2/9/16: Does bank competition reduce cost of credit?

In the wake of the Global Financial Crisis, there has been quite a debate about the virtues and the peril of competitive pressures in the banking sector. In a paper, published few years back in the Comparative Economic Studies (Vol. 56, Issue 2, pp. 295-312, 2014 http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2329815), myself, Charles Larkin and Brian Lucey have touched upon some of the aspects of this debate.

There are, broadly-speaking two schools of thought on this subject:

1. The market power hypothesis - implying a negative relationship between bank competition and the cost of credit (as greater competition reduces the market power of banks and induces more competitive pricing of loans). This argument is advanced by those who believe that harmful levels of competition can lead to banks mispricing risk while competing with each other. 
2. The information hypothesis postulates a positive link between credit cost and competition, as the banks may be facing an incentive to invest in soft information. 

Now, a recent paper from the Bank of Finland, titled “Does bank competition reduce cost of credit? Cross-country evidence from Europe” (authored by Zuzana Fungáčová, Anastasiya Shamshur and Laurent Weill, BOFIT Discussion Papers 6/2016, 30.3.2016) looks at the subject in depth.

Per authors, “despite the extensive debate on the effects of bank competition, only a handful of single-country studies deal with the impact of bank competition on the cost of credit. We contribute to the literature by investigating the impact of bank competition on the cost of credit in a cross-country setting.” The authors take a panel of companies across 20 European countries “covering the period 2001–2011” and study “a broad set of measures of bank competition, including two structural measures (Herfindahl-Hirschman index and CR5), and two non-structural indicators (Lerner index and H-statistic).”

The findings are interesting:

• “bank competition increases the cost of credit and …the positive influence of bank competition is stronger for smaller companies”
• These results confirm “the information hypothesis, whereby a lack of competition incentivizes banks to invest in soft information and conversely increased competition raises the cost of credit.” 
• “The positive impact of bank competition is influenced by two additional characteristics. It is lower during periods of crisis, and the institutional and economic framework influences the relation between competition and the cost of credit.”
• Overall, however, the “positive impact of bank competition is …influenced by the institutional and economic framework, as well as by the crisis.”

The authors ‘take-away lesson” for policymakers is that “pro-competitive policies in the banking industry can have detrimental effects, … [and] banking competition can have a detrimental influence on financial stability and bank efficiency.”

I disagree. Judging by the above, higher costs of credit overall, and higher costs of credit for smaller firms, may be exactly what is needed to induce greater efficiency and reduce harmful distortions from over-lending. As long as these higher costs reflect actual risk levels.

2/9/16: Forecast-based Financing and Blockchain Solutions

As the readers of this blog know, AID:Tech (https://aid.technology/) is a new venture I am involved with that uses blockchain platform for provision of key payments facilitation services for people in need of emergency and continued assistance (refugees, international aid recipients, disaster relief aid and general social supports payments). As a part of the market analysis and strategy, we have encountered an interesting, rapidly evolving services segment relating to disaster relief: the concept of Forecast-based Financing (FBF) worth highlighting here.

Under FBF, aid providers release humanitarian aid-related funding ahead of the adverse event taking place, based on forecast information that aids in predicting the severity, timing and impact distribution of the disaster (natural or man-made). This approach to aid delivery aims to:

• Reduce key risks (e.g. assuring that delivery is timed in line with the adverse shock, focused on key geographic and demographic audiences, uses pre-disaster - and thus more efficient - supply chain networks, etc), 
• Enhance preparedness and response (by increasing quality of aid targeting and allowing to concentrate resources in the areas where they are needed most and ahead of the actual disaster impact), and 
• Make disaster risk management overall more effective by assuring that aid resources are present at the time of the disaster and after the disaster impact, thus reducing losses and delays in delivery of aid that may arise as the result of the disaster (e.g. destruction of roads and disruptions in power supplies, etc).

In general, FBF framework is open to several questions and objections, all requiring addressing.

How does FBF work? 

A humanitarian aid agency and stakeholders (e.g. meteorological services and communities at risk) jointly create a contingency plan, outlining key actions to be taken ahead of the probabilistically likely disaster or shock. They also set out specific metrics that define the trigger for aid pre-delivery, based on a model risk forecast reaching a specific threshold of probability. Linked to severity of forecast shock, specific budgets are set aside for activation. Once the risk probability threshold is breached, aid is delivered to the location of possible disaster, using pre-disaster supply chain management structures before these get disrupted by the event.


