03 June 2008

RBC CM Upgrades Scotiabank and Downgrades CIBC & TD Bank

  
RBC Capital Markets, 3 June 2008

Remain cautious on Canadian bank sector; changing ratings on three of the five banks

We maintain our cautious view on Canadian bank stocks, reflecting our expectations for continued pressure on profitability, the potential for further negative earnings revisions and valuations that are not overly cheap on a historical basis considering the challenges we think the banks will face. Projected returns to our 12-month target prices are in the (4)% to 1% range.

We believe that the stocks of Canadian lifecos should outperform those of the Canadian banks, as we believe that the macro environment, while negative, should not impact lifeco earnings as much and is better reflected in valuations. Projected returns to our target prices are in the 9%-18% range for lifecos.

We believe that the drivers of stock prices over the next 6 months may not be the same as those in 2009. Over the six months we believe that the following three items will be most important for banks:

• Exposure to U.S. lending. Credit quality is deteriorating very rapidly in the U.S. and the Canadian banks with U.S. presences cannot escape from this trend. Based on known exposures and the increase in impaired loan formations seen in recently reported results, we believe that Bank of Montreal is most likely to suffer from deteriorating credit quality in the U.S., followed by TD Bank. National Bank, Scotiabank and CIBC’s exposures to deteriorating credit quality appear less concerning in the near term.

• Strength and performance in retail franchises. Q2/08 results showed continued underperformance at CIBC and National Bank, and, to a lesser extent, Bank of Montreal. Scotiabank had the strongest combination of revenue and net income growth, with TD Bank following. TD’s operating leverage will be less than usual in 2008, in our view, given rapid expense growth in late 2007/early 2008, which is more likely to pay dividends in 2009.

• Exposures to financial guarantors and CDOs of RMBS. We do not think that the bad news on CDOs of RMBS and financial guarantors is necessarily over, and are expecting further writedowns at CIBC ($1 billion). We do not believe that capital markets writeoffs will be as large as in prior quarters for the most affected banks, given improvements seen in many areas of credit markets. Off-balance sheet exposures remain a risk, however, as deteriorating economic conditions will likely lead to declining values in assets held in those conduits, which may ultimately lead to losses for the banks.

For 2009, we believe that the following factors will be important drivers of stock prices:

• Exposure to deteriorating credit quality that spread beyond exposures to the U.S. and impact business lending as well. We expect higher commercial and small business loan losses in Canadian manufacturing, forestry and agriculture sectors for the Canadian banks. Exports account for almost a third of Canada's GDP and about three quarters of exports head to the U.S. so Canadian businesses, particularly manufacturers in central Canada, are likely to feel the impact of lower demand from U.S. consumers and the strong Canadian dollar. Sharply rising energy prices are also negative for many businesses. We believe that National Bank and Scotiabank – two banks that look clean from a credit perspective near term, are likely to not look as clean relative to peers in 2009 given their strength in business lending.

• Strength and performance in domestic retail divisions will, as usual, be a key determinant of relative valuation multiples.

o It is difficult to assume that TD Bank will outperform its peers forever but we believe that betting on TD has a good chance of succeeding as (1) it has a very large customer base to which it can cross-sell products and services, (2) its revenue growth should benefit from sales initiatives and investments made in the past few years, and (3) growth in expenses could be moderated without a significant hit to market share for some time, in our view.

o Scotiabank is coming off a period of rapid asset growth and market share gains, and could probably afford to “milk” those gains at the benefit of the bottom line for some time, if it chose to.

o Of the other three banks, Bank of Montreal and National Bank are both putting a lot of time and effort into centralizing more functions, adding front life staff and improving customer service overall, but they are clearly playing catch up to the leading banks. CIBC’s focus on expenses probably limits its capacity to grow revenues as rapidly as the leading banks in 2009. We also cannot help but think about top management focus in the last 6-12 months, which was probably dealing with the serious issues they encountered in their capital markets businesses, rather than on domestic retail banking.

• Exposure to a rebound in capital markets businesses. This item is hard to quantify but it disadvantages CIBC and Bank of Montreal, in our view. Both firms have been hit by writeoffs that were very large in relation to their income base and we believe that the de-risking efforts that are underway will have a negative impact on risk appetite and net income growth potential on a rebound. The other banks have not been as affected by writedowns and we do not expect a meaningful shift in risk appetite.

• Capital strength will likely be valued more than in a normal environment. Well-capitalized banks (1) are better positioned to withstand surprises and rising credit losses; (2) can take advantage of organic and acquisition opportunities that present themselves as a result of market dislocations; and (3) will be under less pressure to strengthen their capital ratios. We believe that Scotiabank is best capitalized and TD least so when considering the combination of Tier 1 capital, tangible equity ratios, balance sheet ratios and risks to expected profitability (i.e. CIBC has the highest Tier 1 ratio but writedowns related to CDOs of RMBS and financial guarantors remain a risk).

• Larger retail deposit bases may help banks mitigate the negative impact of higher wholesale funding costs on margins. TD Bank stands out as having more of its loan portfolio funded with retail assets, while on the other side, Scotiabank’s domestic retail margins are most dependent on wholesale funding.

Upgrading rating on BNS; lowering ratings TD and CIBC

Our two favourite banks (TD Bank and Scotiabank) are now rated Sector Perform, as we rate them in the context of our financial services universe, which includes life insurance companies, P&C insurance companies and asset managers. In terms of large capitalization stocks (i.e. banks and lifecos), we prefer lifecos, as we discuss in an upcoming section. The relatively narrow distribution of ratings also reflects our view that the differentiation between the top and bottom bank stocks will be lower in the next year than it was in the past year.

