Dr. Peter Bacon:

Thank you, Chairman. I make this statement on foot of your invitation to attend this morning, to assist you in your deliberations. You have requested that I address all or some of the following matters in my evidence: the development of the proposal to establish NAMA, including the options assessed and the conclusions reached; tax policy towards housing and property development; planning and development during the boom; Ireland's housing market in the late 1990s; debate about housing policy prior to the crisis; Ireland’s housing market in the international context; and recommendations made by me in respect of the residential property market in Ireland and their implementation.

I am happy to endeavour to assist you and the committee in relation to these issues on the basis of certain consultancy assignments carried out by me and which have been published or placed in the public domain as follows: An Economic Assessment of Recent House Price Developments, a report submitted to the Minister for Housing and Urban Renewal (April 1998); The Housing Market: An Economic Review and Assessment, a report submitted to the Minister for Housing and Urban Renewal (March 1999); The Housing Market in Ireland: An Economic Evaluation of Trends and Prospects, a report submitted to the Minister for Housing and Urban Renewal (June 2000); and Evaluation of Options for Resolving Property Loan Impairments and Associated Capital Adequacy of Irish Credit Institutions: Proposal for a National Asset Management Agency (NAMA), Abridged Summary of Report (8 April 2009).

I will start with the housing market. Developments in Ireland’s housing market in the late 1990s, in the Dublin region in particular, were symptomatic of fundamental change in housing patterns. House price inflation, especially in Dublin, began accelerating from 1993, reaching 14% per annum in the four years to 1997 and 25% in 1997.

These trends were the result of favourable macro-economic developments in Ireland including lowering interest rates, reinforced by demographic factors and changing social patterns. For example, gross immigration was occurring at an annual rate of 44,000 and concentrated in household formation ages, almost half being aged between 25 and 44 years. By contrast, emigration was concentrated in the younger age of 15 to 25 years, about 62% being of that age. Changing social patterns were reflected in a rise of one and two person households from 41.9% of the total in 1988 to 46.8% in 1997. While housing output increased 80% between 1993 and 1997, the share of completions in Dublin fell.

The recommended policy response contained in the reports considered that, to be effective, a policy response would need to: achieve a better balance between demand and supply in the short term; improve the potential supply of housing in the short to medium term; engage in infrastructure developments; and improve medium and long term planning of development of the east region. The April 1998 report proposed specific policy initiatives under each of these headings, although most debate and commentary focused on the fiscal measures which comprised the following: the repeal of section 23 relief from investment in residential property; the removal of deductibility of interest on borrowings undertaken for investment in residential property against personal income for taxation purposes; and the reforms to the stamp duty code and changes to capital gains tax as it applied to serviced zoned land.

The two subsequent reports of March 1999 and June 2000 contained more detailed proposals directed mainly at improving the supply side response. These latter recommendations were framed in the context of achieving credibility, clarity and certainty. In support of these criteria specific recommendations were made to: achieve higher residential densities; carry out key strategic infrastructure investments to overcome bottlenecks such as the northern fringe interceptor sewer; accelerate the process of securing required planning consents on significant sites in Dublin city and county through the use of strategic development zones (SDZs); improve the deployment of existing planning resources; increase the resources available to the planning system; impose fiscal penalties for non-realisation of potential of SDZs; propose revisions to the stamp duty regime; establish an anti-speculation property tax; establish measures to secure improvements in the quality and availability of rented accommodation; and to strengthen the institutional framework for securing a more effective housing response in the greater Dublin area.

I will turn now to the outcome. Rates of increase in prices of new and existing houses in Dublin and nationally slowed sharply from the middle of 1998. The peak rate of inflation in the new house market was 24.6% in the first quarter of 1998 countrywide and 33.8% in the first quarter of 1998 in Dublin. By the first quarter of 2000, these rates had halved to 12.9% and 16.2% respectively. In the existing house market the peak rate was 36.9% in the third quarter of 1998 country wide and 41.7% in the third quarter of 1998 in Dublin. These rates too more than halved to 17.4% and 20% respectively in the first quarter of 2000. At the same time, the annual rate of new house completions increased about 10%, to approximately 46,500 units, the highest annual rate of completions ever recorded to that time. However, in 2001 the measure to exclude interest deductibility was reversed. Thereafter prices re-accelerated, despite a supply response rising to almost 90,000 units annually, as speculative forces gathered increasing momentum.

