Saturday, November 22, 2014

In Charts: 7.7% Growth & Contract Manufacturing

Here are some slides from a presentation on the recent performance on the Irish economy: statistical recovery or real recovery?

When the Q2 QNAs were published back in September they were meet with wide acclaim.  The headline figures looked very impressive.

Quarterly GDP

The annual growth rates (percentage change over the same quarter in the previous years) are impressive. GDP in the second quarter of 2014 was 7.7 per cent greater than the equivalent quarter of 2013.

Annual GDP Growth

But as was quickly pointed out that while the headline performance might be good “under the hood” things were not all what they seemed to be. 

First we can try to decompose the growth into “net exports” and “domestic demand”.

Contribution to Annual Growth

As an aside it is worth noting two things about the above chart though neither is likely to have a substantial impact.

  1. It doesn’t show the impact for changes in stocks.  Net exports or domestic demand could rise but if some of this comes from existing stocks the effect on overall GDP will be lower.
  2. The CSO would not produce a chart such as the above because the individual series from which the above contributions to overall GDP growth are derived are not additive.  The series are chain-linked individually into constant prices.

These aside we can see that the growth in Q2 2014, in annual terms, was pretty evenly divided between net exports and domestic demand. 

If we now break net exports into its constituent parts we see that exports shot up hugely in the first have of 2014.

Trade Components

What are we selling more of?  On the services side little is evident.  Both have increased but not unusually so.

Trade in Services

However, if we look at trade in good we see this.

Trade in Goods

Goods exports in the national accounts have gone vertical!  And there has been no offsetting increase in good imports.  So what are we making lots more of that we are exporting? 

The CSO’s Trade Statistics show merchandise exports from Ireland and they should reflect the massive increase in exports in the QNA and show us what category the additional exports are coming from.  But as we now know they don’t.  Here is a comparison of quarterly goods exports as reported in the Trade Statistics and as recorded in the National Accounts.  For comparison purposes both are in current prices.

Goods QNA versus TS

The increase in goods exports in the QNA does not appear in the trade statistics.  It is due to “contract manufacturing”.  Manufacturing that is organised by an Irish resident company but happens somewhere else.  However, because the risk is with the Irish-resident company part of the value appears in the Irish national accounts.  Yet another MNC effect to muddy the waters.

The conclusion was that there is growth but that perceptible rather than statistical growth is limited to the blue part of the final bar in bar chart above.  We will turn to that next.

Friday, November 21, 2014

Irish Examiner 21/11/2014

Here is a piece carried by today’s Irish Examiner on the funding of public expenditure on water.  It is based on a recent post - ‘how much do we spend on water?’.  The unedited text for the piece is reproduced here.

Conservation grant a half-measure that muddies the waters

In 2012 public expenditure on water in Ireland was €1.5 billion.  This money is being spent on a system that does not work.  The ongoing protests about water charges fails to understand the key point – it is not about paying for water; it is about paying for the infrastructure to deliver clean water and treat waste water.

Over the past ten years the government has spent €18 billion on water.  Of this, only 10 per cent was to pay for the actual supply of water to households and businesses.  An emphasis on conservation can only save money on this small portion of the expenditure.  The other 90 per cent of expenditure is unaffected by the amount of water we use.

It costs about twice as much to manage and treat waste water and sewage as it does to provide clean water.  A further 30 per cent of water expenditure over the past decade was consumed by staff costs.  The remaining 40 per cent was devoted to investment in the water and waste water infrastructure.  This is the largest component of expenditure on water and the most important.

The most significant austerity measure introduced in relation to water since 2008 is not the botched introduction of domestic water charges – it has been the slashing of investment in our water infrastructure.  Since 2008 capital spending on water has been cut by over 60 per cent.

In 2008, the government provided €1 billion for investment in the public water system.  The same year the ESB also provided €1 billion for investment in the electricity system.  In 2012, the ESB’s capital expenditure was close to €800 million while public capital expenditure on water was cut to just €375 million.