We are upgrading our investment rating on Scotiabank from Underperform to Sector Perform, to reflect our expectations for benign credit deterioration near term, strong domestic retail momentum coming out of Q2/08 results and continuing strong asset growth in international banking, which are offsetting rising loan losses in Mexico, and the negative impact of currency. We continue to believe that deterioration in business lending and lower recoveries will lead to a material increase in loan losses in 2009 but to reduce our exposure to the stock today on that basis is early in our mind, especially since the retail division is performing well. Our 12-month target price of $49 (up from $46) is based on a P/BV multiple of 2.3x, versus the current 2.7x multiple, and it implies a P/E on 2009E EPS of 11.5x, whereas the stock currently trades at 12.1x 2008E EPS. We raised our 12-month target price by $3 per share to reflect better than expected retail momentum, and greater comfort over credit quality near term.

We are lowering our investment rating on TD Bank from Outperform to Sector Perform. We believe that TD Bank faces enough near term headwinds to reduce our rating including (1) recent expense initiatives in domestic retail banking are hurting operating leverage (although it is still positive), (2) loan losses in U.S. banking are likely to rise although we are not as nervous as we are with the other banks with U.S. banking platforms, and (3) two key elements of TD’s capital markets revenues are likely to remain muted near term – equity securities gains are likely to be low, and the bank’s basis trade is going the wrong way (with the rally credit derivatives outpacing the rally in cash assets, as seen in Exhibit 8). Outside of capital strength, we like TD in 2009, so our caution relative to peers is a matter of timing. We believe the bank has a stronger deposit base, it does not have large exposure to business lending (which we believe will cause banks credit headaches in 2009), its domestic retail platform should benefit from recently extended opening hours (via greater revenue growth or at least a slowdown in expense growth) and profitability should benefit from the integration of Commerce Bancorp. Our 12-month target price of $69 is based on a P/BV multiple of 1.7x, versus the current 1.9x multiple, and it implies a P/E on 2009E EPS of 10.9x, whereas the stock currently trades at 12.5x 2008E EPS (the decline in multiples appears high but 2008E earnings do not include earnings from Commerce Bancorp in H1/08 and little in synergistic benefits from that acquisition in H2/08 so true “earnings power P/E” is lower than 12.6x).

We are lowering our investment rating on CIBC from Sector Perform to Underperform. CIBC’s Q2/08 results highlighted weaker than average growth trends in retail banking, which we expect will continue, and the strategic review of the wholesale division is likely to lead to a decline in the earnings capacity of the division as we believe the focus will be on risk reduction. CIBC has the highest Tier 1 ratio of the six Canadian banks, but we expect writeoffs related to CDOs of RMBS as well as exposures to financial guarantors will continue to cause the ratio to fall next quarter. These concerns outweigh our relative comfort on deteriorating loan losses in CIBC’s loan book. Our 12-month target price of $64 (down from $67) is based on a P/BV multiple of 2.0x, versus the current 2.4x multiple, and it implies a P/E on 2009E EPS of 8.6x, whereas the stock currently trades at 9.5x 2008E EPS. We lowered our 12-month target price by $3 per share to reflect our continued concerns over financial guarantors and its potential impact on book value.

We maintain our Underperform rating on National Bank’s shares, but have removed the Above Average Risk qualifier (changed to Average Risk like its peers) given the progress made in the ABCP restructuring plans and lower credit risk spreads compared to the March peaks (seen in Exhibit 10), which has positive implications for the value of the ABCP. ABCP-risk is down in our view, but risk remains as the restructuring is not yet complete, and there has been no external validation of the value of the restructured assets since no notes have traded. In the near term, National Bank’s shares may benefit from limited exposure to structured finance holdings and little in the way of U.S. operations compared to peers, which is one of the reasons why we are relatively comfortable with National Bank’s exposure to credit deterioration near term. However, the retail division’s retail performance remains weak compared to peers and, looking out to 2009, we are more concerned about credit for National Bank given its large business loan book in central Canada. The bank’s greater than average reliance on wholesale earnings is also likely to be a continued drag on the bank’s multiple relative to peers. Our 12-month target price of $52 (up from $48) is based on a P/BV multiple of 1.6x, versus the current 1.9x multiple, and it implies a P/E on 2009E EPS of 9.0x, whereas the stock currently trades at 9.6x 2008E EPS. We raised our 12-month target price by $4 per share to reflect greater comfort over credit quality near term, as well as declining risk related to the ABCP restructuring, in our view.

We maintain our Underperform rating on Bank of Montreal’s shares. We believe that BMO's share price is likely to lag its peers' given (1) our outlook for greater deterioration in credit quality near term; (2) retail banking results that are likely to continue lag the leading banks on a combination of revenue and bottom line growth (although BMO delivered better results than National Bank and CIBC in Q2/08); (3) greater concerns over the sustainability of wholesale earnings than most peers; and (4) continued overhang from off-balance sheet exposures. Our 12-month target price of $44 is based on a P/BV multiple of 1.4x, versus the current 1.6x multiple, and it implies a P/E on 2009E EPS of 8.5x, whereas the stock currently trades at 9.9x 2008E EPS.