I will move now to the development of the proposal to establish NAMA. At the heart of the banking crisis was a concern of capital markets with regard to the adequacy of banks' capital to meet future loan impairments and institutions’ capacity to obtain additional capital externally. Future impairments were of concern because for the previous decade Ireland had experienced rapid inflation in property values and lending to the property sector had become an increasingly important component in credit institutions’ lending. In addition, there was heightened international concern about the health of the financial sector.

Irish banks were facing an extremely unstable outlook in respect of international wholesale deposits, upon which they had become significantly dependent in the previous decade to fund expansion of their assets or lending. They were experiencing major withdrawals of these deposits, a shortening of the average duration of deposits and substantial recourse to the Central Bank for short-term liquidity support. This was not a sustainable trend. In addition, the initiatives taken by Government to that date were considered to be insufficient to achieve rates of capital adequacy that would encourage investors to hold and invest further equity in Irish credit institutions when prospective impairments were considered. As long as this remained the case, it could be expected that share values would remain depressed and deposit liabilities would be likely to experience continued attrition and foreshortening in duration. Such a prospect would hinder economic recovery, complicate further the required adjustment of the public finances and leave Ireland’s international credit rating subject to downward pressures and speculative attacks. Therefore, it was concluded that additional and far reaching measures needed to be undertaken as soon as possible to place the banking system on a sound footing.

Deterioration in the Government debt to GDP ratio was under way as the general Government deficit widened. A significant part of this deterioration arose from the effects of cyclical downturn. Moreover, discretionary budgetary adjustments to curtail the widening deficit would be partially undone by the deflationary impact of the discretionary measures themselves. To some degree, in the absence of international recovery and-or gains in competitiveness and productivity in Ireland, the domestic fiscal adjustment process had the characteristics of a vicious spiral comprising weakening economic activity leading to widening of the Government deficit and indebtedness leading to discretionary adjustments leading to further erosion of economic activity and so on.

The deterioration in Ireland’s credit terms associated with a worsening fiscal position was compounded by the additional contingent liabilities assumed by Government by virtue of the guarantee of the deposits of credit institutions from the previous September. Capital markets were uncertain how to value the additional liability of the Government on foot of the guarantee and the resulting confusion was causing Irish bond spreads to widen unfavourably. Against this backdrop, it was considered imperative that initiatives should be undertaken that would lead to stability in banks' deposits and term debt liabilities and eliminate the need for a renewal of the guarantee in place at the time. To achieve this required removing all doubts about capital adequacy of the credit institutions and their capacity to deal with prospective loan impairments.

There are a number of broad approaches, which are not mutually exclusive, to bank capital support schemes. These revolve around recapitalisation programmes involving stress testing against expected losses, asset guarantee schemes and asset management arrangements. The key features of recapitalisation programmes are future capital shortage is anticipated by testing adequacy of current capital in stress scenarios; the adequacy of capital to absorb losses is assessed; and the regulatory authority may then require more capital, which may be raised from the market, for example, by way of a rights issue or attraction of new shareholders, which may be either private or State. This approach needs to take account of implications of market conditions for cost of capital to bank, dilutive implications for existing shareholders and protection of State capital if the external shareholder is the Government. There have been many recapitalisation programmes put in place in the US and the EU in the current crisis, including in Ireland.

The key characteristics of the asset guaranteed or risks insured by the State approach are troubled assets remain on the balance sheet of the banking system; troubled assets are not subject to upfront mark-to-market write-downs; the bank usually is liable to a relatively small first loss tranche and the State covers elevated losses for a fee; equity capital is not affected as assets do not have to be sold at the current marked-down levels; no initial outlay is required from the State and a fee, premium or compensation arrangement is paid for the guarantee; compensation to the State in the form of convertible preferred shares or warrants is dilutive, of existing shareholders; and such schemes have been implemented at ING, Citigroup, Bank of America and RBS.

