Putting money in the PIIGY Bank? – Bank of Ireland

I have not had much time to write recently for many other reasons. In any case, I have not had a huge amount to write about. The market chugs ever onward and apparently, Apple is the only fruit worth eating. While I should be pleased, as my equity investments have performed well, I must admit that I did not expect to have made such a return so quickly when I returned to the equity markets last year. Also, I am no where near full invested in equities, so asset allocation will have dampened my total portfolio return.

As a student of valuation and market cycles, I hope to invest at the point of maximum opportunity. One of those points was last August/September, but it is not my view that it was the ultimate valuation low in the present cycle. As I have elaborated on in the past, my central belief is that we are in a cyclical bull market in within a broader secular bar market. While stock picking is important, I believe that an appreciation of market cycles and asset allocation is more important.

As a advocate of the secular bear market/sideways market belief, I really should question as to whether I should be a shareholder of a bank. I am a shareholder in Bank of Ireland, having purchased share last October.

Many folks will have seen and read the piece on profit margin sustainability this week from James Montier at GMO. The research is focused on how high US profit margins are relative to history, and that if margins and valuation begin to mean revert then that has implications for total returns from the stock market. My ideal investments all focus on mean reversion. I am principally interested in investing in the equities of companies that have room to grow valuation, revenue and margins. Bank of Ireland to my mind has an opportunity over the next five to ten years to grow profitability and valuation, however in a balance sheet recession banks and wider society are forced to delever. If this is the case, maybe the shrinkage from deleveraging will offset the gains from margin expansion, and I will be left with a pretty poor investment? That is why I am forcing myself to re-evaluate this particular investment.

Bank of Ireland plc 

The purpose of this post is to question the sanity of such an investment in the current market, and to cross check how BKIR is doing versus similar banks in the UK, Scandinavia and continental Europe. The core of this analysis is as follows:

  • How is Bank of Ireland capitalised?
  • How is Bank of Ireland funded?
  • How likely is it that Bank of Ireland will earn a 10% Return on Equity?
  • What does the present valuation imply?
  • Should I own any banks?

As a starting point, before going onto risk weights and Tier 1 ratios I believe that it is important to look at a simple equity/assets ratio. I do this because I believe that the concept of risk weighting assets with weightings less than 100% is fundamentally flawed and is in part to blame for the banking crises of the past few years. To my mind the BIS risk weighting methodology allowed banks (and shareholders) to believe that balance sheet growth was significantly smaller than was actually the case. The issue is important, because although assets can be risk weighted, liabilities (or funding) is not. So, whether a€100  loan on a banks balance sheet has a weighting of 50%, 75% or 100%, it still requires €100 of liabilities to fund this asset. So to look at a balance sheet in any other manner is to my mind delusional.
The table below presents the simple equity/assets ratio of many of the largest banks operating retail and commercial banking in Europe.
Bank Equity Total Assets Equity/Assets
Bank of Ireland 10200 155000 6.6%
RBS 74819 1506867 5.0%
Lloyds Banking Group 45920 970546 4.7%
Societe Generale 47067 1181372 4.0%
BNP Paribas 75370 1965283 3.8%
Credit Agricole 42797 1723608 2.5%
KBC 16255 285382 5.7%
Banco Santander 76414 1251526 6.1%
BBVA 38165 597688 6.4%
Baco Popolar Espanol 8282 130926 6.3%
Banco Sabadell 5887 94550 6.2%
Commerzbank 24104 661763 3.6%
BCP 3889 93545 4.2%
BES 5604 80237 7.0%
Intesa San Paolo 47040 592181 7.9%
Unicredit 64224 929488 6.9%
UBI Banca 10979 130559 8.4%
Monte de Paschi di Sienna 17303 244279 7.1%
Banca Popolare di Milano 3660 51927 7.0%
Danske Bank 125795 3424403 3.7%
DNB NOR 117815 2126098 5.5%
Svenska Handelsbanken 94524 2454366 3.9%
Nordea 26034 716204 3.6%
Erste Bank 13585 205938 6.6%
Average 5.5%
What one will notice is that high equity/assets ratios have been no saviour to the equity investor in terms of share price performance. Some of the best performers (the Scandinavian banks for example) have the lowest equity/asset ratios in the sample group. Poor performers such as the Irish and Italian banks have reasonably high equity/asset ratios. It is not enough that the bank has sufficient capital relative to its assets, those assets must be sound and have a low probability of impairing the capital base. Given the property loan portfolios of the Irish banks and the Italian Government bond portfolios of the Italian banks, the markets has taken a view as to the underlying value of those assets.

