Over the last six months, the S&P 500 Financials sector has rallied 24.2% (banks are up 34.3%) versus 8.3% for the S&P 500. This return has outpaced every other sector, and financials have also had the best return for the last 52 weeks and one month. Most of the drivers for the sector are hooking up, but valuation has also risen so earnings must now deliver to justify the outperformance.
Figure 1: Returns
Source: Spellman, FactSet
Earnings have been improving since the financial crisis (figure 2), and as expected, earnings growth relative to the S&P 500 is somewhat correlated with the sector’s relative returns to the S&P 500 (figure 3). The sector’s latest twelve months (LTM) earnings growth on a monthly basis has been greater than the S&P 500 since June 2015, and in February 2017 financials earnings grew nearly 8% more than the market.
Figures 2 and 3: Financial Earnings (left) and Financial Relative Earnings Growth and Relative Price Returns (right)
Source: Spellman, FactSet
As the sector’s results have improved, its P/E, P/CF, and P/B multiples have risen (figure 4). The P/E and P/CF are above their averages while P/B is below. The current P/E is 15.3 versus an average of 14.7 since the end of 2014, and P/CF is 9.2 versus an average of 6.9. P/B is being held back by ROE which remains low despite the rebound in earnings (figure 5).
Figures 4 and 5: P/E, P/CF, and P/B (left) and P/B vs ROE (right)
Source: Spellman, FactSet
Net profit margin and ROA are back to pre-recession highs, but the improvement in ROE has been limited by lower leverage as financial institutions repaired balance sheets and pared back financial risk. Driven by regulation and other incentives, total debt to total assets is about half of what it stood at before the crisis (figure 6). P/B is 1.34 versus 2.10 at the end of 2016, or down 36%. This compares to ROE which fell 42% from 14.7% to 8.6% today. The decline in P/B appears to be an over-reaction. While ROE is lower, ROE is safer.
Figure 6: ROE Decomposition
Source: Spellman, FactSet
.
Relative earnings helps explain relative returns (figure 3), and so does relative valuation (figure 7). Figure 7 shows a composite of relative P/B, P/E, and P/CF to the S&P 500. The composite places 50% weight on P/B and 25% for P/E and P/CF. Relative 1-year returns are lagged as valuation is predictive. Lower multiples lead higher returns, and vice versa. The recent rebound in multiples suggests that lower returns are ahead; however, multiples are still low versus history and, as noted earlier, drivers are hooking up. Relative multiples have fallen over time as growth has slowed for the sector (the unwinding of the bubble, lower interest rates) and due to memories of financial crisis. The relative multiple (0.60) is about at its five-year moving average (0.62). Thus, if the last five years is the new normal, then the stocks are reasonably priced. Some believe that fundamentals are set to improve further, and eventually memories of the financial crisis will fade. Thus, it is possible that multiples could rise further if fundamentals do, indeed, rise. Yet, much of the easy money has been made.
Figure 7: Relative Valuation and Relative Returns
Source: Spellman, FactSet
The S&P 500 financial sector consists of five industries: (1) banks (46% of market cap), (2) diversified financial services (11%), (3) consumer finance (5%), (4) capital markets (19%), and (5) insurance (18%). Fundamental drivers of the sector can be categorized into six categories below.
1. Interest Rates
2. Business Conditions
3. Lending Trends
4. Consumer Health
5. Investment Banking
6. Regulations
.
The first four drivers are improving, and animal spirits may lead investment banking to rebound as well. Several of the drivers are also correlated with asset prices, which is good for the asset management business of financial institutions. Furthermore, regulations may be relaxed due to Trump and Congress initiatives. Figures 8 and 9 illustrate a composite of the first five drivers. The current reading suggests that fundamentals have hooked up and justify the sector’s recent. The composite is created from a weighted average of three-year percentile (within a range) for each variable contributing to the five drivers. The interest rate category contributes 40% weight to the composite, consumer health 30% and business conditions (12.5%), lending trends (12.5%), and investment banking (5%) make up 30%.
Figures 8 and 9: Fundamental Driver Composite and Relative Returns
Source: Spellman, FactSet
Figure 10: Fundamental Composite
Source: Spellman, FactSet
.
The fundamental driver composite correlates well with relative returns except for the period leading up to the peak of the internet bubble and its burst. During that time, the industry was going through significant consolidation, earnings barely declined during the early 2000s recession, and the sector became undervalued. From January 2003 to present, the R-squared is 0.35. Equal-weighting fundamental drivers with the valuation composite boosts the R-squared to about 49% (figure 11), and better explains the period of the internet bubble and burst (the outperformance of 2000-1 was due to the sector being undervalued – figure 7). The current reading suggests that the sector is stretched.
Figure 11: Relative Valuation and Relative Returns
Source: Spellman, FactSet
.
Let’s take a closer look at the drivers.
