During the last 12 months, pay structures for bank staff have changed dramatically. There have been significant increases in base salaries for some firms and changes to the delivery of incentive compensation. The decision making for these changes has been directed at perceived risk takers in the front offices of banks. However, these changes have also impacted the back offices, i.e., the infrastructure functions. As firms are now beginning the planning process for year-end compensation, should they also reassess the way that the bonus pool for infrastructure functions is calculated and distributed?
How Much is Left?
Historically, there have been three approaches to calculating a bonus pool for infrastructure functions, none of which were particularly scientific!
The first, and more rigorous, approach is to take the previous year’s pool as a starting point and then try to adjust to take care of leavers and joiners. This “bottom up” approach often includes modeling scenarios of an up (or down) percentage based on business performance (though the volatility is typically much less than for front office groups);
The second approach is target based plans, typically represented as a percentage of base salary. However, there is often no “formal” performance based funding associated with this model and incentive awards can often become more “entitlement” based than performance based;
The third approach is the least structured and has tended to be more prevalent at volatile organizations where the infrastructure groups have less representation at senior levels. Here, and particularly in bad years, the approach to the infrastructure bonus pool is very much of the “How much is left after we pay the performers (front office divisions)?”
In the current environment where there is competitive pressure on recruiting and keeping infrastructure staff, and where, for example, some IT staff are as important to the business as the traders themselves, there is now increasing demand from infrastructure groups to increase transparency around the incentive process and increasingly factor individual / group performance measures into pool determination. Beyond IT, there is increased emphasis on the impact of control functions (Risk, Audit, and Compliance). In a rapidly evolving business environment, where in some cases avoiding loss becomes a focus comparable to generating revenue, it may no longer be sensible to pay these key areas with whatever is left over.
Aligning Reward with Performance
Having agreed upon the importance of these groups and the desire to pay them appropriately, the next logical question is how do we build some science around this? It seems easy (or easier) in the front office to measure production, even in an environment where new metrics are being reviewed (risk-weighted assets, cost of capital, etc.). What follows is some thinking on how to do this for infrastructure:
Begin by measuring and benchmarking actual infrastructure payout ratios in the market.
Let’s assume at Firm XYZ the historic rate has been 7%. We find that the benchmark rate for the broad market is 9%.
Next, assess mix of products supported, location strategy, alternate sourcing, use of contractors, etc., as these will all influence this number. For example, you may need to fund an infrastructure group supporting a complex derivatives platform at a higher rate.
While the overall market rate is 9%, based on the complex nature of Firm XYZ’s mix of business and the density of staff in high cost locations, we adjust the number up to 11%. (At this time, we may also consider the use of lower cost locations, automation vs. people strategy, etc.)
Once you have established an overall benchmark, and adjusted for the above factors, you must measure for the “performance” of the infrastructure group: for example, how many heads does the group support relative to other organizations? Consider also the complexity of the products they support. Consider how efficiently this is done, as per the metrics in the diagram below. Consider funding relatively small or relatively inexpensive groups at a higher percentage, if they are effective.
With the adjusted funding rate at 11% based on mix of products, we now see that Firm XYZ has a lean staffing model—their front office to back office ratio indicates that a small number of employees are doing a lot of work, relative to the size of the front office effort. To recognize this, and fund adequately for the kind of talent it requires to support this, we increase the funding rate to 12.5%.
Once you have taken the starting benchmark and adjusted for the considerations listed above, you will then have a good feel for an appropriate funding rate with a method as scientific as the ones used for front office pools.
We now take the 12.5% adjusted benchmark back and look at the typical funding requirements for the front office lines of business. There is no guarantee that the 12.5% funding is available, but now we have the typical multi-line funding “bid-ask” process where each division can make a case for their share of the pie based on their contribution, and the market validated funding rates that are typical for their areas.
Other Less Common Practices Worth Considering
Independent budgets for shared service groups. At a number of organizations, shared service groups have been set up in a quasi-autonomous fashion with rigorously tracked expense budgets. In these cases, the bonus pool is already set at the start of the year. However, issues then arise if year-on-year expense is over budget (pool reduction) or even under budget (generally passed back to the host organization rather than reallocated as additional incentives).
Formal linkage of infrastructure pools to profitability. Linkage of pools to overall firm profitability or return on equity (RoE) is becoming popular. In all cases, there is both an upside and a downside cap. Methodologies vary for calculating the increase/decrease, but the differential is always less than the underlying performance, e.g., a 20% increase in RoE would result in an 8% increase in the pool.
Forced performance ranking to allocate bonus dollars. Within a target-based approach, firms have attempted a number of methods to allocate dollars across performance groups. Simple forced ranking is difficult to achieve whereas a formal appraisal methodology where all staff at a particular level are ranked into bands, e.g., 10% A, 30% B, 30% C, 20% D, 10% E, is more successful. This also simplifies the funding/accrual process.
Reward groups based on their own performance. The relative performance of infrastructure groups is difficult to measure but in at least one organization, an internal ‘client satisfaction survey’ has been used to differentiate between the relative performance of infrastructure groups. This has meant that even in a bad revenue year, a high performing group has been funded well.
Rewarding cost savings. The most tangible way in which infrastructure performance can be measured is by cost reduction. In the case where significant savings have been made, e.g., by implementing new technology or by better negotiating external fees, there is definite potential to link incentive compensation to savings, e.g., 10% of savings added to the pool. Alternatively, the savings made could be reinvested in additional reengineering.
With historic priorities on front office pay and longstanding usage of the previous year’s pool in determining infrastructure compensation, there needs to be a significant push for change. Nevertheless, the radical changes in financial services pay over the last 12 months provide firms an opportunity to implement a better bonus pool process for their Infrastructure staff.
At the same time, regulatory demands toward bonus deferral are potentially going to hit some infrastructure staff who are not risk takers or risk managers. Year-on-year cash income for some employees is going to be significantly impacted.
Change needs to be considered carefully, taking into account benchmarking of both compensation amounts and practices to ensure that reward (and how it is delivered) is competitive. Additionally, it may be timely and advantageous to review grading / salary and pay mix structure to support the bonus pool calculation method.
The firms that adopt change will promote a fairer culture, reward individual and group contribution and thus motivate and increase staff retention. Failure to change will result in the continuing uncertainty in infrastructure groups as to what will be available at year-end and the inability to justify and protect bonuses when, for example, trading or loan losses eat into profitability.