As a semi-state company the ESB can leverage its revenues to borrow money to fund its investment.  As a commercial enterprise the ESB’s management can take a long-run perspective.

On the other hand, expenditure on water was part of general government expenditure and the huge cuts to public investment in water were introduced to meet short-term deficit targets and avoid adjustments elsewhere.  

Huge expenditure is needed to bring our water network up to scratch but it is not clear where the money will come from.  The government remains under pressure to reduce the deficit and the size of the public debt limits the government’s capacity to borrow even though interest rates are at historical lows. 

A public utility can raise the funds and its management can look beyond short-termism or the election cycle.  Investment in water and sewage infrastructure does not have political attractions.  It is disruptive, underground and offers little opportunity for ribbon cutting. 

Although they have been on the agenda for years, water charges were rushed in by the current government because they wanted to get them in the rear-view mirror well before the 2016 election.  The charges were over complicated politically-motivated add-ons such as allowances which required PPS numbers to be administrated correctly. 

If the politicians wanted to compensate people for the introduction of water charges it should have been done through the existing tax and transfer system.  This would have been possible in last month’s budget which had a €1 billion package of tax cuts and expenditure increases but a lack of joined-up thinking resulted in a failure to get a coherent message across.

Now we have a further complication with the €100 water conservation grant.  This is just further pointless administration.  The government want to cap water charges at €160 per household but because Irish Water must get more than half of its revenue from private sources households will actually pay €260 to Irish Water and receive €100 back from the Department of Social Protection. 

The government met major difficulties a few years following the introduction of the €100 household charge; now they are giving €100 to households to try and solve the mess made of water charges.   To avoid been deemed as a subsidy to Irish Water the €100 will be paid to all households.  So instead of collecting money off those on the public system we now have reached a point where we are subsidising those on private water schemes who paid for their own water up to now.

Ireland has a history of introducing measures with names that do not reflect their actual design.  For example, we have the respite care grant and the public sector pension levy.  This measures have their merits but their implementation has nothing to do with respite care or public sector pensions.  They are a transfer payment and a pay cut. 

There are many items on this list to which we can now add water charges and the water conservation grant.  These have nothing got to do with using or conserving water.

The principles of taxation are equity, efficiency, certainty and ease of administration.  The charges announced this week fail on three of these principles.  Both the charge and offsetting grant are at a fixed rate so are regressive.  The fixed charge means that there are no efficiency gains from conservation and the fact that households have to pay €260 to Irish Water to get €100 back from the Department of Social Protection only adds to the unnecessary bureaucracy of our system.

It is a little ironic that the outcome of the protests of the past few weeks has resulted in charges that are regressive and inefficient.  We should have water charges based on use but not necessarily because we need to conserve water.  Conserving water is helpful but it is not as if like we are going to run out of it.  It actually does fall out of the sky.  It is better management of the plentiful supply we have that we need.

Better management can be achieved by metering but the benefit of this will be in identifying leaks and unreasonable usage rather than in getting typical households to control their usage.  Universal metering is unlikely to being significant benefits.  The introduction of metering is to link charges with usage of water but we must have a utility that can adequately fund the 90 per cent of expenditure that is on top of the cost of the actual water.

Many of those protesting want the abolition of Irish Water.  A utility company for water is exactly what we need.  The system that resulted in the broken water network we have now must be dismantled.

Tuesday, November 18, 2014

Is Ireland running a primary surplus?

A bit esoteric but probably not.

The Economic and Fiscal Outlook released with last month’s budget shows that a small primary surplus is expected for 2014.  The primary balance is the balance when interest expenditure is omitted.  This is shown in the following table in item 18.

Primary Balance 2014

The overall deficit is expected to be just below €7 billion and because interest expenditure is almost €7.5 billion, a small primary surplus results.  It is supposed to be a measure of the fiscal position once debt servicing costs are excluded, i.e. is government revenue sufficient to cover all non-debt related expenditure.  The above table suggests this is the case.

