Yesterday, we covered what is contained in the Fed's Proposed Guidance on Sound Incentive Compensation Policies. We would like to offer a little commentary on how to prepare for the implementation of such guidance.
Despite the picture, the French-sounding title and much commentary from readers yesterday, the guidance doesn't make banking more socialistic by limiting pay - au contraire. While we are not compensation experts, we have long been fans of helping banks manage on a risk-adjusted basis, since in our eyes, this is the best capitalistic way to allocate resources and manage a business. Different from the press and Congress, we are of the firm belief that very few bankers knowingly took on more risk for personal reward. We are of the belief that many in our industry either didn't have the tools or convinced themselves that the risk was lower than it actually was. Now, this is about to change.
Like every other regulatory guidance, there is no prescribed path when it comes to creating a risk-encompassing compensation policy. Every bank is different, so the regulators are going to largely leave it up to management to decide. After talking with most regulatory bodies yesterday, it is clear that examiners just want to see a logical and encompassing plan that is commensurate with the size and complexity of the organization. That said, since we have been helping banks for the past 3Ys with risk-adjusted compensation performance (officer, loan and deposit profitability) we can offer insight into some basic best practices.
For starters, it is all about attitude. The board and management must make it clear that risk management is of primary importance. This is not to say that low risk is good, as missing out on opportunities is of equal risk. Rather, stewards of capital should understand and be able to manage the risk that the bank is taking. As a general rule of thumb, we always recommend that banks never talk about return unless it is followed by a statement or risk measure within one sentence. This drives home the point that risk and reward are inextricably linked.
Next comes designing a risk-adjusted compensation system. Here a new language is needed. Basing compensation on firm-wide performance, having deferred compensation and dispensing equity in packages are all good and are all recommended in the guidance. However, know that all 3 of these major elements were present with many banks that have already failed including Washington Mutual and IndyMac. The one thing that failed banks usually have in common is the reliance on GAAP methodology to determine risk. The main goal of accounting is the proper recognition of revenue and expenses. While GAAP may overlap with several risk elements (like reserves), it does not answer the question on the quality of the cash flow stream. As a result, using ROA, ROE, NIM or EPS to manage now has little significance when it comes to managing risk/reward.
What managers really want to know is what the return is after taking into account risk. In other words, to start designing a risk-adjusted compensation system, bankers have to start thinking in terms of "economic return" instead of accounting return. If you are a CFO and you have ever purchased a high premium security and quickly amortized down that premium or have embarked on an investment leverage program - most likely you made decisions based on accounting return in exact opposition to economic return.
As we continue with this series over the next couple of days, we will be working with concepts such as risk-adjusted ROE or risk-adjusted return on economic capital. All this means is that we have calculated return, net of the cost of capital and net of expected losses. In this manner, we now have a basis for looking at assets, liabilities and fee income streams to determine the quality and stability of that cash flow. Tomorrow, in an example of how risk-adjusted compensation doesn't place you in the proverbial glass box, we will cover the core question of how to work with the above elements when it comes to a bank's largest risk - credit.