Forecast errors: are these really costly?

Probabilistic forecasts are never 100% accurate, which means that in some instances, aid will be delivered to the communities where the adverse event (a shock) might end up not materialising, despite probability models generating high likelihood of such an event. In a way, this is the risk of aid agencies providing disaster relief “in vain” or “wasting” scarce resources. It is worth noting that probabilistic errors of “wastage” can be significantly over-estimated, as some disasters can be relatively well forecast in advance (http://www.nat-hazards-earth-syst-sci.net/15/895/2015/nhess-15-895-2015.pdf). Quality of forecasting will, of course, co-determine losses in the system.

To achieve system-wide efficiency and secure gains from implementing an FBF programme, one has to be able to counter-balance the benefits of early response,including those arising from more efficiency in accessing supply chains pre-disaster and reducing the cost of disaster, against the likelihood of a loss due to probabilistic basis for the action. This can be done via two channels:

1. Assuring that during planning, the cost of acting pre-emptively, including the cost of probabilistic ‘waste’, is factored into planning for which forms of aid should be pre-delivered and on what scale; and
2. Assuring that aid supply chain and forecasting models are optimised to delver highest efficiencies possible.

Over time, development of FBF will also require changes in supply chain management to mitigate losses due to “wastage”. For example, putting more emphasis on local (or proximate) sources for supply of critical aid can reduce “wastage” by lowering cost of deliveries and by closely anchoring pre-disaster deliveries to existent markets for goods and services (so at least some pre-delivered aid can be returned into local markets in the case if probabilistically likely disaster does not materialise).

In other words, aid agencies and potentially impacted communities need to have access to timely and accurate information on which resources are needed in responding to a specific disaster, on what scale and, crucially, which resources are already available in the supply chain and in the local or proximate markets. The key element to this is ability to track in real time supply chains of goods and services accessible at differential cost to specific communities in cases of specific disaster events. The agreed (in advance) standard operating procedures (SOPs) that are set between the aid providers and the recipient communities must be both realistic (reflective of measures necessary in the case of specific disaster) and effective (reflective of the balance of cost-benefit).

Put differently, the process of FBF is the process of, first and foremost, planning and data relating to supply chain management.


Are there any tangible experiences with FBF?

One early example of FBF implementation is the case of the Red Cross Red Crescent Movement that has field-tested an FBF programme Uganda and Togo. This project bridged financial and technical support from the German government and Red Cross, and used technical support from the Climate Centre.

Another case is of FoodSECuRE initiative by the World Food Programme that is currently in planning stages. In this programme, private sector partners (aviation services providers, insurance companies etc) are engaged in FBF planning for alleviation of potential flooding due to El Niño effects in Peru (http://www.climatecentre.org/downloads/files/FbF%20Brochure4.pdf and http://www.climatecentre.org/programmes-engagement/forecast-based-financing). Both of these experiences show also the importance of setting aside sufficient response funds for FBF delivery.

Further afield, FBF pilots are being run or planned by the WFP and other organisations in Bangladesh, the Dominican Republic, Haiti, Mozambique, Nepal and the Philippines.

Note: the above cases were provided by the UNFDP research.


Overall, FBF is becoming one of the cornerstones of the global disaster aid delivery programmes and was endorsed by UN OCHA and the IFRC. FBF was also included in the International Federation’s special report ahead of the World Humanitarian Summit in Istanbul. The report included a pledge to facilitate a doubling of FbF within the Movement by 2018.

However, despite the aid agencies enthusiasm, the key problem relating to FBF remains largely unaddressed: currently, with some 20 percent of disaster aid being lost due to insufficient supply chain management, fraud and theft, delivering properly structured FBF requires exponentially greater exposure to data collection and analysis, as well as to strengthening of real-time supply chain visibility systems.

As AID:Tech example shows, these objectives can be supported via private and semi-private blockchain solutions.

2/9/16: Investment in Italy: Banks, Capital and Firms Structures

In my course on the enterprise and financial risk last semester, we talked about the peculiar (or idiosyncratic) nature of Italian firms across a number of matters:

• Relationship banking;
• Firm governance: family ownership, equity distribution and aspects of firm strategy and operations;
• Firm capital structure (leverage risks in particular);
• Firm dividend policy choices, etc.