We remain cautious on the bank stocks

We maintain our cautious view on bank stocks, reflecting our expectations for continued pressure on profitability, the potential for further negative earnings revisions and valuations that are not overly cheap on a historical basis. Projected returns to our 12-month target prices are in the (4%) to 1% range.

We expect the pressure on earnings growth to come from rising credit losses, slower growth in wealth management businesses, slowing loan growth in retail lending, and deleveraging. The Canadian banks have high Tier 1 ratios but we believe that the banks will review capital markets areas that have not historically attracted a lot of regulatory capital but led to losses in recent quarters (i.e. certain trading businesses, structured finance businesses as well as off-balance sheet exposures). We believe that banks will allocate more capital to those areas, or they will reduce the size of those businesses.

Our expectations for a decline in profitability are not unique on the street but we remain concerned that our expectations for profitability (which are already lower than the street’s for 2009) could still be too high given the current headwinds, which include credit quality, wealth management revenue growth, capital markets writeoffs, wholesale revenue visibility, wholesale funding rates, and potentially slowing personal loan growth. The impact of a slowing U.S. economy and rising food and energy price inflation is likely to be felt on many economies outside of the U.S. as well. We had a positive view on bank margins given their ability to reprice loans at this stage of the cycle, as well as the reemergence of a steep yield curve, but recent pricing actions in mortgages indicate the market is still competitive (Exhibit 7), and banks have greater dependence on wholesale funding than in prior cycles, which reprices more quickly than retail deposits. The financial guarantee industry also remains on unsettled ground, in our view.

It is tempting to argue that the worst is over and P/B multiples should increase given the healthier tone in capital markets, but if credit quality indeed deteriorates, loan growth slows and banks are required to hold more capital, there could still be pressure on profitability and on P/B multiples. It was only six years ago that weak credit and shaky equity markets led to bank P/B multiples dropping to 1.65x book value, compared to an average of 2.1x today. Forward P/E multiples declined to a low of 9.9x in that period, compared to an average of 10.4x today. (Exhibit 3 and 4). The banks are not expensive on a forward P/E basis and dividend yield basis, but we believe that in times of limited visibility on earnings, a price to book approach to valuations is more appropriate.

We prefer the stocks of lifecos

We prefer the stocks of lifecos versus those of banks for the following reasons:

• Currency conversions should become more accommodating for lifecos than in recent quarters, if the Canadian dollar stays at current levels (Exhibit 5). Canadian lifecos generate more of their income outside of Canada than banks, and lifeco earnings growth has been more negatively impacted by the rising Canadian dollar. YoY increases in the dollar have been in a fairly tight range since mid-November (between $0.97 and $1.03), so the currency impact should decline from recent quarters.

• Credit deterioration is negative for both banks and lifecos, but more so for banks. Their business models are different and banks tend to have more risk in their loan portfolios than lifecos do in their bond portfolios. Lifeco provision for credit loss rates as a result have been lower compared to banks, and Canadian lifeco ROEs were more stable than bank ROEs in the last credit cycle.

• Rising inflation pressures could put an end to declining long-term rates. Low and declining long term rates are bad for life insurance companies, given the long-dated nature of their liabilities. 10-year bond yields are up from their lows in mid- March both in Canada (from 3.4% to 3.7% today) and the U.S. (from 3.3% to 4.1%). (Exhibit 5)

• We have more comfort in our earnings/ROE forecasts for lifecos than banks. Areas of concern for banks include credit, capital markets revenues, and the potential for declining balance sheet leverage.

• Valuations have come off for lifecos as well as for banks. Lifeco forward P/E multiples declined from 13.4x a year ago to 11.5x today, while the banks’ forward P/E was 12.7x a year ago and is 10.7x today (bank P/B was 2.8x down to 2.2x today).

Overview of Q2/08 results

Q2/08 results were weaker than we and the street had expected.

• The industry’s core cash EPS declined at a median of 4% versus Q2/07 (Exhibit 11). On a GAAP basis, the EPS decline was 18%. Only National Bank’s core EPS came in ahead of our estimates.

• GAAP EPS were lower than core EPS for four banks. CIBC’s GAAP loss per share was much larger than expected after it took another $2.6 billion in capital markets related writeoffs and unusual items. Several banks reported capital markets related gains that partially offset (or in the case of Bank of Montreal, more than offset) the writedowns. (Exhibit 12)

• Core wholesale earnings trended down (29% YoY and 27% QoQ) as the weak operating environment reduced M&A, underwriting and trading results in most capital markets divisions. We should note that some of the volatility that caused trading losses also undoubtedly led to the establishment of trading positions that generated gains that would not have occurred in a normal environment.

• Credit quality continued to deteriorate, with Bank of Montreal being the negative outlier among the banks. Specific provisions for credit losses have grown to their highest level since 2003 but are still below historical averages. New formations of impaired loans (a good indicator of future provisions, in our view), declined slightly from Q1/08, but were at levels not otherwise seen since Q4/02.

• Domestic retail revenue growth slowed to just 2% year-over-year and earnings growth was 8% despite a relatively stronger economy than in the U.S. CIBC, TD and National Bank stood out on the weaker side while Scotiabank fared the best. The difficult equity markets contributed to a slowdown in wealth management revenue, which declined 6% from Q2/07.

• Capital ratios declined QoQ but remain high by global standards (Scotiabank’s Tier 1 ratio rose because the bank benefited from the removal of a transitional adjustment related to the implementation of Basel II, but the ratio would have declined under Basel I). We expect banks will hold higher levels of capital than in the past given that capital markets are uncertain, credit risk is rising and demands on bank balance sheets are likely to increase.