The key features of the asset management arrangements approach are troubled assets are transferred from the balance sheet of the banks at an agreed price; mandatory participation is required; the banks take the impairment loss to profit and loss account now; the bank is cleansed of troubled assets making valuation of the remaining part of the bank less complicated; the removal of impaired loans reduces the risk weighted assets of the bank and releases capital or reduces the shortfall in capital required; a discounted sale of assets may result in a significant reduction in the equity of the seller; and significant financing may be required from the State for the asset management company, impacting negatively on the fiscal position. Examples at the time included UBS and Securum-Nordbanken in the Swedish crisis of the 1990s.

Nationalisation was explicitly considered in the context of the report. Where a bank’s net worth has already been wiped out or would be by future impending losses or where Government are or will become dominant shareholders as a result of recapitalisation or other initiatives, nationalisation may be the most effective means of protecting the interests of all of the stakeholders – Government, equity and bondholders, depositors and the business franchise owned by the bank – and carrying out the required restructuring to enable the bank to stabilise its business in support of the wider economy in the future. For example, nationalisation could be used to facilitate mergers of operations and improve efficiencies of scale in accessing wholesale credit markets so as to bring about required strengthening of management or corporate governance. In effect where taxpayers are liable for guaranteeing the deposit liabilities of banks and also guaranteeing the bank against losses in the value of assets, in whole or substantial part, by any arrangement, such as those described above, nationalisation may be considered necessary to overcome issues of moral hazard. These situations are mostly likely to arise with respect to shareholders, who may be seen to be bailed out or gifted as a result of initiatives to support bank capital.

Another such concern may be the additional cost to the taxpayer in terms of deteriorations of the market's rating of sovereign debt instruments and the premium paid to bondholders in respect of this.

A number of nationalisations were made in the course of the current crisis in the UK, notably Northern Rock and Bradford & Bingley, and of course here in Ireland Anglo Irish Bank Corporation was nationalised in January 2009. A summary comparison of the general attributes of these approaches is contained in a table. I do not propose to go through the table because the committee has it.

The very features which make the asset guarantee approach intuitively attractive - no money up-front from Government and no write-down in banks’ balance sheet assets - contain also inherent fundamental weaknesses, namely, that a contingent liability is created in the balance sheet of the Exchequer. The situation would have significant parallels with the bank guarantee of the six credit institutions. It too was adopted on the basis that it involved no up-front outlay on the part of the Exchequer and on the basis that it would not be "called" and therefore the premium payments by banks would be a net receipt to the Exchequer. In the event, capital markets did not grapple well with the contingent liability created by the deposit guarantee. The tendency was to price Irish sovereign debt unfavourably, reflecting a view that more issuance of Government debt would be required. Indeed, an argument developed that if any part of the guarantee came to be called, in effect all would be called and that would lead to extreme problems for the Exchequer. The point of relevance here is that contingent liabilities are inherently uncertain in nature, are often evaluated in an ill-informed way with resulting errors and the potential for further adverse speculation against Ireland. As a result of the decision to guarantee the debt liabilities of Irish credit institutions the credit rating of sovereign, Ireland became inextricably bound up with the issue of Irish banks’ capital adequacy. A further guarantee approach, this time in respect of banks’ property-related loan assets, would create a further layer of uncertainty through the creation of another contingent liability on the Exchequer. This would further entwine the sovereign rating with Irish banks' capital adequacy problems without actually providing any clarity as to how capital adequacy would be achieved, other than through a calling of the contingent liability.

By contrast, the asset sales approach, while involving the recognition of "pain" at the outset contained the merit of certainty and clarity, provided of course the projection of the extent of impairment was accurate in the first place. In the particular circumstances prevailing it was considered that there was much to be said for recognising and crystallising prospective property-related loan losses explicitly, rather than allowing them to remain on banks’ balance sheets with a concomitant additional contingent liability on the Exchequer.