Now, let’s add to that literature something new. A recent paper from the Banca d’Italia, titled “Investment and investment financing in Italy: some evidence at the macro level” by Claire Giordano, Marco Marinucci and Andrea Silvestrini (Banca d’Italia Occasional paper Number 307 – February 2016) looks at the evolution of the Global Financial Crisis and the Great Recession across the Italian economy in terms of credit fundamentals.

As noted by the authors, “following the outbreak of the global financial crisis, the euro area experienced a large fall in gross fixed capital formation, both in 2008-09 and during the sovereign debt crisis. This drop was dramatic in the countries more exposed to tensions in government bond markets. In Italy, in particular, total real investment has suffered a loss of around 30 per cent since 2007, the pre-crisis peak, reverting to its lowest levels since the mid-1990s. Weak investment also remained a key drag on GDP growth in 2014, although more recent quarterly data on capital accumulation point to a slight increase over the first three quarters of 2015 relative to the corresponding period in 2014.”

In a typical European fashion, investment in Italy must equal debt. In part, as I usually cover in my courses on risk and corporate financial strategies, this is tied to the reluctance of the family-owned firms to release equity. And in part, it is a part of a broader European debt disease. Independent of the reasons, per authors, the worthy corner to check for key investment-crises interlinkages is the credit supply.

“The depressed growth of investment is in contrast with the substantially muted aggregate financing costs, which stem from the low interest rate environment resulting from the strongly expansionary stance of monetary policy in the euro area. In this context, one scenario is that investment demand will remain too low to absorb financial savings, inducing a persistent state of an excess supply of funds in capital markets.” In other words, Italians are discovering secular stagnation: interest rates are too low because investment is too low (and may be, also the other way around).

With that in mind, the authors proceed to show that “medium-term gross fixed capital formation trends in Italy may be summarised along the following lines”:

Pre–2007 capital expansion “was broad-based, both across institutional sectors and asset categories, although less marked for households”, and Post-2007 “exceptional downturn …affected all sectors and components, yet to a different extent. In particular, focusing on the most recent period, the decline in general government and non-financial corporations’ expenditure, cumulatively undertaking about two thirds of total investment in Italy, was sizeable (approximately 25 per cent), yet slightly more contained than the concurrent drop in household investment spending.”

Overall, “the total-economy investment rate in Italy currently stands at its lowest levels since data became available in the mid-1990s; current government and non-financial corporation investment rates are comparable only to those recorded in 1995; the household rate is even lower. From the asset side, in recent years construction investment, in both residential dwellings and non-residential buildings, which represents half of total expenditure, was the hardest-hit item, excluding transport equipment expenditure (small and volatile), whereas ICT investment and the accumulation of intangible assets weathered the recent double-dip recession better.”

But the age of lower interest rates is having another impact on Italian credit markets. This impact is aggressive rolling over of maturing loans amidst deteriorating credit quality of borrowers. As the result, two things can be documented in Italian credit market experience since 2008 bust:

Firstly, overall, the trend toward longer debt maturity is present in household, corporate and government debt sectors (charts below). Secondly, this is doing nothing to repair credit quality on banks balance sheets.

Which is rocking, for now, as low interest rates are keeping debt burden down and allowing leverage to rise. But the problem is that with longer maturity of debt, we are looking at higher long term susceptibility to debt servicing costs. Last time that happened, Italy became on of the PIIGS. Next time it will happen… oh, ok… we shall just wait and hope Mario Draghi says true to his Italian roots long enough for the miracle to happen.

2/9/16: Interest Rates, Financial Cycles and the Real Economy

Claudio Borio and his team at the Bank for International Settlements have just published another interesting working paper, titled “Monetary policy, the financial cycle and ultra-low interest rates” (BIS Working Papers No 569 by Mikael Juselius, Claudio Borio, Piti Disyatat and Mathias Drehmann Monetary and Economic Department July 2016).


In the paper, the authors ask whether “the prevailing unusually and persistently low real interest rates reflect a decline in the natural rate of interest as commonly thought?”

The authors “argue that this is only part of the story. The critical role of financial factors in influencing medium-term economic fluctuations must also be taken into account.” In other words, the authors attempt to control for purely financial factors driving interest rates first, and then consider predominantly real economic variables-determined rates (natural rates).

