Now that Uncle Sam has essentially turned itself into a hedge fund, we thought we would take a moment for a little hedge fund education. If you were a hedge fund manager, and we would argue that you are by running a bank, you would be concerned with "beta" and "alpha." An investment's return is composed of 2 components, beta, the return associated with the entire industry and alpha, the return associated with the particular investment. In theory, the return on any asset, no matter how exotic, can be broken down into both alpha and beta by using a suitable benchmark and a statistical regression. If your bank tracks the industry, then your return is 100% beta. For last quarter the bank industry produce a return of 1.46% on its equity. To the extent that a bank's return was above that, then that amount is attributed to alpha.
Positive alpha is what hedge funds strive for and is why some hedge fund managers earn the big bucks. When evaluating an investment in a hedge fund (or a bank), it is always helpful to determine how alpha is derived. Some managers derive alpha from asset allocation, some from market timing, others from a combination of timing and rebalancing. For banks, the picture is more complicated, as in addition to asset management, alpha can be also be derived from the structuring or pricing of liabilities, geography, the generation of fee income or by superior strategic management encompassing all of the above. The beauty of banking is that banks are semi-homogenous and somewhat transparent; both factors that make for easy analysis. By taking each banking component, we can regress the data to pre-tax operating income and discover what factors are important to help banks produce a return above beta.
Looking at quarterly data going back to 2002, banks that had the greatest influence on above average return managed credit quality and their associated returns. This is an important statement that helps explain our current situation of low earnings as an industry. In 2004 and 2005, banks produced the most alpha by reducing ALLL. Conversely, banks are currently producing the largest negative alpha by adding to their ALLL. If you normalize for ALLL, then a slightly different picture emerges.
Holding loan loss reserves constant, the best way to increase alpha in this environment is to reduce non-interest expense. Put another way, if you are looking for the single biggest thing you can do to affect 4Q earnings, reduce overhead expenses.
While reducing expenses has the greatest short-term affect on operating income (next to reducing loan loss reserves), loan asset allocation and then pricing, are the next 2 most correlated to earnings over the past 6Ys. Behind that, liability structure and liability pricing come next. Note that while a bank derives most of its economic value from liabilities, managing loans has a greater influence on earnings. Also note that while pricing is important, choosing the amount of exposure to individual loan and deposit sectors is a greater driver of earnings. For example, low construction exposure contributed to positive alpha in both an up and down market. On the other side of the balance sheet, reducing CD issuance and wholesale funding also produced outsized alpha (as did having higher balance in checking and well-priced money markets). After that, banks have a myriad of items such as fee income generation, geography, investment portfolio performance and asset-liability position that all have a less than 5% correlation to earnings.
There are many other items that we did not get around to measuring (such as strategic positioning, culture, brand, ownership of branches, etc.) so this is by no means a comprehensive list. Further, every bank is different and some banks are more earnings sensitive to certain items (such as liability pricing), than others. Hopefully, by getting a rough idea of how to produce positive alpha, management may have more control over future returns.