BID® Daily Newsletter
Sep 13, 2006

BID® Daily Newsletter

Sep 13, 2006

RETURN ON EQUITY


Yesterday we focused our discussion on whether it made sense to use net interest margin (NIM) as a primary performance metric. We discovered that NIM ignores as much as 40% of a bank's net income, missing too much of the picture to add much value. As such (and given it is baseball season), we sent it back down the minor leagues. Today we turn our analysis to return on equity (ROE). As we all know, ROE is calculated by taking revenues minus expenses and dividing it by equity. It was created in an effort to quickly report how much money a bank earned in comparison to the total amount of shareholder equity it held. The theory goes that a bank with a high ROE is more capable of generating cash for investors and therefore is being managed more effectively than peers. In short, the higher a bank's ROE compared to its peers, the better. The other reason ROE was absorbed into the fabric of the banking industry so quickly was because it was also very easy to calculate, understand and explain. In fact, wander into any bank and ask an executive what their ROE (or NIM) is and it will roll right off their tongue in a matter of seconds. It is closely tracked, continually and prominently reported to the board of directors and often used to support compensation structures for executive managers. Given the complexities of the banking landscape that have evolved over the years, however, the value of ROE as a key metric has begun to slip. As with NIM, sometimes keeping things very simple can mean they are too simple to remain relevant. As we will come to see in the discussion that follows, we are not yet ready to send ROE down to the minors just yet, but it is certainly heading in that direction as a top performance metric. The reasons are many, but one of the biggest is that ROE does not incorporate risk in its calculation. As a result, this accounting notion of equity can be a very poor indicator of a bank's risk. Some of the highest ROE banks we know are either in specialty lending categories (credit cards, sub prime, etc.), or have lots of loans in higher credit loss categories (i.e. C&I, construction, etc.). By ignoring the risk of the assets on the books, the value of this metric loses much of its luster. Another problem with ROE is that while it is a bank-wide performance measure, it does nothing for individual business lines or transactions. This leads to a problem where banks assume each successive transaction carries the same risk as the existing portfolio risk. This leads banks to under-price higher risk transaction and over-pricing lower risk transactions. Ignoring specific transactions can also create an environment allowing for greater "exception" pricing. Left unchecked or unmeasured, these can both be damaging to long-term shareholder value. ROE is also a historically-based measurement stick and as such does nothing to incorporate forward looking views. Some of the banks with the highest ROEs one year can go out of business the next. Measuring where a bank has been, with no idea where it may be heading, is not only risky, but may not surface poor strategic decisions. As the graph above and the one from yesterday indicate, large banks can have a NIM 150bp below independent banks, yet still produce an ROE 150bp higher. Tomorrow we will uncover how they are doing this and discuss critical ways independent banks can better compete.
Subscribe to the BID Daily Newsletter to have it delivered by email daily.