You might think that the currently low rates are facilitating the real economy, right? If so, then actual observed (already low) rates today should be coincident with even lower ‘natural’ rates (if real economy drags down the financial economy). Alas, as the authors find: accounting for the different sources of pressure on the interest rates (financial vs natural), in the case of the United States, “yields estimates of the natural rate that are higher and, at least since 2000, decline by less.”

Oops… so persistently low interest rates today are below natural rates and reflect the needs of the financial intermediation sector.

Notice the difference between the observed rates (yellow) and the ‘natural rates’ (red). Or as the lads from BIS put it: “As a result, policy rates have been persistently and systematically below this measure.”

But never mind. With time, things should get rebalanced, as the authors also find that “monetary policy, through the financial cycle, has a long-lasting impact on output and, by implication, on real interest rates. Therefore, a narrative that attributes the decline in real rates primarily to an exogenous fall in the natural rate is incomplete. The influence of monetary and financial factors should not be ignored. Exploiting these results, an illustrative counterfactual experiment suggests that a monetary policy rule that takes financial developments systematically into account during both good and bad times could help dampen the financial cycle, leading to higher output even in the long run.”

Yah, yah… lots of talk. What’s the meaning? Ok, the authors take two drivers of financial sector impact on the real economy: leverage and credit.

Leverage gap is defined as basically a credit to assets ratio for the economy - or how much credit does economy create per each unit of assets. Meanwhile debt service gap is the ratio of debt service payment, or more precisely, “the ratio of interest payments plus amortisations to income”.

To understand the dynamics of the monetary (interest rates) policy impact, the authors do a couple of experiments. The main one is worth discussing. The authors start with a leverage gap of -10%, so there is an excess of assets over credit in the economy and hence there is room to borrow, driving leverage gap up. Note: as the authors point out, the -10% leverage gap assumption is consistent with historical reality: in the late 80s and mid-2000s, “at their trough”, leverage gaps were -11% in 1987 and -20% in 2006 respectively.

So, as I noted above, “a negative leverage gap initially induces a credit boom that then turns into a bust… Initially, the negative leverage gap is followed by rapid credit growth, which in turn feeds into a positive, albeit small, increase in private sector expenditure. But as credit outgrows output, the credit-to-GDP ratio and with it the debt service gap start to rise, putting an increasing drag on output and asset prices. A severe and drawn-out recession follows.”

The dynamics match the Great Recession: “…at the start of 2005, the real-time estimate of the leverage gap was significantly negative while the debt-service gap was positive. Given this starting point, the adjustment dynamics of the system would have predicted much of the subsequent output decline during the Great Recession. This suggests that the recession was not a “black swan” caused by an exogenous shock but, rather, the outcome of the endogenous dynamics of the system – a reflection of the interaction between the financial factors and the
real economy.”

And here is the actual run of annual estimates of the two gaps:

Remember, the cyclicality? Negative leverage gap —> credit boom —> positive leverage gap and positive debt service gap —> bust.

Good thing we are not going to repeat THAT cycle this time around, right?.. Not with all the low interest rates not being lower than ‘natural’ rate… right?

2/9/16: Remember Banks Stress Tests: Tripple Farce and Still No Joy for Ireland

Couple of older, but still relevant notes have stacked up on my virtual desktop over the last few weeks. Catching up with these, here is a post on the banking sector 'bill of un-health' produced this summer by the EBA.


European banks street tests conducted by EBA last month combined the usual old farce with the novel new farce. Just to make sure the punters were not too scared of the European economy’s champions.

Based on Basel III criteria - CT1 ratio of 7% post shock - all but two banks (Italy’s Banca Monte dei Paschi di SAiena Spa and Austria’s Raiffeisen Zentralbank) have managed to escape the tests with CT1 ratios post-shock within the Basel III parameters. Or in other words, everyone passed, save for two who didn’t. Systemically, therefore, EBA can assure us all that euro area banking is just fine. Nothing to see, nothing to worry about.

However, the farce of the tests goes deep than this predicable and historically conditioned outcome. Because this time around, EBA no longer even bothered with determining who ‘failed’ and who did not. Like in Breznev’s USSR, in the EBUSSR, ‘friendship wins’ and ‘no one loses’.