Credit quality is deteriorating

We believe that banks are past the turning point after many years of below average loan losses, and Q2/08 results showed the trend is worsening, particularly at Bank of Montreal. The deterioration of credit quality that started in early 2006 accelerated during the first half of 2008 (Exhibit 39 provides credit quality metrics for each bank). We expect credit losses to continue to rise in 2008 particularly for those with U.S. exposure (Bank of Montreal and to a lesser extent, TD Bank) and those banks that saw large increases in impaired loan formations. In 2009, business lending in Canada and potentially higher consumer loss rates will likely add on to the banks’ credit problems, in our view.

• Specific provisions for credit losses have risen to their highest level since 2003 but are still below historical averages for the industry. The specific PCL ratio rose to 0.40% from 0.29% in Q2/07, and compares to a 17-year average of 0.58%.

• New formations of impaired loans (a good indicator of future provisions, in our view), declined slightly from Q1/08, but were at levels not otherwise seen since Q4/02, and they were up by 156% from Q2/07. (Exhibits 13 to 15)

• Q2/08 allowance ratios declined as gross impaired loans have risen more rapidly than reserves. The total coverage ratio of 113% was down from 129% in Q1/08 and 158% in Q4/07, representing the weakest coverage ratio since Q2/04. (Exhibit 16)

o Specific allowances as a percent of impaired loans are on a declining trend. They are lowest at Bank of Montreal and strongest at CIBC and Scotiabank.

o Reserves in relation to historical losses are strongest at Scotiabank and Bank of Montreal, and weakest at National Bank.

• Recent provisions for credit losses at Scotiabank and National Bank were lowest relative to historical averages. The two banks look clean from a credit perspective in the near term, but their provisions are more at risk of normalizing in 2009 in our view given the two banks’ strength in business lending.

• We are expecting specific provisions for credit losses to rise from $2.0 billion in 2007 to $3.0 billion in 2008 and $5.3 billion in 2009. This implies loss rates of 0.27% in 2007, 0.40% in 2008 and 0.52% in 2009, compared to a 16-year average of 0.52%, adjusted for current loan mix. (Exhibit 17)

U.S. credit losses likely to rise further at BMO (and less so at TD)

• Mortgage delinquencies are at record levels, home equity loan defaults are steadily rising and residential construction and land loan nonperforming assets are skyrocketing for lenders with excess exposure to weakest housing markets in the U.S. We believe losses will continue to rise especially in HELOCs, credit cards, automobile lending, construction lending, commercial real estate and leveraged lending.

o U.S. construction loan exposures are problematic for three of the Canadian banks given the challenged state of the U.S. residential mortgage market, but their size for the three banks that have exposure is manageable. TD Bank has 1.6% of its total loans in U.S. construction loans, and Bank of Montreal 1.2%.

o Exposures to U.S. commercial real estate are also likely to lead to losses. Commercial real estate loans have grown rapidly over the last five years while at the same time credit quality improved and loss ratios reached unsustainably low levels. TD has 6% of its loan portfolio tied up in U.S. commercial real estate following the closing of the Commerce Bancorp acquisition, a higher percentage than for Bank of Montreal (2%).

• Bank of Montreal’s U.S. loans represent 31% of total loans, highest among Canadian banks and the percentage would rise to 33% if including the exposures within Fairway, an off-balance sheet conduit which Bank of Montreal sponsors and has taken over troubled assets from.

• TD's U.S. exposures (26% of total loans) are worth paying attention to but we believe that issues are more likely to arise in 2009 than 2008 as TD fair valued Commerce Bancorp's loan book as at the acquisition closing date. The bank would have had a fair bit of visibility on potential near term impairments, in our view, and would have fair valued loans that had the potential to become impaired in the near term.

Business loans are the loan category witnessing loan losses most below long-term averages and, many indicators are pointing to a turn.

• Rapidly widening credit spreads are usually an indicator that loan losses are about to rise, and we do not believe that the current spread widening is an exception to the rule. (Exhibit 18) The magnitude of the widening does not, however, in our view, solely reflect higher expected defaults and losses. We believe that other factors are also contributing, including risk aversion and deleveraging in the U.S. and Europe.

• Business loan losses have generally followed increases in rating agency downgrades versus upgrades, such as the one we are currently seeing. (Exhibit 19). According to a recent report by Standard and Poor’s, global corporate credit rating downgrades in Q1/08 reached its highest level since 2002 and will likely accelerate because the U.S. has entered a recession.

• Initial signs of this credit contraction were evident in the U.S. senior loans’ officer surveys, as tightening of lending standards has historically been associated with rising credit losses (Exhibit 20). The percentage of institutions indicating that they had tightened lending standards for commercial and industrial loans has been increasing since early 2006, a marked increase from the loose lending standards reported in the period from 2003 to 2006. Even Canadian businesses are feeling tighter conditions. (Exhibit 21)

• Declining corporate output and lower corporate earnings have also been good indicators of rising business loan losses, and those have both started to point downwards. (Exhibits 22)

Outside of the U.S., we expect higher commercial and small business loan losses in Canadian manufacturing, forestry and agriculture sectors for the Canadian banks. Exports account for almost a third of Canada’s GDP and about three quarters of exports head to the U.S. so Canadian businesses, particularly manufacturers in central Canada, are likely to feel the impact of lower demand from the U.S. consumer and the high Canadian dollar. Sharply rising energy prices are also a negative for many businesses.

