Mr. Pat McArdle:

Therefore, I am going to take, Chairman, my witness statement of 10 April and supplementary statement of 27 April as read, and highlight some of the issues that arise.

While my main focus as chief economist in Ulster Bank was external, and while I was not formerly part of the internal UB decision-making process, I accept that my voice was probably an influential one within the bank, particularly as regards to so-called soft landing. My team's emphasis was on short-term forecasting and we produced and published quarterly forecasts for GDP, inflation, employment, etc. These high-frequency short-term forecasts are designed to paint a picture over a two-year time horizon or occasionally less. Longer-term projections in Ireland are published by the ESRI, which produces forecasts for up to a quarter of a century ahead every two or three years. And, you've already discussed these with Dr. FitzGerald. For the economic input into the five-year planning circle ... cycle, Ulster Bank used the ESRI medium-term forecasts. As all ESRI forecasts, benchmark, high and low variants, envisaged a soft landing - and I refer you to my table in my supplementary witness statement - it follows that all UB medium-term planning was also predicated on a soft landing. I effectively buttressed this view in terms of my rolling quarterly short-term economic updates and ad-hocreviews. My forecasts, and those of all other short-term forecasters, remained positive up until early 2008. The turnaround was then dramatic. At that time, the central expectation was that 2009 GDP would grow by 3.6%. Eighteen months later, the consensus was that it would fall by about 8.3%. This about-turn came too late to influence events, as the credit that fuelled the boom had been extended between 2004 and 2007. I was generally regarded as being towards the more pessimistic end of the range of short-term forecasters ... certainly over that period.

I should add that any views in the UB publications were mine alone and there was no difference between the internal and external positions I adopted. I was given freedom of speech by Ulster Bank, even if some of the things I said made them uncomfortable on occasion. I was a trenchant critic of the loose fiscal policy adopted and I was an outspoken advocate of a soft landing. In general, I did not input directly into the credit, the risk or the lending functions in Ulster Bank. Instead, my job was to produce regular economic briefings and they used them as they saw fit. However, there was the occasional exception and I recently came across a case where UB credit drafted a report based on my output and, somewhat unusually, I received a copy of it. It summarises my views in early 2004 and if you don't mind, Chairman, I'll quote from it:

UB Economic Research unit (Pat McArdle) views on the ROI Housing Sector remain positive in the updated report dated 25 February 2004. Its rating however is amber, taking due recognition of the cycle. ... Neither of the usual major negative influences on the housing market (rapid rises and interest rates and unemployment) is forecast. ...

Major changes are due on the supply side - last year['s] completions were in excess of 68 [thousand] but underlying demand is believed to be [around] 50 [thousand]. The difference is accounted for by pent up demand and holiday homes. There are signs that the building industry is aware of this and has the capacity to orchestrate a smooth transition but this is not certain. Pat McArdle does not rule out the possibility of a price correction of up to 20%. Even if this were to happen we do not think that this would have major knock-on effects, given bank capital requirements and stress testing.

A 20% fall in nominal prices would've equated to a much greater fall in real prices, about 35%, that's when you're after making allowance for inflation. In effect, therefore, I warned, in February 2004, that real house prices could decline by 35% ... that's within the 30% to 50% range predicted by the two main contrarians some three years later. My main regret, Chairman, is that I did not continue in this vein. I should have stepped up my warnings as the housing market initially slowed and then regained momentum and even stronger growth. However, if anything, my later pronouncements are more consensual in that I did not disagree with the projected 15% to 20% fall in real prices advocated by the Central Bank and the ESRI. Of course, had I continued in the vein I just outlined, I would've been treated as a contrarian. Few agreed with them, and while it's cold comfort, it's unlikely that I would've been listened to either. Indeed, it's ironic that even though I disagreed with the contrarians later, we had stress tested in Ulster Bank for their more extreme predictions. The problem was not house prices, per se, but developer and real estate lending and we did not spot it or stress test for it, certainly sufficiently.

I want to make the following points. A soft landing and it's perhaps no harm to dwell on it for a minute since this has received such currency and it may not be as understood in the sense that I understand it. So a soft landing would've been quite bumpy, involving substantial job losses and up to €3 billion, I had estimated, in tax revenue foregone. In other words, it would've left a major hole in the budget. The regulators, management and board bear most responsibility for the collapse here as elsewhere, and I'll come back to that. Ulster Bank rejected the hard landing hypothesis, but stress tested for house price declines of 56% in real terms. That was at the very upper end of the range predicted by the contrarians. This was insufficient because the actual recession was much greater than anyone predicted.

I go on now, Chairman, to talk about the implications of a soft landing for a moment. It could've been, as I said, quite bumpy. A soft landing that saw house completions fall from their peak around 90,000 to the 50,000 sustainable medium-term level estimated by the ESRI would've had significant implications. I'd calculated that every 10,000 houses represented one percentage point off growth and that a soft landing would involve 30,000 construction job losses and up to €3 billion in tax revenue foregone.

I go on to speak about the regulator. There were no strong incentives for banks to make arbitrary judgments about the prudent limits of credit expansion. Hence, the infamous comment by Chuck Prince, chairman of Citigroup: "When the music stops ... things will be complicated, but as long as the music is playing, you've got to get up and dance." The regulator's job is to stop the music, or as it's more often put, "To take away the punch bowl just as the party gets going". All regulators failed to do this. This was a major failing and much of the focus post-crisis has centred on improving bank supervision. Indeed, responsibility for bank supervision of euro area banks has been removed entirely from national supervisors.

