16 May 2012

Preview of Banks' Q2 2012 Earnings

Scotia Capital, 16 May 2012

Banks Begin Reporting May 23

• Banks begin reporting second quarter earnings with Bank of Montreal (BMO) on May 23, followed by Royal Bank (RY) and Toronto-Dominion Bank (TD) on May 24, Bank of Nova Scotia (BNS) on May 29, Canadian Imperial Bank of Commerce (CM) and National Bank (NA) on May 31, Laurentian Bank (LB) on June 6, and Canadian Western Bank (CWB) closing out reporting on June 7.

• Scotiabank GBM’s earnings estimates are highlighted in Exhibit 2, consensus earnings estimates in Exhibit 3, and conference call information in Exhibit 4.

Second Quarter Earnings – 4% YOY Increase, Down 5% Sequentially

• We expect second quarter operating earnings to increase 4% year over year (YOY) but decline 5% sequentially from the strong first quarter.

• We expect Wholesale earnings to decline QOQ, Wealth Management to be up moderately, with Domestic/Retail Banking earnings resilient, although growth is slowing (see Exhibit 8). Domestic/Retail Banking earnings are dependent on net interest margin performance as volume growth slows. Overall, our second quarter earnings estimates are in line with consensus.

• We expect bank profitability to remain solid this quarter, although down from highs, with return on equity at 17.5% versus 18.8% in the previous quarter. Return on risk-weighted assets (RRWA) remains extremely high at 2.31% versus 2.41% in the previous quarter.

• The main earnings variables to focus on this quarter, in our view, are trading revenue and the retail net interest margin.

• Trading revenue will likely be negatively impacted by a deterioration in sentiment and increased uncertainty as European sovereign debt concerns resurfaced in March and April. We expect trading revenue to decline 21% QOQ and 7% YOY due to lower fixed income trading revenue (lower underwriting) and lower equity trading volume.

• The sequential trading decline is expected to be impacted by lower fixed income trading revenue, with 10-year Canadian and U.S. government bond yields increasing 15 bps and 12 bps, respectively, and fixed income underwriting activity declining. Canadian government bond and corporate bond underwritings declined by 25% and 20% QOQ, respectively, which is expected to translate into lower fixed income trading volumes. TSX average trading volumes (equity) were relatively soft, declining 3% quarter over quarter (QOQ) and 19% YOY, with block trading volumes down 6% sequentially. The S&P/TSX Composite Index declined 1% in the quarter; however, the majority of the decline in the index occurred in April, with the average value of the index in the quarter up 3% QOQ.

• We expect bank group underwriting and advisory revenue to increase 12% sequentially to $861 million, although down 10% YOY. Underwriting revenue is expected to be led by a 54% QOQ increase in equity issuance, although the lower margin fixed income underwriting is lower as previously noted. M&A activity was up 2% sequentially, although declining 5% YOY.

• We continue to forecast net interest margin compression of 2 basis points per quarter out to the end of 2013. Rational pricing is required to mitigate some of the margin pressure.

• Credit trends remain stable with loan loss provisions expected to be flat at $1.6 billion or 0.37% of loans.

Dividend Increases Unlikely – Timing Discretionary

• Dividend increases this quarter are unlikely with perhaps the exception being NA, as TD, RY, and BNS increased their dividends last quarter, and CWB, NA, and LB increased their dividends in the fourth quarter of 2011.

• The bank group’s dividend payout ratio is currently 45% of our 2012 earnings estimates versus the bank group target payout ratio range of 40% to 50%, so at the midpoint of the range.

• In our view, the strongest candidates for dividend increases in the remainder of fiscal 2012 based on target payout ranges (see Exhibit 5) are LB, NA, CWB, and BMO, followed by CM, TD, BNS, and RY. We expect dividend growth in 2012 to mirror earnings growth in the 6%-7% range. Timing continues to be extremely discretionary.

Bank Share Performance & Valuations - Canadian Housing Concerns

• Canadian bank share prices have outperformed the TSX by 6% in the quarter and 6% year-to-date as at May 14, 2012. As systemic risk from European sovereign debt fears eased in the first 3 months of 2012, the bank beta trade was on, with Bank of America, Citigroup, and JP Morgan up 79%, 42%, and 39%, respectively, versus 11% for Canadian banks. However, an uptick in systemic risk and uncertainty in Europe has resulted in a correction for bank stocks, with Bank of America, Citigroup, and JP Morgan down 26%, 25%, and 22%, respectively from the 2012 highs, versus an 8% correction for Canadian banks.