Business loan losses have been essentially non-existent in recent years for the Canadian banks, averaging (0.07)% in 2005, 0.03% in 2006 and 0.09% in 2007. Loan losses have been low given a solid North American economy, recoveries from loans classified as impaired in the early part of the decade and the relatively easy availability of refinancing options for companies that were getting in financial difficulty. Those factors are all likely to shift in the opposite direction.

• The banks most directly exposed to business lending are Scotiabank, National Bank and Bank of Montreal. CIBC and TD are less exposed.

• If business loan losses rose from 2007 levels to their average of the last 16 years, it would have a material negative impact on the industry’s profitability. Loan losses for the industry would rise by $2.5 billion, or 2.5% of common equity (Exhibit 23).

• Banks could have business credit losses that are lower than the peaks of the early 1990s and early 2000s because of (1) better leverage and liquidity positions today, (2) lower industry and single name limits compared to those prior credit cycles, and (3) more tools to manage risk through credit default swaps, loans sales and securitizations.

Canadian consumer loan losses have not been a worry but a potential increase should not be totally discounted. We believe that the Canadian consumer is healthier than the U.S. consumer, partly because of a healthier housing market. There signs, however, that indicate that losses could inch up, including:

• Canadian employment growth, which peaked in Q4/07 at 2.4%, has declined to 2.1% (Exhibit 24). The labour market beat forecasts once again, creating another 19,200 jobs in April beating market forecasts for a 10,000 job increase, but the unemployment rate edged up to 6.1%.

• Retail sales growth was muted, up a modest 0.1% in March. Excluding autos and parts, sales registered no growth. Although March’s retail sales report showed an increase in real retail sales activity, manufacturing shipments tumbled pointing to subdued GDP growth in March.

• Nationwide housing affordability deteriorated in every quarter throughout 2007 to end up at the worst level since the housing bubble peaked in 1990, according to RBC Economics. (Exhibit 25).

• We are not very worried at this time, but nonetheless retail loan growth and credit quality is unlikely to be as strong in upcoming years as it has been in recent years.

Regulatory capital ratios declined but remain at high levels

Canadian banks’ regulatory capital ratios declined despite raising capital in Q2/08, but capital positions remain strong by global bank standards. Tier 1 ratios are between 9.1% to 10.5%, well above the regulatory minimum of 7.0% (Exhibit 26), but we believe that the banks will review capital markets areas that have not historically attracted a lot of regulatory capital but led to losses in recent quarters, and may reduce the size of those businesses or allocate more capital toward them.

We expect capital ratios to remain well in excess of minimum required regulatory levels (Tier 1 ratio of 7.0%) as we believe it is prudent for banks to hold excess capital in an environment where capital markets are uncertain, credit risk is rising (which should lead to higher RWAs under Basel II) and demands on bank balance sheets are likely to increase as corporate clients increasingly turn to banks for their borrowing needs.

• We believe that rating agencies, regulators and bank boards of directors will all pressure banks to hold higher capital than in the past. For example, The Financial Stability Forum recently proposed several actions on capital requirements to the G7 Ministers and Central Bank Governors, which we believe will be implemented over time:

o To raise Basel II capital requirements for certain complex structured credit products;
o Introduce additional capital charges for default and event risk in trading books;
o Strengthen the capital treatment of liquidity facilities to off-balance sheet conduits.

• We think that regulators and banks themselves will pay more attention to unadjusted balance sheet leverage ratios. Many of the issues that have led to writeoffs came from areas that attracted little in capital requirements on a risk adjusted basis (i.e. certain trading businesses, structured finance businesses as well as off-balance sheet exposures). As a result, banks with strong Tier 1 ratios but high balance sheet assets relative to capital may not be in a position to fully take advantage of what appears to be a strong capital position.

o Bank assets-to-capital multiples declined as shown in Exhibit 27, which is a function of banks having raised approximately $1.8 billion of Tier 1 capital (and more Tier 2 capital) during the quarter with flat asset growth. TD Bank’s multiple increased as its acquisition of Commerce Bancorp boosted assets, and CIBC’s multiple rose as it recorded large writedowns and did not raise any regulatory capital this quarter.

Capital strength will likely be valued more than in a normal environment. Well-capitalized banks (1) are better positioned to withstand surprises and rising credit losses; (2) can take advantage of organic and acquisition opportunities that present themselves as a result of market dislocations; and (3) will be under less pressure to strengthen their capital ratios. We believe that Scotiabank is best capitalized and TD least so when considering the combination of Tier 1 capital, tangible equity ratios, balance sheet ratios and risks to expected profitability (i.e. CIBC has the highest Tier 1 ratio but writedowns related to CDOs of RMBS and financial guarantors remain a risk). Exhibit 28.

The implementation of Basel II in Q1/08 was timely for Canadian banks as Tier 1 ratios under Basel I would be much lower.

• Bank of Montreal's Tier 1 ratio of 9.4% was down slightly (from 9.5% in Q1/08) as risk weighted assets grew 4%, with the highlight being an increase in risk weighted assets related to securitizations, which came as a result of the bank's initiatives to restructure Apex/Sitka. The bank's assets to regulatory capital ratio declined from 18.4x in Q1/08 to 16.2x, highlighting that the bank raised capital to delever the balance sheet. Assets were almost unchanged sequentially while total regulatory capital grew 7%.