What I can say, is that relative to its peer group, Bank of Ireland appears well capitalised. This should not be a surprise given that the bank has raised equity from shareholders and liability management exercises many times over the past three years. What I need to look at is, how does the bank look at in an absolute sense of capital adequacy.

For a bank, capital exists (in part) as a buffer to loan losses. Since the 2008 financial year, Bank of Ireland has realised impairments and loan losses of €12.6 billion on an asset base of €194billion and a loan portfolio of €134 billion at the height of the market.

To put that in context, almost 9.5% of the peak balance sheet loan book has been impaired thus far.

Problem Banks Accumulated LLPs 2008 Loan Portfolio Writedown
Bank of Ireland 12631 134000 9.4%
Lloyds 48588 677958 7.2%
RBS 39936 874722 4.6%
Danske Bank 64767 1127142 5.7%
Banco Populare Espanol 5115 91705 5.6%
Banco Sabadell 3398 65629 5.2%
So it would seem that on an absolute and relative basis, that Bank or Ireland has more capital than its peer group average, and that its most recent destruction of capital by way of loan losses have been similarly large. Will this prove to be sufficient given the current state of the loan portfolio and impaired loans?
Current loan portfolio and impaired loans:
During the summer of 2011, BKIR began a large rights issue that would ultimately raise sufficient capital to meet the requirements set out by the EU and ECB in terms of banking stress tests. As part of the rights issue prospectus, the Bank published details of the adverse stress test scenario carried out by Blackrock.
Source: Bank of Ireland Rights Issue Prospectus, 2011.
As part of the stress test, Blackrock Solutions  estimated that Bank of Ireland would require additional provisions of €6.6bn to cover a worst case scenario in terms of loan losses between 2011 & 2013. Since then, Bank of Ireland has reported its 2011 full year figures, where they recorded an impairment charge of €1.96 billion. While this would put the bank inside the run rate of the Blackrock stress test, the recent results did show a deterioration in credit quality across several loan books.
I have used the most recent accounts to come up with my own stress test.
Irish Mortgage Portfolio
The model assumes that
  • house prices fall 20% in the next three years.
  • provisions required are calculated after stock of existing provisions and collateral (after price falls)
Loan Book NPDOI Past Due Impaired Total
Irish Residential Mortgage Book 18458 1823 582 20863
Irish Buy to Let Mortgage Book 5398 828 765 6991
Total 23856 2651 1347 27854
Total Provisions 1026
Provision Coverage 0.0% 0.0% 76.2%
Avg LTV 100% 125% 125%
House Price Fall (2011-13) 20% 20% 20%
% of portfolio to go into arrears. 20% 50% 100%
Total Bad Loans 4771 1326 1347
Existing Provisions 1026
New Loan Loss 4771 1326 321
Collateral after 20% price fall 3976 914 0
Provisions Required 795 411 321 1528
*NPDOI = Neither past due or impaired
This model assumes that the Irish mortgage portfolio will generate a futher €1.5bn of provisions, which I assume will be applied between 2012 & 2014.
Property and Construction Loan Portfolio

Property & Construction 11915 1042 7623 20580
Provisions 3205
Assume Default Rate 20% 50% 100%
Loan Losses 2383 521 7623
Collateral  (Guess work) 1787 313 3049
Provisions Reqd 596 208 1369 2173
Consumer Loan Portfolio
Consumer Loans NPDOI Past Due Impaired Total
Other Consumer Lending 2818 158 338 3314
Provisions 278
Assume Default Rate 10% 40% 100%
Loan Losses 282 63 338
Provisions Reqd 282 63 60 405
UK Mortgage Portfolio
UK Mortgage Portfolio NPDOI Past Due Impaired Total
Standard Mortgages (£m) 10407.0 487.0 11.0 10905.0
Buy To Let Mortgages (£m) 9232.0 511.0 83.0 9826.0
Self Certified Mortgages (£m) 3449.0 563.0 12.0 4024.0
Total (£m) 23088.0 1561.0 106.0 24755.0
Total Provisions
Provision Coverage 0.0% 0.0% 0.0%
Avg LTV 87% 87% 87%
House Price Fall (2011-13) 15% 15% 15%
% of portfolio to go into arrears. 10% 35% 100%
Total Bad Loans 2309 546 106
Existing Provisions 129
New Loan Loss 2309 546 -23
Collateral after 15% further price fall 2264 536 0
Provisions Required 45 11 -23 33
Provision Reqd (€m) 39
The UK Portfolio will according to my model generate almost €39m in further provisions.
Commercial Loan Portfolio