When I worked in the banking area of the Department of Finance in the '70s and '80s, I had the distinct impression that when the regulator said, "Jump", the bank's response was, "How high?" I was therefore shocked to read in the Honohan report that the regulator spent almost a decade in fruitless correspondence with one financial institution without ever achieving anything. Clearly, the boot had shifted to the other foot and the regulated, instead of the regulator, were now calling the shots. However, it seems, and I've only learned this recently, it seems that this only applied to prudential regulation, as bank representatives have testified to this inquiry last week that the regime on the consumer side remained quite strict, which would've been my memory of it from 20 years earlier. Therefore, I disagree with the conclusion in the Honohan report that the major responsibility lay with the directors and senior managements of the banks that got into trouble. In my view, the regulator had a higher degree of responsibility. And this should go without saying, really, because the regulator's job is to promote the safety and soundness of the banking system.

If the banks were capable of regulating themselves, there would, of course, be no need for rules and regulations and regulators. In saying this, I just want to emphasise that I am by no means trying to absolve the banks from blame. As is clear from the Nyberg and other reports, that there were many actors involved and that the banks were up there at the top. However, the regulator was the only one who had full information on large exposures and, critically, the only institution that could have sought to curb excessive balance sheet growth. It was only when the loans were transferred to NAMA in 2010 that it was revealed that the big developers had multiple exposures to the different banks. The regulator should have had full details from the large exposures reports that it received - I think they were quarterly - and it would've been a simple matter to add them up. Finally, in this section, in 2009, Patrick Honohan, then professor of economics at TCD, wrote, and this has been quoted here yesterday:

A very simple warning sign used by most regulators to identify a bank exposed to increased risk is rapid balance sheet growth. An annual growth rate of 20 per cent real is often taken as the trigger ... Anglo Irish Bank, crossed it in eight of nine years, and indeed its average annual rate of growth 1998-2007 was 36 per cent ... So this was a very obvious and public danger sign.

On this basis the alarm bells should've been ringing for the best part of a decade.

Stress testing. Ulster Bank stress tested for a 36% fall in nominal house prices in its severe stress scenario. That was equivalent to a real fall ... a fall in real prices, sorry, of 56%, i.e. it was at the very upper end of the range predicted by the contrarians. Indeed, it was slightly above the upper end, I think. However, the real problem was commercial and not mortgage lending, and this was not adequately stress tested. In 2007, at the behest of the Bank of England FSA, global adverse scenarios for the UK, US and euro area were produced by RBS Group economics and we commissioned the ESRI to help translate them into three Irish shock scenarios: a mild, once in a decade, a medium, once in a quarter century, and a severe, once in century recession. I supplied the ESRI with the external ... these external global assumptions for the three shock scenarios in April 2007. These scenarios are based on global economic shocks, which severely affected growth, employment, credit, etc., and we left it up to the ESRI and their model to determine the impact of this on the Irish economy and on Irish house prices.

The severe scenario envisages a downturn of a magnitude not seen in the post-war period, greater than the 2009 UK recession and greater than any ... anyone else ... there was no other one that was known at that time. It entailed GDP everywhere going negative for a few years, unemployment rising to low double digits, short-term interest rates falling to be 1% or below, and Irish credit contracting by 5% per annum for a few years. This results are given in the accompanying table. In general, the outcome was worse than predicted ... that's the outcome in the event was worse than the scenarios we utilised I just should say for clarification ... and this was mainly because the impact of the construction rate of decline ... related decline, was vastly greater than expected, with new house completions virtually coming to a halt, whereas the model had predicted a fall of 20,000 units only. The predicted nominal house price fall, however, was 36% and, of course, the actual, as we now know, was 50%.

In conclusion, Chairman, the financial crisis has been described as an example of groupthink. In my experience, and I've now been through it, groupthink is almost impossible to understand unless you've experienced it. I suggest that a useful way to approach it is, perhaps, to look at two more recent examples. First, the Central Bank's proposed macro-prudential limits on mortgage lending were eminently sensible and less severe than the regime that applied when I took out my first mortgage, yet they were almost universally opposed, even by the Department of Finance, and were watered down in the event. Second, the expansionary 2015 budget was condemned by IFAC, the Irish Fiscal Advisory Council, a body expressly established to advise on policy and also by most economists, yet the Government persisted with widespread support from beneficiaries and politicians. So in my opinion, groupthink may well be alive and well.

The Bank for International Settlements was the only major institution I know of to warn of the crisis and I want to end with a quote from it. In a presentation in 2013, Bill White, who was the former chief economist there, reflected on why his warnings were ignored, and this is also relevant to groupthink. He said, and I quote:

There were a few who did warn that there were serious problems building up under the smooth surface of the Great Moderation. I would like to believe that we at the BIS saw it more clearly than many others, though certainly the timing and the precise nature of its unfolding eluded us ... Why were these warnings (both public and private) not heeded? Why were the historical antecedents not given more emphasis? I am going to suggest in the immortal line of Flanders and Swan, that it was "A Tale of Seduction". All of the parties who contributed to the crisis (borrowers, lenders, regulators, central banks, academics and politicians) were each seduced by [the] various influences into believing different things that were not true. Moreover, since seduction normally involves more than one party, the relationships between these various parties also contributed to their having 'no eye to see and no ear to hear'.

Thank you, Chairman.