• Scotiabank's capital position is strongest among its peers, in our view. The Tier 1 ratio of 9.6% was up from 9.0% in Q1/08 as the bank benefited from the removal of a transitional adjustment related to the implementation of Basel II. The Tier 1 ratio would have otherwise declined by 0.2% in spite of a decline in the assets to regulatory capital ratio. We estimate the bank holds $2.2 billion of excess capital and will generate about $560 million in the next twelve months.

• CIBC’s Tier 1 ratio of 10.5% (down from 11.4%) is highest of the big six Canadian banks. Tangible equity to risk weighted
assets is also higher for CIBC than other banks. We expect a further decline in the Tier 1 ratio in Q3/08, but for CIBC to get into capital difficulties, we believe that financial guarantors have to fail, CDO of RMBS values have to collapse further, and CLO values need to continue deteriorating. In such a scenario, it is important to consider that all bank stocks would be very weak because (1) the failure of financial guarantors would increase concern over the health of the financial system; (2) CLOs trading at a large discount to par would only be economically justifiable if a very nasty credit cycle was about to hit corporations, which would likely have negative implications on the corporate loans of all banks; and (3) concerns over counter-party risk for derivatives would broaden beyond financial guarantor exposures.

• National Bank’s Tier 1 ratio of 9.2% was down from 9.3% in the prior quarter as a 7% increase in risk weighted assets on RWA growth (for both credit and market risk) more than offset a 5.5% increase in Tier 1 capital. The bank's Tier 1 ratio is at the low end of its Canadian peers, partly because the bank will not be ready to adopt Basel II's Advanced IRB approach until fiscal 2010. Its Tier 1 ratio should rise by approximately 50 basis points when that occurs.

• TD Bank’s Tier 1 capital ratio of 9.1% is down from 10.9% in Q1/08 following the close of the Commerce Bancorp transaction. The Tier 1 ratio is the lowest of its peer group and the excess capital the bank generates between now and Q4/08 will be offset by a negative impact of 1.3% on the bank's Tier 1 ratio given changes to the way the bank accounts for its investment in TD Ameritrade. If the year goes as management plans, then we believe TD can improve the ratio quickly after Q4/08 because TD generates about 20-30 basis points of Tier 1 capital per quarter. Our forecast Tier 1 ratio of 8.1% at the end of Q4/08 gives TD very little room for slippage against profitability estimates, particularly when considering ratios based on tangible equity.

Growth in retail businesses slowed in Q2/08; bottom-line growth to slow from recent years

Domestic retail revenue growth was only 2% YoY as pressured margins and difficult equity markets offset solid loan growth. Net income was up 8% from Q2/07 helped by strong growth at Scotiabank (15%), which helped to offset a decline at CIBC (–2%) and no growth at National Bank.

• Scotiabank had the strongest combination of revenue and net income growth, with TD Bank following. We also saw continued underperformance at CIBC and National Bank, and, to a lesser extent, Bank of Montreal. (Exhibits 30 and 31).

o TD’s operating leverage will likely be less than usual in 2008, in our view, given rapid expense growth in late 2007/early 2008, which is more likely to pay dividends in 2009.

o It is difficult to assume that TD will outperform its peers forever but we believe that betting on it has a good chance of succeeding as (1) it has a very large customer base to which it can cross-sell, (2) revenue growth should benefit from sales initiatives and investments made in the past few years, and (3) growth in expenses could be moderated without a significant hit to market share for some time, in our view.

o Scotiabank is coming off a period of rapid asset growth and market share gains, and could probably afford to “milk” those gains at the benefit of the bottom line for some time, if it chose to.

o Of the other three banks, Bank of Montreal and National Bank are both putting a lot of time and effort into centralizing more functions, adding front life staff and improving customer service overall, but they are clearly playing catch up to the leading banks. CIBC’s focus on expenses probably limits its capacity to grow revenues as rapidly as the leading banks in 2009. We also cannot help but think about top management focus in the last 6-12 months was on dealing with the serious issues they encountered in their capital markets businesses, rather than on domestic retail banking.

• Retail loan growth was strong (13% YoY and 5% QoQ) but we believe it will decline as we have started to see deterioration in some factors that drive rapid loan growth – employment growth has slowed, the housing market shows signs of cooling, and the Bank of Canada has signaled an end to lower interest rates.

• Retail net interest income margins remained pressured, down 8 basis points from Q2/07 and up just 3 basis points from last quarter (Exhibit 29). The slight improvement from last quarter was less than we expected given the increase in the Prime/BA spread.

o We had a positive view on bank margins given their ability to reprice loans at this stage of the cycle, as well as the reemergence of a steep yield curve, but recent pricing actions in mortgages indicate the market is still competitive (Exhibit 7), and banks have greater dependence on wholesale funding than in prior cycles, which reprices more quickly than retail deposits. We are not modeling margin improvements for the remainder of the year.

o Larger retail deposit bases may help banks mitigate the negative impact of higher wholesale funding costs on margins. TD Bank stands out as having more of its loan portfolio funded with retail assets, while on the other side, Scotiabank’s domestic retail margins are most dependent on wholesale funding.

• Wealth management revenues were down 6% YoY (and down 1% QoQ, shown in Exhibit 32) due to difficult equity markets, weakened retail brokerage activity including new issue flow and the negative impact of the Canadian dollar for Bank of Montreal. Short term prospects for revenue growth look favourable compared to Q2/08 with the S&P/TSX trading near record levels even after starting the year off on the wrong foot. Asset management and retail brokerage (the main drivers) are heavily influenced by equity market direction.