Other Commercial Loans 22245 430 4043 26718
Provisions 1723
Assume Default Rate 5% 30% 60%
Loan Losses 1112 129 2426
Collateral  (Guess work)
Provisions Reqd 1112 129 703 1944
The total level of loan losses over the next three years from my stress test would be €6.09billion.
On the basis of the existing level of pre-provision profitability (excluding exceptional gains) of €500m per annum, it would seem reasonable to assume that €1.5 billion of total pre provision profitability could be generated over the next three years.
This would mean that €4.5 billion of capital would be required to cover loan losses. While I think that this would a pretty extreme scenario, it is not unlikely given any likelihood of a renewed recession.
This would leave the Bank with less than €6 billion of equity, in other words the bank may lose 45% of its equity if loan losses continue to deteriorate from here.
The capital ratios at that time would be dictated by the amount of assets on the balance sheet at that stage, but at a level of €5.5 billion of equity Bank of Ireland would have an equity/assets ratio of 3.7%. This would require a further equity raising in my opinion.
I should remind readers that there is a €1 billion of contingent capital that would convert to straight equity if core tier 1 ratio falls below 8.25% (42% below the current level).
What this all means is that Bank of Ireland may have sufficient equity, but that a further capital raising exercise should not be ruled out. The present core tier 1 ratio at Bank of Ireland is 14.3%. This is excessive by historical norms, and it exists to cover the likelihood of increased loan losses in the foreseeable future.
It is my view that regulators, policy makers and the media are overly obsessed with the levels of capitalisation at many banks, and are not nearly as focused on funding.
At a very simplistic level, risk weighting of assets contributed in part to the rampant growth of the asset side of balance sheets in many banks in many countries.
This growth had been funded through the growth in the shadow banking system (effectively non deposit funding). With the collapse of the banking debt market, this funding is now by and large being met through central bank funding. So, private shadow banking has been replaced by a public funding market.
Unless banks can get to a balance sheet that can be funded through stable funding, then the future in terms of credit provision looks bleak. This is particularly an issue for the Irish banks.
The bank of Ireland Balance sheet is funded as follows:
Balance Sheet & Funding 2011 2010
Customer Loans 102 114
Customer Deposits 71 65
Loan to Deposit Ratio 144% 175%
Wholesale Funding 51 70
 – Monetary Authority Funding 23 33
 – Covered Bonds 6 7
 – Securitisations 4 5
 – Private Market Repo 7 8
 – Senior Debt 9 13
 – Bank Deposits 2 3
 – Comm Paper 1
 o/w > 1 year 27 22
 o/w < 1 year 24 48
Central Bank Funding 22 31
There are several issues in terms of funding for Bank of Ireland:
  • The company is covered under the Irish state Eligible Liabilities Guarantee (ELG) scheme.
The ELG is the Irish state guarantee of certain debt securities as well as deposits in excess of €100,000. It is a very expensive insurance scheme, with the cost during 2011 up 31% to €449 million. That is equal to 100% of pre-provision operating profit during the year.
The sooner the bank returns in a meaningful way to the unguaranteed funding market the better. Initially they will be required to pay up for funding (corporate deposits or debt). It will be worth it in my view. In terms of milestones, unguaranteed funding will be the most important issue for me going forward. During 2011, a total of €44 billion of liabillities (funding) was covered by the guarantee scheme at a cost in excess of 1%. This compares with €69 billion in 2010, but at a cost of 50bps.
Of note during 2011, was the issue of €4.2billion of unsecured collateralised funding (of the UK mortgage assets) at a cost of 250bps over 3month Euribor. I am assuming that these assets we over-collateralised to some degree. However it is progress ad I will be looking for further evidence of access to the unguaranteed funding market during 2012.
The following chart illustrates the pressure that the present funding arrangements are having on profitability.
Margins on lending (the asset margin) is expanding, but all of that is being eroded by a significant increase in the cost of funds.
I should mention that some of the margin erosion over the past four years has been down to prevailing low interest rate environment. Many banks have deposits that pay little or no interest. The impact of this is greatest when interest rates are high (it is known as the free funds effect). As interest rates fall, the absolute amounte charged for lending falls, and the advantage of free funds falls in tandem.
  • Under the Prudential Liquidity Assessment Review (PLAR) the bank has to deleverage its balance sheet within a prescribed timeline in order to reduce reliance on short term funding and central bank liquidity support. The bank has successfully competed 86% the 2011-13 deleveraging mandated under the PCAR.
While they are to be commended for this. However the balance sheet is still significantly reliant on central bank/monetary authority financing.
Monetary Authority funding amounts to €23billion for 2011 (down from €33b in 2010). This amounts to
The principal issue around deleveraging, is that it is difficult to take advantage of opportunities as you delever. In the case of the bank, we know not what type of profitable lending opportunities that must by turned away due to the focus on slimming. I have said before that you cannot shrink to greatness. 