However, there has been lots of circularity in the public finances over the past few years with money moving between entities (the Exchequer, the NPRF, Nama, the IBRC and the Central Bank) and hence in and out of the definition of the “general government sector” under Eurostat rules.  The NPRF and the IBRC are within the general government sector while NAMA and the Central Bank are not.

As a result of the Anglo bailout and the Promissory Note bond swap the Central Bank holds around €28 billion of Irish government bonds (it was originally €28.5 billion but €0.5 billion have been sold.)  These bonds will see around €900 million of interest paid to the Central Bank during 2014 with around €750 million coming the Floating Rate Notes and €150 million from the 5.4 per coupon 2025 bond.  A large part of this is repaid back to the Exchequer Account and is counted as Revenue in the above table.

Because of this cross-holding in the broader government sector the primary balance can appear different with no actual change at all.  For example, if the interest on the bonds held by the Central Bank was to increase to €2 billion.  This would increase interest expenditure by €1 billion and also see revenue increase by €1 billion when the Central Bank surplus is paid to the Exchequer.  The overall balance would be unchanged.

However, the primary balance would show a €1 billion improvement as the one €1 billion increase in interest would be excluded when calculating the primary balance but the €1 billion increase in revenue from the Central Bank surplus would be included.  This is not how the primary balance should change as it is supposed to be unaffected by interest changes.

So with a reported primary surplus of €575 million and around €900 million of interest being paid to the Central Bank and flowing back to the Exchequer it can be seen that the primary surplus is somewhat notional.  This would be revealed if the Central Bank was to sell the bonds it holds meaning that the interest payments would not be recycled back to the Exchequer.  Alternatively, the primary surplus would disappear if the interest paid on the bonds held by the Central Bank was nil.

Excluding these circular interest payments between the Exchequer and the Central Bank, government revenue is still not sufficient to cover non-debt related expenditure. It should happen in 2015.

Monday, November 17, 2014

How much do we spend on water?

Here is data from Eurostat on government expenditure in Ireland on water (divided between “water supply” and “waste water management”)

Expenditure on Water 2012

This includes expenditure for commercial as well as domestic supply of water.  Around €300 million is collected annually in commercial water rates.  Two things are worth noting:

  • Two-thirds of the expenditure relates to the management of waste water and one-third to water supply.
  • One-sixth of the total relates to the intermediate consumption involved in actually supplying water (€250 million).

Here is a table with the same data for the decade from 2003 to 2012. Click to enlarge.

Expenditure on Water 2003-2012

Over the decade the government spent €18 billion on water services.  The intermediate consumption of actually supplying water is 11 per cent of the total (€1,981 million out of €18,057 million).

The element showing the greatest variation over the years is gross capital formation (investment in infrastructure).  This was around €600 million in 2003, rose to €1,100 million in 2007 and was reduced to less than €400 million in 2012.

In financial terms, the most significant austerity measure in relation to water is not the introduction of domestic water charges; it has been the slashing of capital expenditure on water infrastructure.  This has short-term benefits for the deficit but the money will have to be spent at some stage.

Here is a comparison of public capital expenditure on water (funded directly by the government) and on public capital expenditure on electricity (funded indirectly by the ESB).  No value for money judgements are implied.Water versus Electricity Capex

The table does not include private capital expenditure and there have extensive subsidies and incentives for private capital expenditure in electricity (wind farms etc.) in recent years.   It also excludes private capital expenditure on water by the many not connected to the public water system.

In 2008, public capital expenditure in the two areas was almost identical.  By 2012, public capital expenditure on water had been cut by 65 per cent and was almost 50 per cent below its 10-year average of €716 million.  Capital expenditure by the ESB was also reduced but was  20 per cent below its 10-year average of €966 million. 

The cuts to capital expenditure on directly funded water services were far greater than those to capital expenditure on indirectly funded electricity services.  This was done mainly to achieve deficit reduction targets.