There was another predictable trend in the EBA results. No matter how ‘flexible’ the models fort testing get, no matter how being the ‘stress assumptions’ get, Irish and Italian banks remain the sickest puppies in the entire ward of already not too healthy ones:

But, hey, despite much of the stock markets hullaballoo over recent months, the bidding of banks’ equity has not really done much in terms of beefing sufficiently their capital buffers. So here are some comparatives on 2014 stress tests against current ones.

Note: 2014 stress tests estimated impact of a shock out to 2016, while this year tests are estimating impact out to 2018.

So behold (via @FT):

Italy:

• 2016 state: Transitional CET1 ratio of 6.14 per cent v 8.42 per cent average - under performing the average by 228 bps
• 2018 state: Fully loaded CET1 ratio of 7.62 per cent v 9.2 per cent average - under performing the average by 158 bps
• Signals improvement, on the surface, but this is a cross comparative over tow somewhat different benchmarks

Ireland:

• 2016 state: Transitional CET1 ratio of 7.05 per cent v 8.42 per cent average, undershooting the average by 137 bps
• 2018 state: Fully loaded CET1 ratio of 5.21 per cent v 9.2 per cent average, underperforming by 399 bps
• Signalling, even if we are to totally disregard differential quality, this does not bode too well for Irish financial ‘giants’

FT did provide a handy chart showing changes in stress test shock-level CET1 ratios for Adverse Stress Scenarios in 2014 tests and 2016 tests (never mind the ‘actual’ levels as of 2015, as these are subject to market valuations etc).

What the above shows?

For a tiny banking system, Ireland’s one is sicker than any other. And this comes on foot of years of repairs, recapitalisations, arrears resolutions etc etc etc. Green Jerseying ain’t working, folks. All Spanish banks are performing better than the two Irish flagships. Majority of Italian banks (save for one) are better than the two Irish ‘giants’. All Portuguese banks are stronger than the Irish systemically-important institutions. And none have spent anything close to Ireland on ‘repairing’ their lenders.

Maybe, if we wait long enough, EBA will include a bunch of Greek and Cypriot banks next… to make ours look better…

2/9/16: Getting Back to the Blog

Folks, the joys of moving to my new job have been quite all-consuming over the month of August, and the joys of preparing for the move - over June and July - were testing as well. Now, with things more settled, it is time for me to come back to blogging, so stay tuned and read.

30/7/16: Europe's Bank Stability in Question

My comment for Euromoney on banking sector risks: http://www.euromoney.com/Article/3574602/Europes-bank-stability-is-in-question-say-country-risk-experts.html.

29/7/16: Tax Regime, Apple, Fraud?

We have finally arrived: a Nobel Prize winner, former Chief Economist and Senior Vice-President of the World Bank (1997-2000) on Bloomberg, calling Apple's use of the Irish Tax Regime 'a fraud': http://www.bloomberg.com/news/articles/2016-07-28/stiglitz-calls-apple-s-profit-reporting-in-ireland-a-fraud?utm_content=business&utm_campaign=socialflow-organic&utm_source=twitter&utm_medium=social&cmpid%253D=socialflow-twitter-business.

This gotta be doing marvels to our reputation as a place for doing business and for trading into Europe and the U.S.

The same as Facebook's newest troubles: http://www.irishtimes.com/business/technology/facebook-tax-bill-over-ireland-operation-could-cost-5-billion-1.2738677.

But do remember, officially, Ireland is not a tax haven, nor is there, officially, anything questionable going on anywhere here. Just 26.3 percent growth in GDP per annum, and booming corporate tax revenues that the Minister for Finance can't explain.

22/7/16: Brexit сближает Ирландии...

My comment for Kommersant (in Russian) on Brexit risks for Ireland: http://www.kommersant.ru/doc/3043700.

21/7/16: Article 50: Facts

There is a lot of poorly informed nonsense being pushed around the media (especially 'new media') about Article 50 process relating to the Brexit. The best, most cogent and brief summary of what actually is involved in this process was provided back in February by the Open Europe think-tank here: http://openeurope.org.uk/today/blog/the-mechanics-of-leaving-the-eu-explaining-article-50/.

20/7/16: McKinsey's "Generation Worse"...

A new study from McKinsey looks at the cross-generational distribution of income as a form of new ‘inequality’, in words of the authors: “an aspect of inequality that has received relatively little attention, perhaps because prior to the 2008 financial crisis less than 2 percent of households in advanced economies were worse off than similar households in previous years. That has now changed: two-thirds of households in the United States and Western Europe were in segments of the income distribution whose real market incomes in 2014 were flat or had fallen compared with 2005.”