Interest Rates
The rally in the financial sector has coincided with the Federal Reserve’s plans to increase interest rates. A bank’s bread and butter is its profits from the lending business, which has declined for years as the yield curve flattened and net interest margins declined. Figure 12 shows the relationship between the rate index and relative returns, which has been tight except for the internet rate bubble and burst. Figure 13 illustrates how the spread between mortgages and Fed Funds rate has declined since 2009 from about 5% to 3% at the narrowest level before widening over the last year. Annual relative returns are driven by the one-year change in spreads (figure 14) and overall rates (figure 15). Except for the internet bubble and during the financial crisis when spreads widened (while loan losses grew), change in the spread has been a good coincident indicator of returns.
Figures 12-14: Interest Rates Index (top left), Mortgage Spread (top right), Change in Mortgage Spread (bottom left),
Change in 10 Yr Treasury Bond Yield (bottom right)
Source: Spellman, FactSet, FHLMC
.
The interest rate model places a 50% weight on 1-year change in the 10-year Treasury Bond yield. The inverse of high yield credit spreads (BofA Merrill Lynch High Yield YTM – 10 Yr Treasury Bond Rate) also has a 50% weight. The yield on the 10-year Treasury Bond has risen since last summer, and higher long rates are a proxy for overall improvement in the economy which benefits financial institutions. Lower credit spreads are positively related to relative returns, this suggests that banks recover as lending for the most risky companies becomes more relaxed. The current spread, which has narrowed over the past year, is 3.6%, versus a 27 year low of 2.8%, max of 18.6%, and average of 5.4%.
Business Conditions
Business conditions have rebounded sharply over the last year (figure 15). Surveys of manufacturers (ISM (NAPM) Manufacturing Index) and small businesses (NFIB Small Business Economic Trends) have rallied. Assuming this leads to business expansion, this should be good for bank lending growth and the percent of non-performing loans. Figure 16 shows that the ISM Manufacturing Index Leads Private Nonresidential Construction by about 12 months. The current reading of 57.7 suggests that construction is about to rise.
Figures 15 and 16: Business Conditions Index (left) and ISM Mfg Index
vs Private Nonresidential Construction
Source: Spellman, FactSet, US Census Bureau, Institute for Supply Chain Management
.
Lending Trends
While interest rates and business conditions are driving up financial stocks, trends in lending are less sanguine (figure 17). Loan officers are tightening standards, charge-offs are beginning to rise, and loan growth has recently slowed. Easy credit has been available for some time, and corporate debt and consumer debt has been growing, so tighter credit and losses are not unexpected. However, if the economy continues to chug along and regulations are loosened, charge-offs should remain controlled and loan growth could remain moderate or improve. On the other hand, these variables are signs of cracks in the foundation of this expansion, so a slow-down in the economy could make this situation unravel (and the low credit spreads) quickly and propel a negative feedback loop.
Figures 17-19: Lending Trends Index (top), Tightening and Loss Trends (bottom left),
and Tightening and Loan Growth (bottom right)
Source: Spellman, FactSet, Federal Reserve System
Consumer Health
The consumer health index has been running at high levels since 2012 (figure 20). Some of its components include consumer confidence, debt growth, and housing prices. Consumer confidence is at a high level (current 115 versus average of 91, low of 25, and max of 145 over the last 27 years). Consumer confidence has rebounded 29 points over the last year and is correlated with lending trends (figure 21). Confidence leads credit card debt growth, and housing prices lead consumer confidence (figure 22). Housing prices are up 5.8% in the latest month and the growth rate trending up, so this bodes well for lending trends.
Figures 20-22: Consumer Health Index (top), Credit Card Growth and Consumer Confidence
(bottom left), and Consumer Confidence and Housing Prices (bottom right)
Source: Spellman, FactSet, S&P Case-Shiller, Conference Board, Federal Reserve Bank of New York
.
Investment Banking
.
Investment banking, as measured by transactions – M&A and initial public offerings – is subdued. While this is not highly correlated with relative returns of the financial sector (figure 23), it benefits large financial institutions and it is a gauge of the overall level of optimism of corporate America (M&A) and the public in financial markets (IPOs). M&A and IPO trends are highly correlated with the return of the overall market (figures 24-5), so it is somewhat surprising that these activities are weak. Although, IPOs trends are starting to pick up.
Figures 23-25: Investment Banking Index (top), US M&A Volume (bottom left),
and IPO Transactions (bottom right)
Source: Spellman, FactSet
.
Regulations
.
Reductions in regulations being discussed by Trump could benefit financial institutions in the short-run. Whether changes will materialize anytime soon is yet to be determined, but it appears stocks have rallied in part due to potential regulation relief. Already, the DOL’s new fiduciary rule is being delayed. As noted earlier, bank capital levels have risen and dragged ROE down; although, ROE is healthier than before. When dire events occur (financial crisis), people often over-react. Is the Dodd-Frank Bill of 2010, which was 2,319 pages (versus 61 for the Sarbanes-Oxley Act of 2002 and 31 for the Federal Reserve Act of 1913) too extensive? As time proceeds, memories fade and people focus more on recent information. Since the sector has improved, regulators and investors are more apt today to ignore risks. Therefore, substantial unravelling of regulations could be detrimental long-term. Besides reduced regulations, tax reforms could help the sector as its effective tax rate (figure 26) is higher than the market.
Figures 26: Sector Tax Rates
Source: Spellman, CNBC, The Earnings
Scout