The water charges debate has focussed on the cost of the water that comes out of the taps.  For the past ten years that has been about 10 per cent of the cost of public water services.  We need to make sure we can fund the other 90 per cent.

Wednesday, November 12, 2014

Preferential IP regimes, R&D expenditure and Ireland’s Knowledge Development Box

The most significant announcement on Corporation Tax in the recent budget was the planned introduction of a ‘knowledge development box’. The changes to the residency rules under the heading of the “double-irish” will be ineffective in changing tax outcomes and the profits, as well as the low-tax outcomes, will remain offshore. A blind-eye could not be turned if similar low-tax outcomes were achieved using an onshore ‘knowledge development box’.

In the Financial Statement the Minister for Finance said “I plan to legislate for it in next year’s Finance Bill or as soon as EU and OECD discussions conclude.” A significant element of these discussion is this joint Germany-UK statement on patent boxes.

It is not clear what impact this will have on Ireland. The Minister for Finance’s intention that “the Knowledge Development Box will be best in class and at a low competitive and sustainable tax rate” is a relative concept. When trying to examine this it worthwhile to look at the level of R&D expenditure in Ireland.

The most significant change in ESA2010 is the reclassification of R&D expenditure. In essence it has been moved from intermediate consumption to capital investment. The revised NIE accounts provide figures on GFCF (Gross Fixed Capital Formation) in R&D back to 1995.


The six per cent rise in the level nominal GDP arising from the introduction of ESA2010 is almost entirely due the capitalisation of R&D expenditure as the CSO explain here. The estimate of 2013 nominal GDP was revised from €164 billion under ESA95 to €175 billion when ESA2010 was applied.

While there is a huge amount of R&D expenditure in Ireland (€7 billion in 2013) but there is little evidence of activity resulting from it. The reason for this is that most R&D in Ireland is imported. The expenditure comes from Ireland but the activity takes place somewhere else.

We can see this in the Balance of Payments where R&D imports are recorded (payments made from Ireland to pay for R&D activity elsewhere).  International standards for the collection of Balance of Payments data have also being revised and the CSO have moved from applying BPM5 (Balance of Payments Manual) to BPM6.  For that reason we get two measures in the following chart as the CSO have not published revised historical data using BPM6.

R&D Imports BoP

The main difference for R&D in BPM6 is that outright purchases of patents are now included in the current account whereas under BPM5 such transactions were in the capital account.  The €2 billion difference between the BPM5 and BPM6 R&D imports figures for 2012 is due to a large acquisition of a patent by an Irish-resident company. 

There is also a small amount of R&D exports from Ireland and this series seems relatively unaffected by the change from BPM5 to BPM6.  It is the net amount between exports (R&D assets going out) and imports (R&D assets coming in) that contributes to gross fixed capital formation (GFCF).  The following table gives a breakdown of the R&D component of GFCF shown in the first chart for the past four years (all figures in €millions).

R&D Breakdown

Net capital transfers are the irregular outright purchase and sale of patents and other IP.  Net service imports mainly reflects MNCs paying for R&D with funds from their Irish branch/subsidiary.  This allows the Irish MNC to contribute to the cost of the R&D and also to use the resulting patent or IP.  This can be achieved through things like cost-sharing agreements.  Here is a short description:

An agreement between two parties to share the cost of developing an intangible asset, such as computer code, production methods, or patents. Such an arrangement is used to reduce or avoid taxes on the transfer of assets. For example, if a parent company wanted a foreign subsidiary to use one of its patents, tax authorities might consider the transfer a taxable transaction. By establishing a cost sharing agreement, the parent company and the subsidiary share in the cost of developing the patent, so that both are entitled to use it, and it is not transferred from one to the other.

The final sub-part of the above table labelled “in-house R&D” refers to R&D activities in the private sector that take place domestically (i.e. within Ireland) and also in the public sector such as in health and education.