In other words, McKinsey folks are looking at the “proportion of households in advanced economies with flat or falling incomes” - the generational cohorts that are no better than their predecessors.

Key findings are frightening: “Between 65 and 70 percent of households in 25 advanced economies, the equivalent of 540 million to 580 million people, were in segments of the income distribution whose real market incomes—their wages and income from capital—were flat or had fallen in 2014 compared with 2005. This compared with less than 2 percent, or fewer than ten million people, who experienced this phenomenon between 1993 and 2005.”

So that promise of the ‘sharing economy’ and the ‘gig-economy’ where people today are enabled to derive income (and thus wealth) from hereto under-utilised ‘assets’… pwah! not doing much. The ‘most empowered’ - web and gig-economy wise cohorts? Ah, they are actually the “worst-hit” ones. “Today’s younger generation is at risk of ending up poorer than their parents. Most population segments experienced flat or falling incomes in the 2002–12 decade but young, less-educated workers were hardest hit”.

For those of us who, like myself, tend to be libertarian in our view of the Government, McKinsey study tests some of our accepted ‘wisdoms’: “Government policy and labor-market practices helped determine the extent of flat or falling incomes. In Sweden, for example, where the government intervened to preserve jobs, market incomes fell or were flat for only 20 percent, while disposable income advanced for almost everyone. In the United States, government taxes and transfers turned a decline in market incomes for 81 percent of income segments into an increase in disposable income for nearly all households.”

Except, may be it did not, because counting in disposable income while allowing for taxes and subsidies is notoriously difficult and imprecise. And may be, just may be, all the fiscal imbalances that were accumulated in the process of achieving these supports in some (many) countries will still have to be paid by someone some day?

There is a reduced connection between current growth metrics and income outcomes on the ground (don’t we know as much here in Ireland, with 26.3% jump in GDP in 2015?): “Before the recession, GDP growth contributed about 18 percentage points to median household income growth, on average, in the United States and Europe. In the seven years after the recession, that contribution fell to four percentage points, and even these gains were eroded by labor market and demographic shifts.”

And the forward outlook? Bleak: “Longer-run demographic and labor trends will continue to weigh on income advancement. Even if economies resume their historical high-growth trajectory, we project that 30 to 40 percent of income segments may not experience market income gains in the next decade if labor-market shifts such as workplace automation accelerate. If the slow growth conditions of 2005–12 persist, as much as 70 to 80 percent of income segments in advanced economies may experience flat or falling market incomes to 2025.”

There are some wrinkles in the study. For example, in the U.S. case - cross time comparatives do not provide for the same data base, as pre-2014 data does not include state and local taxes. VAT and sales taxes are omitted across the board. And some other, but overall, the paper is pretty solid and very interesting.

So here is the key summary chart, positing the massive jump in the numbers of households on the declining side of market incomes:

And the chart showing that the taxes and transfers side of income supports is no longer sustainable over time:

Which brings us to the main problem: on the current trend line, politics of income supports from the fiscal policy side are unlikely to be able to contain growth in political discontent. Advanced economies are heading for serious tests of democratic institutions in years to come. Buckle your seat belts: the ride is going to get much rougher.

15/7/16: A Booming Tax Haven? And Why Not?

Yes, we all know, Irish GDP in 2015 grew by an officially idiotic 26%. And yes, I am no longer gracing these illusionary / delusionary numbers with an attempt at any serious analysis. Doing so would be too big of an intellectual subsidy to the world of Irish officialdom. So here are two quite opposite (in their top-line accuracy) views of the same problem, that both, in the end, arrive at the same conclusion:

One view is from Fortune magazine http://fortune.com/2016/07/13/ireland-tax-haven-gdp-up/ where the headline says all you need to know

And another is from the Irish Times, that rows in with a more 'diplomatic' (aka - easier to chew by the Dons of the Irish Civil Service) discourse that is worth reading, despite it containing some pretty delusional (see my twitter stream from today on this) statements: http://www.irishtimes.com/business/economy/ireland-s-gdp-figures-why-26-economic-growth-is-a-problem-1.2722170#.V4i7kuyVTic.twitter.

In the nutshell, Fortune got it right at the top level, Irish Times got some beefier discussion of the details.