Capital Markets challenges will continue, but writedowns are likely to get smaller

Bank of Montreal’s exposure to Fairway Finance takes the spotlight from SIVs and ABCP in the near term as our focus shifts on real credit deterioration rather than secondary market credit spreads.

• Fairway Finance - the bank's $9.9 billion U.S. asset backed commercial paper conduit - again led to Bank of Montreal repurchasing assets on which it incurred losses. BMO has funded $851 million in assets, $590 million of which are classified as impaired. Bank of Montreal maintains that the credit quality of what is left is fine but, given the rapid speed at which credit quality is deteriorating in the U.S., Fairway Finance remains a source of credit risk for the bank in our view. $7.2 billion of the $9.9 billion was outstanding as at March 31, 2008.

• BMO-sponsored SIV assets continues to decline as asset sales continue. There is now US$9.5 billion invested in Links and €840 million in Parkland. Negatively, the net asset value of the funds is also declining; as at April 30, it was US$382 million for Links and €108 million for Parkland. In other words, a decline of more than 4% in the market value of Links' assets (13% for Parkland) is all that is required for senior debt holders to incur losses. BMO has undertaken to provide support for the senior funding of Links and Parkland as it matures.

• The bank's exposure to Apex/Sitka led to a recovery of past provisions of $85 million. The bank completed the restructuring of the trusts following the quarter end, and we expect further provision releases in Q3/08. In essence, BMO avoided crystallizing losses by restructuring the trusts and will be proven right if credit spreads remain at current levels over the next five years and/or North America avoids a deep recession. It will, however, suffer greater losses down the road if BBB-rated bond defaults surge to record levels. The bank's exposure to the senior-funding facility (some of which we believe would be drawn if credit spreads widened back to the March peaks) rose from $850 million to $1.0 billion. We do not expect drawdowns on the facility if investment grade spreads remain at current levels. Scotiabank has not been very affected by exposures to hot topic issues so far.

• The $5.9 billion U.S. automobile lending portfolio originated and managed by GMAC but financed by Scotiabank remains an unknown to us. The bank structured the agreement with over-collateralization to protect itself on the downside from a credit perspective (and gave up yield as a result) and management appears very comfortable with its exposure. We do not know, however, at what portfolio loss levels (whether a function of rising delinquencies and/or declining used car values) would Scotiabank begin to incur losses. The bank and GMAC have amended their agreement to include some Canadian automobile loans as part of the $6 billion portfolio.

CIBC’s troubles related to financial guarantors and CDOs of RMBS are not over, in our view.

• We do not think that the bad news on CDOs of RMBS and financial guarantors is necessarily all out, and are expecting further writedowns at CIBC ($1 billion in Q3/08E). CIBC's writeoffs and other unusual items of $2.6 billion in Q2/08 were much larger than the $1.5 billion we had expected. The bank has put its structured finance operations in "run-off" and will look to reduce its positions over upcoming quarters and years. The bank still has $2.9 billion in fair value of hedges with financial guarantors, and an additional notional $1.7 billion in CDOs of RMBS hedged with financial guarantors. Outside of those two items, which we perceive of being at a high risk of further writedowns, the bank has an additional $22.9 billion in structured finance exposures that relate mainly to corporate debt/CLOs that are hedged with financial guarantors, and $2.2 billion in of various other unhedged holdings of structured credit.

National Bank still faces risk with ABCP but that risk is down today, in our view.

• We view the bank’s exposure to ABCP as less than problematic than before given the progress made in the ABCP restructuring plans and lower credit risk spreads compared to the March peaks (Exhibit 10), which has positive implications for the value of the ABCP. Risk remains as the restructuring is not yet complete and there has been no external validation of the value of the restructured assets since no notes have traded.

• There was no change to the bank's allowance for ABCP (we had estimated an allowance of $150 million, which would have taken the allowance from 25% of par value to just over 30%). We have moved that estimated allowance to Q3/08 in our model, with key drivers of value between now and then being (1) a successful restructuring of the ABCP under the Montreal accord, and (2) the direction of investment grade spreads. We should point out that Q3/08 results should benefit from an $88 million ($59 million after tax) securities gain related to the merger of the Montreal Exchange and TSX Group.

Capital markets businesses weak in Q2/08 – we expect a rebound in H2/08

Bank wholesale earnings were negatively impacted by large writedowns that totaled $3.7 billion as shown in Exhibit 12. Furthermore, the capital markets turmoil made it a tough environment for banks to grow core earnings.

• Core net income declined across the board, down 29% YoY and 27% QoQ. Core wholesale revenues also declined on a core basis as well as on a GAAP basis (except Bank of Montreal which saw a 5% increase from net gains). (Exhibit 37)

• Writedowns were partly offset by gains. CIBC recorded the largest writedown ($2.6 billion). Four banks also recorded gains in the quarter led by Bank of Montreal’s net gain of $213 million.

• Trading revenues were $(1.9) billion, versus an industry run rate of $1.2 to $1.6 billion per quarter. Excluding writedowns, trading revenues were weak at CIBC and TD. Certain trading businesses such as foreign exchange and commodities benefited from higher volatility and activity, but liquidity declined in other areas related to fixed income and credit trading.

• Provisions for credit losses of $88 million for the six banks are low by historical standards but still grew from the $25 million recovery in Q2/07. Bank of Montreal saw the largest loan loss increases. (Exhibit 38)

We expect a rebound from the weak performance in the second half of the year at most banks, but the profitability of the wholesale divisions compared to the “pre-disruption quarters” could be weakest for CIBC and Bank of Montreal, in our view.

