There has been much volatility in the banking sector over the last few months: Barclays is exiting Africa, Old Mutual may be selling its stake in Nedbank, credit rating downgrades loom. So how does one think about banks as an investment – is there value to be found or should one approach with caution? Mark Dunley-Owen investigates.
The market capitalisation of Standard Bank, FirstRand, Nedbank and Barclays Africa, South Africa’s big 4 banks, peaked at just under R900bn in April 2015. As shown in Graph 1, it fell by R338bn or 38% in rands over the remainder of the year. In dollars, it more than halved. Macro conditions, such as low economic growth and questionable government policies, have been blamed for the fall. A better answer is to view price as equal to valuation times earnings, and to consider each of these in turn.
Valuation
The simplest explanation for the recent fall is that banks were overvalued. Graph 2 shows the price-to-earnings (PE) multiple for the big 4 banks since 2000, the red line being the absolute PE and the grey line the banks’ PE relative to the FTSE/JSE All Share Index (ALSI). In April last year the banks’ absolute PE was at the top end of its range. Historically this has been a good time to sell and, in hindsight, it is unsurprising that bank share prices fell.
Fast forward to today, and bank valuations appear to have overreacted on the downside. In absolute terms they are approaching historic lows and relative to the ALSI they are at the global financial crisis (GFC) low. Negative sentiment linked to weakening macro conditions may have caused this overreaction, as well as uncertainty following the proposed sales of large stakes in Absa and Nedbank by their parent shareholders. We believe these are temporary headwinds and, given time, valuations are likely to rise from current levels. How this will affect the prices of bank shares depends on what happens to bank earnings.
Earnings
At its simplest, a bank makes money by borrowing from savers and lending to borrowers. Much of South Africa’s banks’ earnings thus depends on how much money they lend out, the difference between the interest rate they charge borrowers and the interest rate they pay savers, and how much of the money borrowers can’t or won’t repay. The first is known as their loan book, the second is the interest margin, and the third is bad debts.
It is reasonable to believe that some or all of these variables have worsened recently. But by how much? For context, during the GFC in 2009, the big 4 banks’ loan book fell by 4%, their interest margin reduced by 0.55% points and their bad debts as a percentage of loans rose by 0.9%. These look like small numbers but the cumulative effect was that headline earnings of the big 4 banks fell by about a quarter in 2009. We don’t expect bank earnings to fall by the same amount as they did during the GFC: economic conditions are less extreme, and the balance sheets of South African banks are stronger today than in 2007.
History provides useful guidelines, one of which is that bank earnings fall a lot after periods of rising asset prices and debt levels. Graph 3 shows the change in South African house prices and the change in private sector credit growth, both over five years, with inflation taken out. These graphs highlight the differences between today and the GFC. Real house prices and real private sector credit more than doubled in the five years prior to the GFC, meaning that banks entered the GFC with large loan books backed by high-priced assets. It was probable that credit growth and asset prices would revert to their long-term trend lines and, as they did, that bank earnings would fall materially.
Contrast this to today. Real house prices and private sector credit have barely grown over the last five years. Banks have made fewer loans, backed by more fairly valued assets and serviced by less indebted clients. Even if South Africa is entering another crisis, bank earnings are starting from a lower base with less downside risk.
Standard Bank’s mortgage loan book illustrates this point. This doubled over the three years prior to the GFC, from R124bn to R252bn and, by 2008, the weighted average interest rate on these loans was prime minus 2.5%. From a shareholders’ perspective, Standard Bank had lent out too much money too quickly at too low a margin. The subsequent poor performance of these loans confirmed this and, seven years later, this portion of Standard Bank’s mortgage book is still struggling to cover its cost of funding. In contrast, Standard Bank’s mortgage loan book has grown by only 30% over the seven years since the GFC, to R325bn, and the weighted average interest rate of new loans is prime plus 0.5%. There is more ‘fat’ in these recent loans, which should mean they perform better through the cycle.
Tail risks
Low valuations and sustainable earnings make banks attractive investments at current prices. However, it may take some time before the market agrees with us and their share prices rise, so we also take into account tail risks, or the small chance of a big loss. Banks are geared businesses and rely on the confidence of their customers who make deposits and funders who buy their debt. This makes them particularly exposed to tail risks: the word bankruptcy is, after all, derived from the Italian ‘banca rotta’, meaning ‘broken bank’.
“A DOWNGRADE WOULD BE NEGATIVE FOR SOUTH AFRICA, BUT WE EXPECT THE EFFECT ON SOUTH AFRICAN BANKS TO BE LIMITED”
The commonly mentioned tail risk is a downgrade of South Africa’s sovereign rating to junk status. It seems probable that this will happen and, when it does, that less foreign money will be invested in South African assets. At best this may result in less demand and lower prices for some South African assets. At worst it could result in a loss of confidence in the South African government’s ability to fund itself at reasonable rates, and a material fall in all South African asset prices.
A downgrade would be negative for South Africa, but we expect the effect on South African banks to be limited. It would not be a surprise and the banks are positioned accordingly. More importantly, our banks are more regulated today than any time in history, most notably via the international regulatory framework known as Basel III. Essentially Basel III requires banks to fund themselves with more equity and more stable debt, and invest in lower risk assets. This reduces the probability of a liquidity event (run on a bank) or insolvency (asset write-downs wipe out equity), making banks more likely to survive tail risks.
Most metrics confirm that SA banks today are less risky than they were prior to the GFC. One of the simpler risk metrics is the assets-to-equity ratio, defined as total assets divided by total equity. A high number means the bank is highly geared, with a lot of debt being used to fund its assets. A small drop in asset value results in a large drop in equity value. Deutsche Bank, for example, has an asset-to-equity ratio of roughly 25x. Each EUR100 of assets is funded by EUR4 of equity and EUR96 of debt. Since equity bears the first loss, a 4% loss in the value of Deutsche Bank’s assets would wipe out all of its equity and result in technical insolvency.
In contrast, the asset-to-equity ratio of South Africa’s big 4 banks is 11.5x, having steadily fallen since the GFC as regulation and reality has reduced risk tolerance. Graph 4 provides a long-term perspective, showing Standard Bank’s asset-to-equity ratio since 1975. Standard Bank is less geared and thus more able to absorb tail risks now than almost any time in history.
The investment case for banks
The best time to buy shares is normally when valuations and earnings are at historical lows, typically following a crisis such as the GFC. Political and economic events caused a mini crisis in the second half of 2015 and, as a result, South African bank valuations have fallen to historic lows. Earnings have not fallen, and there is a risk they may do so should conditions worsen. Despite this, South African banks are well positioned to withstand negative events and their earnings should prove resilient. This combination of low valuations and sustainable earnings makes them attractive long-term investments for our clients.