The table clearly shows that the most significant contributor to R&D in Ireland is R&D imported by the MNCs.  It is important to note that this only has a once-off level effect on GDP.  Up to this year these R&D imports where included in the national accounts (subtracting from GDP).  It is only with the introduction of ESA2010 that these R&D imports are counted as part of gross fixed capital formation (adding to GDP).  From now on the net or growth effect will be zero as the positive investment effect will be offset by the negative import effect.  It will affect the composition of GDP though.

How do these R&D expenditures relate to the Germany-UK statement on preferential IP regimes?  Here is a key extract (emphasis added):

The OECD Forum on Harmful Tax Practices (FHTP) has led work in relation to BEPS Action 5, Countering Harmful Tax Practices More Effectively, Taking into Account Transparency and Substance. Work within the FHTP has led to the development of proposals for new rules, known as the Modified Nexus approach, based on the location of the R&D expenditure incurred in developing the patent or product. This approach seeks to ensure that preferential regimes for intellectual property require substantial economic activities to be undertaken in the jurisdiction in which a preferential regime exists, by requiring tax benefits to be connected directly to R&D expenditures.

One source of R&D expenditure is domestic or “in-house” R&D activity. Unfortunately, national accounts data using ESA2010 for all EU countries are not available from Eurostat.  From the CSO we know that such “in-house” R&D expenditure is relatively small in Ireland.

The other source is net R&D imports.  This is the main contributor to R&D GFCF in Ireland.  We can use Eurostat Balance of Payments data to see how much of an outlier Ireland is.  In 2012, in nominal terms, Ireland had the third highest level of R&D imports, behind only Germany and France with the UK on practically the same amount. 

R&D Imports EU28

It is when you look at the figures as a proportion of GDP (albeit using the ESA95 figures which Eurostat provides) that we can see just how much of an outlier Ireland is.  These expenditures are equivalent to around 3 per cent of Irish GDP.  Finland is the only other country with a level above 1 per cent of GDP.

If the R&D expenditure from Ireland is counted as “qualifying expenditure” for these patent boxes then the ‘knowledge development box’ announced in the budget has the potential to be a significant component of Ireland’s Corporation Tax regime – whether this is positive or negative depends on one’s position of the issue.

However, this is only one part of the potential of this scheme.  Thus far we have looked at the R&D expenditure that currently takes place here.  There is also the R&D expenditure that might take place here.  We know MNCs do not generate their IP here but they do exploit it here.  The largest payments from Ireland for IP are not R&D cost-sharing expenditures but patent royalties.  When it comes to these in a European context Ireland is off the chart.

Royalty Imports EU28

In Ireland these outbound payments are the equivalent of almost 20 per cent of GDP.  The next highest is the Netherlands but the figure is less than 3 per cent of GDP.

Why are these significant in terms of a ‘knowledge development box’ when these are payments for the use of IP not its development?  These payments are made to companies which own the intellectual property.  In many cases these will be Irish-registered companies (though heretofore non-resident for tax purposes).  This is the two Irish company structure in the “double-irish”.

The big change for these two-tier structure is not the proposed changes to Irish residency rules.  That will have little impact.  The big issue are the moves as part of the OECD’s BEPS project to align profits with substance, particularly employment.  At the moment many of the (particularly US) MNCs operating in Ireland have their profit-generating IP assets located in Caribbean Islands where they have little more than brass-plate operations.  If the BEPS proposals come to fruition the ability of companies to declare profits in locations where they have little or no real substance will be reduced.  In the case of IP, substance will be judged on the basis of DEMP – developing, enhancing, maintaining or protecting the IP assets.

In this instance the companies will have a couple of options:

  • First, they could move their staff involved in the DEMP of their IP assets to these Caribbean Islands to continue to avail of the low tax rates.  Given the lack of infrastructure in these locations this is unlikely.
  • Second, the companies could move the assets to where their DEMP staff are currently located.  For US companies, this is the US.  If the 35 per cent federal corporate income tax rate remains moving the assets to the US is unlikely.
  • Thirdly, the companies could move both the assets and some of the key DEMP staff to another country that makes itself attractive to both.

