Going Long Canadian Banks
Making up more than 20% of the S&P/TSX Composite Index and acting as a critical enabler of total economic activity, the health of the Big 6 Banks is of paramount importance to Canada, and Canadians at large. Since the global financial crisis of 2008, we have often seen US hedge funds take short positions in Canadian bank stocks and subsequently flood the press with reasons to be short these stocks. Each time we have seen these short sellers go away quietly as their short thesis does not play out as planned. Note the S&P/TSX bank index has outperformed the broad S&P/TSX Composite benchmark by approximately 700 basis points on an annualized basis since the March 2009 recession lows.
Recently, a well-known hedge fund manager has commented on their new short position in select Canadian banks with expectations for a significant decline in profitability particularly related to higher loan losses. We do not see this as a likely scenario over the near to medium term. We compared many of the recent short seller’s views to our own views on the Canadian economy, bank profitability, loan loss expectations and stock valuations. We believe Canadian bank stocks are attractive at current levels, trading well below historical average valuations, and we expect a re-rating in the coming period driven by solid profitability and dividend growth.
Canadian Economic Backdrop
Let’s look first at the Canadian economy, an important driver of bank revenue. Although GDP growth expectations in Canada have declined, the broad consensus is that we will still experience positive growth for the remainder of 2019 and into the beginning of 2020. The Canadian economic cycle has traditionally followed the US cycle closely, and we expect that will be the case going forward. Employment trends in both the US and Canada remain strong, with unemployment rates at historically low levels. We follow several US economic indicators (jobless claims, unemployment, credit conditions, Conference Board LEIs and the change in the oil price) that show little probability of recession in the near term.
When it comes to the consumer, it has been well telegraphed that Canadian consumer debt levels are at record levels relative to incomes. However, we believe this risk is mitigated by the fact that household assets and resulting net worth are also at record levels.
Bank Earnings and Valuations
Strong cost control has allowed Canadian banks to generate positive operating leverage over the last several years. The banks have controlled expense growth which has increased at a slower rate than revenue growth, and we expect this will continue given the initiatives outlined by current management teams. We expect the banks will grow their revenues by mid-single digits in 2019, derived from a diverse mix of business operations. Although consumer loan growth is expected to slow, we believe that commercial and corporate loan growth will remain strong. We would like to remind investors that interest income generated by personal and commercial loans only accounts for 25% of overall bank revenues, an important point for our discussion on loan loss provisions that follows.
Based on the above, and other market considerations, consensus estimates for 2019 are that aggregate bank EPS growth will come in around 5%. The team believes these expectations are reasonable given the outlook for the Canadian economy. Overall, we think that banks are attractively valued at this juncture and offer a good margin of safety relative to our estimates of fair value. Further, an examination of valuations relative to their own historical trading range (Figure 1), and relative to Canadian equities broadly, shows that valuations are on the lower end of their long-term trading ranges (Figure 2). In both instances, the banks sat roughly one standard deviation below their historical relative trading range at the beginning of May. Should earnings growth remain healthy, and we believe it will, then we would expect some reversion to the mean with bank valuations broadly.
Figure 1: S&P TSX Banks - Historical Forward P/E
Figure 2: S&P TSX Banks - Relative to S&P TSX Composite
Loan Loss Provisions
Recently, short sellers stated publicly that they believe nearterm Canadian bank loan losses will normalize at higher than current levels (Figure 3). We do not subscribe to this scenario in the near-term. Consumer loan losses are highly correlated to unemployment levels and less so to debt levels. Current unemployment levels sit near historical lows and we do not expect there will be a significant rise in unemployment in the next 12 months.
When we look beyond consumer loans and examine residential mortgage and business/government loans, we see a healthy aggregate loan portfolio. Business loan losses, traditionally more volatile than consumer loan losses and more likely to spike during a recession, are less concentrated by sector than they have been in previous recessions (e.g. real estate in the early 1990s and telcos in the early 2000s). We believe this more diversified loan portfolio will reduce aggregate potential business default risks during recessionary periods (Figure 4). Losses on residential mortgages, which account for 42% of all loans, have been well managed traditionally, given the underlying collateral levels. We would point out that 43% of residential mortgages are insured and ultimately backed by the Government of Canada while the uninsured mortgage portfolio has an average loan-to-value of 55%1.
Figure 3: Canadian Bank Provisions for Loan Losses
Figure 4: Canadian Bank Provisions by Loan Type
Another issue raised by this hedge fund manager is that banks have been taking less loan loss provisions than required to boost reported earnings. The short seller implied that 90% of 2018 earnings growth was generated through a reversal of Stage 1 loan loss provisions. Although as a group the banks did see some reversal of Stage 1 provisions, in our view the magnitude of that reversal was insignificant. Furthermore, under new IFRS 9 accounting rules, Stage 1 and Stage 2 loss provisions are taken on performing loans and, in our view, should be assessed on a combined basis. In 2018, collectively, the banks took a reversal of $167M on Stage 1 provisions, however, this was more than offset by the $370M provisions taken on Stage 2 loans. In reality, these provisions on performing loans are dwarfed by the $7.6B in provisions taken on non-performing loans.
In aggregate, we believe that current loan loss provisions appear appropriate, equating to 29bps2 of loans.
We have run our own stress tests on potential loan losses. Under our worst-case scenario, a very deep and prolonged economic recession resulting in a -40% correction in housing prices and significant losses in each of the banks real estate and commercial loan books, we would expect loan losses and provisions to increase to roughly 100bps of loans. As we noted above, none of the economic indicators we follow are signalling a recession in the near term. Thus, this worst-case scenario is not one we would expect over that period.
Under our stress test, we also found that capital declines ranged between 100 and 200bps on a common equity tier one (CET1) ratio. The CET1 ratio is an Office of the Superintendent of Financial Institutions (OSFI) minimum requirement of high-quality capital to risk-weighted assets for the Big 6 Banks. Despite this requirement, all six Canadian banks remained profitable and generated profits in excess of their dividend commitments during the stress test period. Additionally, given that all six banks have a CET1 ratio over 11% (Figure 5), this amount of capital decline remains manageable. Capital levels would remain well above OSFI’s mandated 8% level before dividends would need to be cut.
Bank capital levels have increased significantly over the past several years and balance sheets are better prepared for an economic recession than they have been in years past, with risk weightings on underlying assets higher than they have been historically. Higher levels of tangible equity capital - defined as share capital plus retained earnings, less goodwill and intangible assets - has resulted in much lower leverage (Figure 6). Even with this higher equity base, the banks have been able to generate mid-teens returns on equity.
Figure 5: Canadian Bank Group - Tangible Common Equity Ratio
Figure 6: Canadian Bank Group - Leverage Ratio
We currently see an attractive opportunity in Canadian bank stocks. These banks are high quality franchises operating in an oligopoly. The banks have historically proven themselves to be prudent risk managers while consistently earning ROE’s in excess of their cost of capital generating strong excess capital.
Over the near term we expect the banks to generate mid single digit earnings growth. This continued growth should allow for a re-rating of the stocks which currently trade near the bottom of their historical valuation range. We expect these factors, together with strong dividend yields, to drive decent returns from the stocks over the near to medium term.
To find out more, ask your Mackenzie representative. Let's work together.