• We believe Canadian bank P/E multiple expansion has been delayed by increased headline risk and concerns over a Canadian housing correction and slower consumer loan growth, in addition to the uptick in uncertainty from Europe. We believe these concerns have been factored into bank valuations, with U.S. hedge funds potentially short selling Canadian bank shares to take advantage of the increased headline risk.

• We have analyzed the banks' exposure to a housing correction, concluding that price declines would be very manageable for the banking sector especially given the soundness of the Canadian banking system and Canada's relatively strong and stable fiscal position. The Canadian banks have roughly $900 billion in residential mortgage lending exposure (including helocs), with on average 64% insured and 36% uninsured. The insured/uninsured mix varies between the banks, with CM carrying the lowest proportion of uninsured mortgages at 21% and RY the highest at 64%. The average LTV on the uninsured books is 53%, ranging from 48% at CM to 59% at NA. The average LTV on the insured book is 66%, ranging from a low of 49% at CM to 86% at NA (see Exhibit 6).

• We stress tested bank earnings assuming that 10% of total uninsured mortgages were originated at peak house prices and at the maximum LTV of 80% with house prices declining 25%-35% (see Exhibit 7). The stress test also assumes mortgage holders will default when the value of their mortgage is greater than the value of their house. Under this severe stress test, the estimated direct bank earnings impact would be a decline of 5% to 16% with 2012E EPS as a base (see Exhibit 7). We note that the largest peak to trough house price correction in Canada occurred between February 1989 and May 1990, with house prices declining 14%.

• We estimate condos to represent approximately 10% of bank mortgage portfolios, with a guesstimate of 40% of the condo portfolio in Toronto and 20%-25% in Vancouver. If our stress test was applied to the condo exposure only, the direct earnings impact is estimated at approximately 1% to 2%.

• Additionally, historical data shows a strong correlation between house prices and employment levels, with current Canadian employment rate at a very strong 62%. It seems that a 10% house price correction is a reasonable assumption but anything significantly greater than 10%, we believe, would require a major drop in employment levels. A major factor in US house price declines was the collapse in the employment rate compounded by a collapse in underwriting standards which we don't believe we have seen in Canada. CMHC is no Fannie Mae.

• The market continues to chase high dividend yielding sectors, creating valuation premiums in Pipes & Utilities and REITs, while systemic risk has muted bank stock participation. The negative impact of systemic risk on valuations is evident when you compare bank dividend yields versus Pipes & Utilities (see Exhibit 19) and REITs (see Exhibit 20), where systemic risk is relatively low. As systemic risk eased in the first three months of the year, bank stocks began outperforming, with an apparent sector rotation out of Pipes & Utilities and into banks; however, the recent uptick in systemic risk has capped the outperformance. Bank dividend yields are now 1.0 standard deviations above the mean versus Pipes & Utilities and 2.7 against REITs.

• Bank earnings yield relative to Canadian corporate BBB bonds is at an all-time high, even higher than in Financial Crisis I (Lehman Brothers collapse), which is reflective of the level of systemic risk that still persists in bank valuation (see Exhibit 21).

• We believe the recent bank share price weakness offers an attractive entry point into bank stocks.

Maintain Overweight Recommendation

• Our share price targets remain unchanged. Our share price targets are based on 12.0x our 2013 earnings estimates and we believe are extremely conservative in the context of the interest rate environment, dividend levels, and capital generation rates.

• In fact, if Pipes & Utilities can trade at 20x to 22x earnings, we believe a banks stock can trade at 15x to 17x in a period of low systemic risk and perhaps a period where the market recognizes the decline in the risk premiums post the full implementation of Basel III and the global restructuring of the banking industry.

• We maintain 1-Sector Outperform ratings on TD, RY, and CM, 2-Sector Perform ratings on CWB, BNS, and LB, and 3-Sector Underperform ratings on BMO and NA. We believe CM is the most undervalued stock in the bank group by a wide margin.