• Both firms have been hit by writeoffs that were very large in relation to their income base and we believe that the de-risking efforts that are underway will have a negative impact on risk appetite and net income potential on a rebound. The other banks have not been as affected by writedowns and we do not expect a meaningful shift in risk appetite.

We still expect capital markets writedowns to continue, particularly for banks exposed to financial guarantors and CDOs of RMBS, but they will likely be smaller.

• We do not think that the bad news on CDOs of RMBS and financial guarantors is necessarily over, and are expecting further writedowns at CIBC ($1 billion). However, we do not believe that capital markets writeoffs will be as large as in prior quarters for the most affected banks, given improvements seen in many areas of credit markets. Off-balance sheet exposures remain a risk, however, as deteriorating economic conditions will likely lead to declining values in assets held in those conduits, which may ultimately lead to losses for the banks (particularly at Bank of Montreal).

• Banks that have reduced risk or entered into trading strategies involving cash assets and credit derivatives may face unexpected losses given the divergence between spreads on cash assets and derivatives. The spreads on credit derivative indices (what banks typically use to hedge their cash positions) have tightened more rapidly than have the spreads on cash assets in recent months, which may cause the banks to record losses related to what were hedged positions.

Valuation and Price Target Impediments

• BMO (Underperform, Average Risk): Our 12-month price target of $44 is based on a price to book methodology. Our P/B target of 1.4x book value in 12 months is at the low end of the banking sector given a lower ROE and higher risks related to off-balance sheet exposures. Our target implies an approximate P/E multiple of 8.5x 2009E earnings. The 5-year average forward multiple is 12.2x. Risks to our price target include the health of the overall economy and sustained deterioration in the capital markets environment, the US housing market, its exposure to ABCP conduits and Structured Investment Vehicles, and its commodities trading portfolio.

Additional risks include greater-than-anticipated impact from off-balance sheet commitments, the potential for non-accretive acquisitions and/or related execution risk, litigation risk, declining domestic market share and a rising Canadian dollar. Risks to our cautious view relative to peers include better than expected operating leverage, a turnaround in retail earnings and changing sentiment toward bank mergers.

• BNS (Sector Perform, Average Risk): Our 12-month price target of $49 is based on a price to book methodology. Our P/B target of 2.3x in 12 months is higher than our target average for the banks given a higher ROE as its lower exposure to headline risks and solid performance in the rapidly growing international division is offset by greater exposure to business lending and the strong Canadian dollar, while the domestic franchise (although improving) lags the leading banks', in our view. Our target implies an approximate P/E multiple of 10.9x 2009E earnings. The 5-year average forward multiple is 12.6x.

Risks to our price target include the health of the overall economy, sustained deterioration in the capital markets environment and greater than anticipated impact from off-balance sheet commitments. Additional risks include the potential for non-accretive acquisitions and/or related execution risk, deterioration in the Latin American political and economic climate, litigation risk, a rising Canadian dollar and rising business loan losses.

• CM (Underperform, Average Risk): Our 12-month price target of $64 is based on a price to book methodology. Our price to book target multiple of 2.0x reflects a higher risk premium offset by the bank's high ROE relative to peers. Our target multiple relative to ROE is the among the lowest of its peers reflecting more exposure to sub-prime CDOs and financial guarantors, below average retail banking trends, and lower confidence about unknown exposures. Our target implies an approximate P/E multiple of 8.6x 2009E earnings. The 5-year average forward multiple is 11.8x.

Risks to our price target include the health of the overall economy, sustained deterioration in the capital markets environment, the US housing market and the monoline industry, litigation risk, and greater than anticipated impact from off-balance sheet commitments. Additional risks include loss of domestic market share, a decline in underwriting activity and weakening retail credit quality.

• NA (Underperform, Average Risk): Our 12-month price target of $52 is based on a price to book methodology. Our P/B target of 1.60x in 12 months is among the lowest of our target for banks given a higher risk premium, based on more exposure to ABCP and wholesale earnings, and deteriorating credit quality. Our target implies an approximate P/E multiple of 9.0x 2009E earnings. The 5-year average forward multiple is 11.2x.

Risks to our price target include additional write-downs of Asset-Backed Commercial Paper assets held on its balance sheet, the health of the overall economy, sustained deterioration in the capital markets environment, the health of the Quebec economy, an unexpected acquisition and a change in the competitive or political environment in Quebec. Additional risks include rising business loan losses, greater than anticipated impact from off-balance sheet commitments, and litigation risk.

• TD (Sector Perform, Average Risk): Our 12-month price target of $69 is based on a price to book methodology. Our P/B target of 1.7x in 12 months is slightly lower than our target for banks given a lower ROE offset by its relatively lower exposure to headline risks and leading domestic retail franchise. It implies an approximate P/E multiple of 10.9x 2009E earnings. The 5-year average forward multiple is 12.2x.

Risks to our price target include the health of the overall economy, sustained deterioration in the capital markets environment and greater than anticipated impact from off-balance sheet commitments. Additional risks include an unexpected acquisition, integration risk with Commerce Bank, TD Ameritrade and TD Banknorth, pricing pressure in the discount brokerage industry, a rising Canadian dollar, litigation risk and a worse than expected impact from Enron-related litigation (although it appears that risk has declined, given a court ruling in another Enron trial).
;