Last week's appointment of a Federal Compensation Czar represents one of a number of steps the federal government is taking to insert itself into executive compensation - initially at companies receiving federal bailout funds, but potentially much beyond.
One aspect of executive pay that is drawing particularly acute attention is its relationship to institutional risk. As a result, we are seeing a lot of pressure and attention from government officials not only on the levels of executive pay but also on the steps taken - in pay design and structure - to mitigate risk.
Problem is, this is neither as simple or straightforward as it looks. New research from Watson Wyatt brings us an appropriate cautionary note here, as it demonstrates how some compensation elements conventionally believed to aggravate risk may actually do the opposite ... and vice versa.
Watson Wyatt conducted an empirical analysis of the executive compensation architecture at more than 1,000 firms from the S&P 1500 from 2005 to 2007, examining each individual program element in view of the relationship between the executive's realizable pay and the Z-score (a measure available through S&P or Blooomberg which has been historically used to measure credit risk, and which has been shown to be predictive of bankruptcy). The purpose of this exercise was to identify which executive compensation elements acted as "risk mitigators" (i.e., strengthened the relationship between pay and risk reduction, encouraging executives to manage risk more effectively) and which as "risk aggravators" (i.e., weakened the relationship between pay and risk reduction, potentially encouraging executives to take excessive risks). For the most part, the study results contradict widely held beliefs, as the chart below illustrates:
Image Source: Watson Wyatt
Note, by way of example, that one of the design changes which has been pushed by government representatives recently - increasing the proportion of fixed (salary) versus variable (incentive) pay in the overall mix - turns out to be the one least related to high credit risk.
The upshot? I don't know all the details and particulars of the Watson Wyatt research, and there may even be elements of its methodology that are open to challenge. I do believe, however, that it brings home some very important points, not the least of which are:
That executive compensation design is a complex process full of subtleties and unknowns, which demand rigorous study and deep understanding in order to grasp the consequences of different structural actions and restraints, and
That, therefore, there is peril in letting popular opinion - much less political whim - drive mandates on the level and architecture of executive pay. There may also be peril in appointing otherwise intelligent and capable people without deep expertise in executive compensation (and who will, as a result, be forced to rely on their "common sense") to positions of broad oversight in this matter.
The Watson Wyatt study, and the article in yesterday's Wall Street Journal on this topic, also sound a note of caution with respect to the efforts at stamping out risk. Economic recovery requires that companies, whether they have accepted bailout funds or not, produce returns on shareholder investments. Too much emphasis on avoiding risk may eliminate the prudent risk taking that goes hand-in-hand with the kind of growth and innovation that will be necessary to turn these organizations - and the economy at large - around.
More on this to come tomorrow, when the Obama administration releases its proposal for overhaul of financial system regulation.