Scott Sumner  

The Irish miracle

Hayek's Views on Emergence... Midweek Musings...

There's been a lot of recent discussion of the new GDP statistics out of Ireland, which show 2015 RGDP growth of 26.3% and NGDP growth of 32.4%. Almost everyone agrees that the data is somewhat fishy, but it's not clear whether that means "wrong" or "accurate but misleading". Based in what I've read, I vote for "wrong".

Half of the growth in NGDP is due to a massive jump in depreciation. This sector had been rising at a very gradual pace from many years, from 25,087 in 2010 to 30,891 in 2014. Then it doubled to 61,558 in 2015. That almost certainly did not actually happen; rather it probably represents multinationals writing off certain assets in a lumpy fashion.

Paul Krugman links to another explanation:

The increase in investment, although you can't see it in the national accounts, is being driven by airline leasing. My hunch is that this has increased by about 110%. Airline companies of the world are effectively transferring their financial activities (as new aircraft machinery) into Ireland for tax purposes. As a student of mine nicely put it: imagine all those massive Boeing planes flying around the world, then imagine them in Ireland, and hundreds of people working on them. Where are they?

In truth. We couldn't even fit these planes in Ireland. It's just around 20 people managing a financial fund for tax avoidance purposes. Then using the generated money for profit redistribution. That's what's really go on.

The increase in exports, although more real, and somewhat more complicated, is a result of a similar dynamic. It's large corporations transferring assets and IP patents into Ireland - with no real connection to employment - and then booking it as real investment, for tax purposes.

This makes no sense to me. Suppose assets like airliners or patents were transferred to Ireland. That would indeed represent investment, as their capital stock would rise rapidly. But it should also show up as a boost in imports, which subtract from GDP. So there is no first order effect on GDP (which isn't to say it might not make Ireland more productive, and gradually boost NGDP over time). Even worse, the article suggests the airplanes were not even located in Ireland.

I've argued that NGDP targeting is not always appropriate for small open economies, citing examples such as Australia and Kuwait. Actually, it's probably much more appropriate for Australia than Ireland. The key is whether NGDP tracks total labor compensation fairly closely. Where it does, as in the US, then NGDP targeting is appropriate. Where it doesn't, as in Kuwait, then you want to target total labor compensation, perhaps per capita.

In terms of monetary policy, it doesn't matter whether the Ireland data is wrong, or correct but misleading. Either way that huge surge in reported NGDP does not reflect labor market conditions, and hence is not an appropriate target for NGDP. (I didn't even check the wage data, it's one of those things you just "know", if you are numerate.)

And the Irish really need to get some better national account statisticians.

HT: Tyler Cowen, Gordon

Comments and Sharing

COMMENTS (9 to date)
Miguel Madeira writes:

I had the idea that the point of the airplanes were something like an american air company renting an airplane from a company registered in Ireland, and the rent being added to the Irish GDP.

Shane L writes:

It has been known for a long time that Irish GDP is inflated by the headquartering of multinational corporations here. This causes complications with understanding things like debt-to-GDP ratio or tax revenue as a percentage of GDP. Ireland looks richer than it really is. (Interestingly, this causes Ireland to pay higher contributions to the EU than it would under a more accurate accounting of economic size.)

"Over the last few years a number of companies have relocated their headquarters to Ireland without generating any real activity in the economy in terms of employment or purchases of domestic inputs... However, the benefits of the retained profits of redomiciled plcs are attributed to their foreign owners – there is no benefit to the Irish economy. Nonetheless, this has the effect of raising the measured current account surplus in the Balance of Payments and increasing the level of nominal GNP arising in Ireland."

The above paper describes an adjusted-GNP that takes into account this phenomenon, which showed that the economic crisis had been even worse than suggested by the official figures.

Today the Central Statistics Office has published a note exploring the odd increase in GDP. They point to much more modest changes in personal consumption of goods and services (+4.5%) and net expenditure by government on current goods and services (+1.1%) between 2014-15.

They add that they will convene a "high-level cross sector consultative group" that will explore the indicators and consider new indicators, which seems sensible!

In a way it is a bit of a shame that this has gotten so much attention since it distracts from the general consistent economic recovery experienced in Ireland in recent years, somewhat against the odds. Latest unemployment figures were at 7.8%, down from 15+% in 2011, while Purchasing Managers' Indices show Ireland leading the recovery in the EU.

Frank writes:

[Comment removed for supplying false email address. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Scott Sumner writes:

Miguel, Yes, that's possible.

Shane, Thanks for that info. And just to be clear I do agree that Ireland has done relatively well in recent years, recovering faster than the other "PIIGS", for instance.

For international comparisons of living standards, consumption might be better than GDP.

Philo writes:

You are known for advocating that the monetary authority target the level of NGDP. That's fairly simple; complicating the message would breed confusion. But then you say "you want to target total labor compensation, perhaps per capita." If, but only if, "NGDP tracks total labor compensation fairly closely . . . , then NGDP targeting is appropriate." In other words, NGDP is just a proxy for total labor compensation, to be targeted only if the two are highly correlated and the data on NGDP are easier to get. Some readers will be confused about which variable you really want to target.

Furthermore, whichever is chosen it is not the present value of the variable that you want to target but rather the expected value n months into the future. Unfortunately the value of 'n' is uncertain. Now add the uncertainty inherent in "perhaps per capita," and the average reader's head will be spinning. Politically your ideas are looking like a hard sell (though you have been somewhat successful so far; good luck in the future!).

Paul T writes:

If Amazon transfers the asset of an ebook to its Irish subsidiary, and then sells millions of copies of that ebook around the world for profit, doesn't that count as a net export from Irelands perspective?

James Alexander writes:

This was written in March and predicted trouble ahead for GDP data.

It also includes a sensible discussion of much lower consumption growth:
"Despite this household domestic consumption increased with a sturdy 3,5% which can be explained, more or less, by 2,3% employment growth in combination with an increase of hours per job, though de 0,5% decrease (!) of average wages in the fourth quarter was, of course, a negative for consumption."

Scott Sumner writes:

Philo, Yes, I'm known for advocating NGDP targeting, but that's primarily because I focus on the US, where the two targets are almost identical (especially if focused on expectations.)

I would rather risk "muddying the message" than expose my ideas to ridicule by claiming that they apply to cases (like Ireland) where they clearly do not apply.

Paul, Yes, but the import of the ebook asset would tend to offset the export of the actual ebooks.

James, Good call.

Prakash writes:

Minor quibble

" NGDP does not reflect labor market conditions, and hence is not an appropriate target for NGDP. "

Shouldn't that be

"NGDP does not reflect labor market conditions, and hence is not an appropriate target for monetary policy"

Comments for this entry have been closed
Return to top