It is through this option that Ireland has a massive advantage that cannot be replicated by other countries – regardless of how attractive their patent box is.  The advantage Ireland has is that the IP of many US MNCs is already vested in Irish-registered companies. 

The MNCs do not want to be transferring their assets between companies as this could potentially trigger tax liabilities and administering the changes consumes significant time and resources.  They will want to restructure in the most tax-efficient and straightforward manner possible.  One of the ways to achieve this is to move the location of the IP assets but not to change the ownership of them.

It is also the case that for many US MNCs, these Irish-registered IP holding companies have entered cost-sharing agreements with the US parent to gain the rights to the MNCs IP.  Keeping the IP in the same company means that the cost-sharing agreement can be continued.  These agreements lead to R&D expenditures discussed above which will come from Ireland if the company relocates here (and possibly making the profits eligible for Ireland’s ‘knowledge development box’).

The changes announced to Ireland’s residency rules have a grandfathering period where they won’t apply to existing companies for a period of six years.  Ostensibly this was done to allow companies the time to restructure their operations.  In actuality this grandfathering period is to ensure that the companies don’t change anything – at least until Ireland’s ‘knowledge development box’ is in place.  Six years gives more than enough time for the OECD and EU discussions on patent boxes to be concluded and allow Ireland the time to legislate for a “best in class” version. 

The intention is that US MNCs will keep their IP with Irish-registered companies (and the lead-in period to the end of the year is probably with the hope that a few more companies might do so).  And the further hope is that if the BEPS proposals come to fruition the companies will choose option 3 and move their IP assets and some key IP staff to Ireland.  Lots of “hope” and lots of “maybes”.  The benefits may be massive or they may be nil.

How do we get US MNCs to choose option 3 and to do so in Ireland?  First, the tax in the ‘knowledge development box’ has to be attractive enough so that it is not worthwhile to transfer the assets out of their Irish-registered companies.  A tax rate of around five per cent should achieve that.  It will be pretty easy to relocate the Irish-registered companies to Ireland as they are only brass-plate operations so a few board meetings with appropriate decision-making agendas will achieve that.

Next, we need the MNCs to move some key staff associated with their IP to Ireland (assuming they are not here to begin with).  This may not be as straightforward.  Dublin is not London or Paris.  Ireland’s Income Tax regime is very different to our Corporation Tax regime.  Schemes like the Special Assignee Relief Programme (SARP) can help but are not well-received.  It is just one piece of a big jigsaw and there are just too many pieces that make explaining where it fits in very difficult.

Getting the necessary IP staff here will not be easy.  Ireland does not have a wide or deep pool of high-quality researchers so focusing on the D and E (develop and enhance) of DEMP is unlikely to lead to much.  Maybe we can achieve something on the M and P (maintain and protect) but it is not clear than any progress is being made in that direction.  At present we just exploit the IP and more is required to satisfy the proposed OECD criteria.

All we can say at this stage is that the ground is moving quickly.  Ireland has a significant head start over similar countries trying to gain in the area of FDI.  The outcome may be beneficial but there is no guarantee of that. We may find the ground moving out from under us and where will we end up is difficult to project.  With so many moving parts prediction is close to impossible.

The focus now is on intangibles, substance, DEMP, patent boxes and the like.  That is where the game is.  However, the purpose of Ireland’s Corporation Tax regime is to generate employment and maximise overall tax revenue.  Just because the focus is on intangibles does not mean that is where our focus should be.  The underlying primary objective has to be employment and there may not be many jobs in managing intangibles.

The tax profession will argue for a ‘knowledge development box’ because there will be plenty of work for them in that.  The changes to our Corporation Tax regime over the next few years have to be about more than on-shoring low-tax outcomes.