at trying to calculate the amount of tax companies pay based on where they are incorporated is completely flawed, in my view. It would have as much sense as saying: "How much tax does an Irish citizen living in Dubai pay in Ireland?" The concept of residence is where a company resides and where it exists.
We could, of course, if we wanted to, do as the United Kingdom did in the late 1980s and amend our rules to provide a dual test. We could provide that a company is resident in Ireland if it is managed and controlled here or if it is incorporated in Ireland. We could make that policy choice, but heretofore we have not done that. That is not a particularly unusual feature of the Irish tax system.
One needs data to see what effective taxes are paid. One needs to know how much tax is paid by Irish resident companies on their worldwide profits - that is both incorporated and non-Irish incorporated companies. One wants data on the amount of tax paid by non-Irish resident companies that have Irish branches. The best source is my view is to go to the Revenue Commissioners because it collects the data. All the companies have to file tax returns. They file CT1s, corporation tax returns, and that data is aggregated in the Revenue document which I distributed. If members turn to page 6 of that document, they will see the Revenue Commissioners have aggregated the data and have come up with gross trade profits of €73.8 billion being the gross profits earned by Irish resident companies and Irish branches of foreign companies.

They deduct a number of expenses, which are not special tax breaks or reliefs but deductions for genuine business expenses. The document shows a figure of €8.5 billion for capital allowances. A capital allowance is essentially a deduction for the depreciation cost of an asset that one buys. As I note in my paper, the effective tax rate we produced in our calculations is understated. One of the reasons is that one does not obtain capital allowances on every asset one buys, which means companies buy certain assets for which they do not receive a tax reduction for depreciation.

Trade losses forward are a normal deductible expense. A company that makes a loss in one year can carry this forward to be offset against profits for the subsequent year. This is common practice in virtually every tax system in the world.

Current year trading losses occur where one makes a trading loss in one trade and offsets it with a second trade. This, too, is not an unusual deduction.

Trade charges refer to royalties, patent payments and so forth, which are deductible in every sensible corporation tax system in the world.

Group relief is where the tax system recognises where a group has profits in one company and losses in another. Again, virtually every corporation tax system in the world recognises that this type of relief should be available where one organic economic enterprise is split into different entities.

These are all normal deductions which reduce the figure for net profits to €38 billion. Tax should only be levied on net profits after legitimate expenses. One then has other items of income such as gross rental income and other normal deductions. The real net profit base in Ireland is €40 billion. This is the correct denominator in the calculation of an effective tax rate.

The figures on page 5 allow us to calculate the correct numerator. One starts off with a figure of €4.594 billion for the total tax less reliefs plus surcharges. This figure should then be adjusted for two items only, the first of which is research and development tax credits. These appear below the line with the €4.594 billion figure and two figures are shown. The first, €152.3 million, is an amount of research and development tax credits which were set against corporate tax liabilities. The second figure is a research and development tax credit of €106.2 million, which is the amount of research and development tax credits which are refunded to companies. This figure refers to research and development costs that are not offset against liabilities but refunded. These two items reduce the figure of €4.594 billion, whereas the other three items are not, in my view, adjusting items. I will discuss these relatively minor items in detail if members wish but the Department of Finance calculations treat them as adjusting items.

Another number should be added back. The table shows a figure for double taxation relief of €567.1 million. One could also do as the Department of Finance has done and produce two figures, one for before double tax relief and one for after double tax relief. Double tax relief is, however, a normal form of tax relief and should not be deducted.

This is how KPMG produced the numerator in our paper. Using this numerator and denominator, we arrived at an effective tax rate in Ireland of 12.24%. As I indicated, this is probably understated for a number of reasons. There are a number of items of genuine business expense for which tax deductions are not available in Ireland, for example, tax depreciation on certain assets and certain entertainment and other expenses. These figures must be added back. As the Revenue figures do not show these add-backs, the gross profit figure is somewhat overstated.

Ireland's loss relief rules are imperfect, which means companies receive loss relief in some but not all cases. On that basis, one could argue that the Revenue figure for total income is overstated. For example, the explanatory note accompanying the paper notes that, for example, the total net rental income in the State in the period in question amounted to a loss of €158 million, in other words, losses were made in aggregate as opposed to profits. Despite this, €520 million was charged in rental profits and, as such, the loss relief was imperfect. The paper also shows total trading losses forward of €150 million, yet this is not the number that emerges. Only a small portion of these losses are relievable.

The loss relief rules are imperfect and companies do not always receive relief for losses.