While the cost of monetary authority funding is highly favourable, it does have implications implications  for the cost of other financing raised by a bank, as it pledges assets as collateral against the borrowing. Furthermore, it is a sign of weakness and as such may act to dissuade potential funders from doing business with the bank. There are advantages, principally that banks funding is now less questionable than it was before the latest ECB Long Term Refinancing Operation (LTRO).
Funding at BKIR has improved, in that funding is now more secure than it has been in a number of years due to:
  1. Lower LDR,
  2. Lower quantum of funds being guaranteed (albeit at a higher cost),
  3. €7.5bn of LTRO plus an equivalent amount of NAMA bonds,
  4. Some (small) return to the unguaranteed funding markets.
What will be interesting in 2012 is that almost €23 billion of funding is within one year. I will be watching for how this funding is replaced and at what cost.
Return on Equity
I have run a simulation in order to help me ascertain, what type of assumptions are needed in order for the bank to reach a 10% Return on Equity. This is not an estimate or prediction of future returns or profitability – that would be an exercise in futility. In any case, future returns are more than likely to be off a lower level of capitalisation, as loan losses in the next few years eat into the current equity capitalisation.
Bank of Ireland 8% RoE 10% RoE 12% RoE
Net Interest Income 2665 3165 3665
Net Interest margin 1.97% 2.34% 2.71%
Net Insurance Premium 929 929 929
Net Fee Income 500 500 500
Net Trading Income 0 0 0
Life Assurance Investment Income 247 247 247
Other Operating Income 196 196 196
Non Interest Income 1872 1872 1872
Total Operating Income 4537 5037 5537
Insurance Claims and Liabillities -1089 -1089 -1089
Total Income 3448.33 3948.33 4448.33
Operating Expenses -1793.1 -2053.1 -2313.1
Cost Income Ratio 52.0% 52.0% 52.0%
Pre Provision Operating Profit 1655.2 1895.2 2135.2
Impairment Charges -695.2 -695.2 -695.2
Operating Profit/(Loss)
Share of JVs after Tax
Loss on Disposal
Profit/(Loss) before tax 960.0 1200.0 1440.0
Taxation -144.0 -180.0 -216.0
Net Profit/(Loss) 816 1020 1224
Total Assets 135000 135000 135000
Loans to Customers 99314 99314 99314
Shareholders Equity 10200 10200 10200
RoE 8.00% 10.00% 12.00%
RoA 0.60% 0.76% 0.91%
The simulation works as follows:
  • A level of returns (8%, 10% and 12%) are specified,
  • The equity base is assumed to be €10.2 billion and the asset base is assumed to be €135 billion.
  • The net income is then calculated ( Net Income = RoE * Equity)
  • Pre Tax income is then calculated assuming a tax rate of 15%,
  • A Loan Loss Provision rate of 70bps is assumed (that is equivalent to the 20 year average, with the range being -3bps to 4.0%),
  • This figure is the grossed up to derive Pre Provision Operating Profit,
  • A cost income ratio of 52% is then assumed,
  • From this the level of Total Income is calculated,
  • The level of non interest income is calculated as average over the past 5 years,
  • Then the balancing figure is the Net Interest Income,
  • From the Net Interest Income, the Net Interest Margin is calculated.
So in order to get a 10% RoE from the assumptions that I have used, one would need a 2.34% net interest margin. While this is 88% greater than the present level of net interest margin, I think that it is worth recalling that NIM used to be in excess of 2.5% for much of the period between 1991 and 2007. In fact margins declined over the 17 year period to 2007 as the shadow banking market allowed access to cheap funding and leverage. I have a suspicion that the next cycle will see an expansion of banking margins as rates eventually rise, risk pricing returns to lending and a more prudent approach to funding is mandated.
So margins required to meet a 10% RoE target may seem excessive by today’s level, but it is entirely within the bounds of probability on the basis of 20 years of net interest margin history. The reality will be different from my simulation – I have run it simply to sanity check whether these guys can achieve a 10% RoE again. I believe that they can, but it is unlikely to occur within the next 3 years at a minimum.
Valuation (Share Price €0.13)
Valuation is problematic.  The present book of €0.34/share also includes over €1.5 billion so it needs to be amended, to €0.28. More importantly though is that while the bank is trading at a large discount to book value – this discount is very likely to be eroded over the next three years as loan losses eat into profitability (and the gains from bind buybacks are not repeatable).
Share Price 0.13
P/E 97.9
P/B 0.46
Dividend Yield 0%
P/PreProvOpProfit 1.7
G&D PE 6.9
Mkt Cap/Deposits 5.6%
I notice that Davy Stockbrokers, are forecasting a €0.22 tangible NAV by the end of 2013. I am going to assume that my stress test losses of almost €4.5billion over the next three years actually come to pass. This could push the book value/share down €0.16.
If this comes to pass, then I would be prepared to purchase more shares in bank of Ireland at a P/B of 0.5x or €0.08. I believe that at that level, my margin of safety is sufficient from both a risk and reward stance. To buy any higher means that I leave too little upside relative to the ongoing risks that go with investing in banks during a balance sheet recession.
As usual, comments and critiques welcome.

