Allocation of Risk Capital in Financial Institutions

Financial Management (Financial Management Association), Autumn, 1999 by Francesco Saita

The example clearly shows that different criteria lead to different profitability rankings. Considering the return on utilized CaR, the fixed-income-trading unit has the best performance, while FX has the highest return based on allocated CaR.

But most importantly, meeting profitability targets based on utilized CaR is not sufficient to meet profitability targets at the corporate level. Despite the fact that each individual unit is able to meet the 30% target on utilized CaR, the return on the total amount of capital allocated to the dealing room is inadequate (23.81%) because of unutilized CaR, since

RAROC on Total CaR = RAROC on utilized CAR(1-excess CaR/total CaR) (4)

Therefore, establishing profitability targets based on utilized CaR can have the effect of eliminating each unit's responsibility to exploit fully the resources that have been assigned to it by corporate headquarters on behalf of the bank's shareholders. [15]

How can the problem of choosing between utilized CaR and allocated CaR be solved? Should cautious units be punished, and full capital usage encouraged, or vice versa? A solution can only be found by considering the way the capital-allocation process operates. In fact, the greater the influence of individual units on CaR assignments (as in the case of an internal market for capital), the greater their responsibility for the capital they have requested. Therefore, units should be evaluated on allocated CaR, since otherwise they would have an incentive to request as much CaR as possible, while taking no responsibility for its efficient usage.

On the other hand, when capital-at-risk is allocated in a rigidly centralized, top-down fashion, individual units should not be blamed for not utilizing an amount of capital that may be disproportionate to the risk-taking opportunities they face. Adopting utilized CaR would therefore make it clear that the potential gap between the results of individual units and aggregate bank performance is caused by an excess of capital whose responsibility lies with top management alone.

An intermediate solution would be to adopt utilized CaR together with a penalty on unutilized CaR. This would combine ex ante CaR limits and ex post CaR usage into a single performance index. When a negotiation process is used, the "penalty rate" on unutilized CaR can be set at any level between zero and the target return on capital, depending primarily on how much the business units can influence the amount of capital they receive. The lower the penalty rate, the closer the effects to the utilized CaR solution. In the opposite case, if the penalty rate were set equal to the target return, then allocated CaR would be fully charged to each business unit. The level of the penalty rate could also depend on individual units' characteristics. Since forecasting CaR needs is more difficult for some of them than for others, a comparatively lower penalty rate would be appropriate.

It is useful to point out that the problem of choosing between utilized CaR and allocated CaR is not only interesting per se but is also an example of the approach that is needed in order to solve risk management implementation issues. Organizational theory alone cannot provide a clear-cut solution to the problem concerning the transformation of a daily CaR measure into a monthly or yearly equivalent. However, combining it with financial theory is critical to understanding when and for which purpose a given measure is optimal, and makes clear why this is more important than determining which is the best measure of capital to use.