A feature of the guarantee approach is that assets remain on the balance sheets where they have been created. Another side to this is that they continue to be managed by the officers and executives of banks which created the problem assets in the first place. In the case where assets are complex financial instruments, such as many of the assets acquired by banks that were originated in the US and based on sub-prime borrowers, their valuation and resolution may best be undertaken in the banks which acquired them and which have the financial skills appropriate to this task. The nature of impaired property loan assets simply was not of this character. They were loans created and secured by property assets, development land, work in progress, completed but unsold residential stock and under-performing property investments, which are now worth significantly less than was envisaged by the loan. There is not a great deal banking skills can do to resolve this dilemma. Moreover, the property development companies involved in these transactions are almost entirely privately owned, championed by entrepreneurial characters and mostly without equity or recourse to equity markets, and in many cases do not have the depth of management skills to engage in the kind of portfolio sales and work-outs which ultimately are required to resolve the impairment issue.

Asset management companies, AMCs, offer prospects for avoiding many of the shortcomings associated with a continuation of the existing bank-property developer relationship. Potential advantages include: economies of scale in administering work-outs and in forming and selling portfolios of assets; benefits from the granting of special powers to the government agency to expedite loan resolution; and the interposing of a disinterested third party between bankers and clients, which might break connections that otherwise could impede efficient transfers of assets from powerful enterprises.

Sweden’s AMCs provide examples of some of these potential advantages, but other countries have found it difficult to realise them. First, government agents may lack the information and skills of private market participants. Second, government agencies do not operate in a vacuum. They, too, are creatures of the societies that create them, and government agents must negotiate, rather than dictate, solutions, just as private market participants must do. In negotiations with government agencies and private participants alike, the strength of one’s position depends on one’s “threat point”, the ability to credibly threaten adverse consequences for one’s bargaining opponent, if agreement is not reached.

Notwithstanding these shortcomings, it was considered that AMCs, by virtue of the potential advantages they contain, that I have noted, have the potential to bring about better economic resolution of the impaired loans of Irish property developers than relying on existing bank management and banker, developer relations, which brought about the problems in the first place.

The point I am about to make is one I made extensively at the time. It was never picked up and I considered at the time and still consider that it is one of the most important points. A further important consideration relates to the future financing requirement of impaired assets. Many of the impaired assets will be capable of achieving higher values if they can be worked out rather than disposed of. A key issue for successful work-out will be access to additional capital, equity and debt required for the work-out. It is extremely difficult to see how existing property developers will be able to access capital markets effectively for such equity and banks’ capacity to extend credit will be limited by the absence of collateral available from most of them. Potentially the amounts involved are large and a feature of Irish property developers is they are not publicly quoted companies and have not had a history of recourse to equity markets for their funding. I made the point when the crash occurred that the first thing that happened in the London market was that British Land, Land Securities and Hammerson, and some others, went straight to the market and raised capital to bolster their balance sheets because they were quoted companies. There was not a single one in Ireland then or now. Instead, they have relied on retained earnings for equity and bank lending for the balance. This shortcoming cannot be put right now and it represents a significant impediment looking forward to resolution of the impairment issue, at least cost. There has been a development in the past two years in the sense that there are now three publicly quoted real estate investment trusts, which is certainly a welcome development to absorb very fine commercial property assets from the resolution process.

An AMC, however, does have the potential to at least mitigate this issue in two respects. First, it has the potential to achieve scale and overview of developments and projects. Banks will be concerned about the security they hold and how that can be maximised and realised. In many instances more than one bank will be involved in the security and their individual interests may not correspond. An AMC would be able to achieve project oversight. Second, if properly structured and resourced with relevant property-related skills, such an entity would have the potential to attract long-term capital in a manner that individual development companies would not.

In conclusion, it appeared to me that the asset management approach had the potential to offer greater assistance in achieving resolution of the impairment issue upfront and maximising taxpayer returns, over the long term.