8 Responses to “Putting money in the PIIGY Bank? – Bank of Ireland”

  1. 1 rijk March 25, 2012 at 9:24 am

    excellent analysis!

    seeing that you purchased at 0.117 and currently trades around 0.13, and your statement that you are a buyer at 0.08 because of continued loan losses, are you selling?


    • 2 jmcelligott March 25, 2012 at 7:07 pm

      No, I wont sell for the minute. Though I readily admit that my initial purchase was misjudged in terms of entry price. There is a lot to admire about BKIR in terms of fundamental improvement over recent history. In saying that, it is still very highly speculative.

  2. 3 Loglorry March 26, 2012 at 8:47 am

    Great article thanks for posting it up. It does highlight how precarious a situation a lot of the Euro banks are still in.

    I think the big take-away here though has to be that they need time to heal to work off the impairments and central banks will as much as possible pump in liquidity to make sure they don’t have any funding problems. Simply put the soverign backstop will keep them afloat. To me therefore the best place to invest has to be in the bank sub-debt and preference share markets where one can shelter from dilution to some extent.

  3. 4 HARDROAD April 1, 2012 at 8:13 am

    like you i bought shares last year at 10.4 cent, with a view to holding on to them for 5 years. If bank of ireland returned to profit, what would your estimate be on the share price.

    • 5 jmcelligott April 3, 2012 at 8:06 am

      Hi Hardroad. I think it depends on the profitability and what the book value is when that eventually happens. I have satisfied myself that an RoE of 10% is achievable at higher net interest margins. If that passes and the final book value per share is 15.5 cents then I think that the stock is worth a minimum of 15.5.
      COuld it be worth more. Yes it could.
      On a very very simplified basis, if the BV was 0.155 and the RoE was 10%, then one could make a case that it would trade at 1.55xBV, or 24 cents.
      At this stage one would need to realise that the book value will alo begin growing each year.

      So for the moment, I am saying that 15.5-24cents seems plausible. But this is very much dependent on profitability and adequate returns, both of which seem like a long way from here.

  4. 6 feuerball April 22, 2012 at 5:40 pm

    The problem with all the UK and most other European banks is that the net interest margin is only ~2%. If you build a model with 50% efficiency ratio and some loan loss provision, you end up with less than 1% ROA even in good times. I don’t see how a decent ROE can be achieved, now that 30:1 leverage ratios are out of question. That is why I am passing on most European banks

  5. 7 Michael MacNicholas April 26, 2012 at 9:02 am

    Isnt there a concern that the mortgage book losses could be very much higher again than the Blackrock adverse stress scenario, and that your additional loss estimate, because the ability of the Irish banks to realise any value from the security unpinning defaulting mortgages is completely circumscribed by political considerations? That is to say: there will be no repossessions in Ireland

  1. 1 Market Musings 25/3/2012 « Philip O'Sullivan's Market Musings Trackback on March 25, 2012 at 3:19 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


John McElligott


Enter your email address to follow this blog and receive notifications of new posts by email.

Join 104 other followers

%d bloggers like this: