UPDATE: All of our latest thinking can now be found at rewards.aon.com/insights. To read our latest articles, please click here.
By all accounts, 2009 was a year that most of us want to forget. The credit crisis came home to roost, government intervention and regulation reached an all-time high, executives in general and bankers in particular were vilified, and so many things that we took for granted were turned upside down. But, from all this, are there lessons that we can learn? As I reflect on the year, while the challenges are still fresh in my mind, there are a few key points that I hope to use for the future:
The past cannot be the prologue to the future: high water market pay levels earned in 2006 / 2007 may not return in large segments of public companies.
Computer models cannot replace judgment, and, if it seems too good to be true, it probably is.
Risk assessment is difficult work that cannot be delegated.
A balance of power between those paying for services and those providing them is healthy.
The government is incapable of passing legislation to ensure prudent decision making.
Equity, in all its forms, is an excellent way to pay people over the long-term.
Rewarding those who stay at the expense of those who leave is a winning strategy
Pay for performance relationships require a real understanding of true performance and the measurement of risk over a realistic time frame
Performance measurement cannot be abdicated to a simple formula; a more comprehensive review looking at multiple factors against important benchmarks is needed
In order to align pay and performance, pay determination must be extended to match the performance cycle
The Past Cannot Be Prologue to the Future
Most people in this industry believe that their best year (i.e., highest pay) is the truest indicator of their market value. This belief is often validated by the marketplace when people can market themselves to new employers at these peak rates. This cannot continue for a number of reasons.
The first reason is that many financial institutions will never again generate the net earnings that they have in the past. Nearly every financial institution has been forced to add capital, to reduce leverage and to avoid many lucrative lines of business (e.g., securitization, excessive interest rates on credit cards, late fees and penalties). There is also considerable external pressure by regulators, politicians and independent directors to reduce levels of compensation.
Taken together, we believe there will need to be a recalibration of both individual expectations and underlying bonus and LTI targets (e.g., below the 2006 / 2007 high water mark). Any overt indication that pay levels will be permanently reduced may cause employees to test the market. However, most employees will find that there are few takers at an inflated pay rate in this new market environment. There will, of course, be stars who can continue to command extraordinarily high levels of pay as long as they can produce earnings commensurate with their expected levels of pay. The challenge is separating the stars from the rest.
Computer Models Cannot Replace Judgment
Large banks, securities firms and other financial institutions have created a plethora of sophisticated computer models designed to test risk, guide decision making and predict outcomes. The problem is that these models are based on the assumption that future events will be similar to past events and that people and markets both behave rationally. The credit crisis has shown us that those assumptions have been proven to be wrong on more than one occasion. It has often been said that if the outcome looks too good to be true then it probably is.
The credit crisis was fueled by people (e.g., borrowers, lenders, syndicators, asset managers) who all believed that they were getting a great deal despite the obvious fact that many of these loans were being taken by people who could not afford them. The belief that real estate values would continue to rise indefinitely was at best naive and at worst reckless. Real, experienced people need to evaluate the risk/reward tradeoff and not be swayed by computer models that “prove” what we want to believe. A healthy dose of skepticism is required.
Risk Assessment Is Difficult and Cannot Be Delegated
To be done correctly, risk management must be a holistic process. First, all of the elements of risk must be understood (see McLagan Alert, “What Do They Mean by Unreasonable Risk?”), measured and viewed in the context of compounding and interlocking risk. Most importantly, those at the top of house must make it their personal responsibility to know exactly what types of risks are being taken on and that the institution is being properly protected from and/or adequately rewarded for assuming that risk.
The CEO of one of the banks that experienced significant challenges in the credit crisis was quoted as saying that “as long as the music is playing, you’ve got to get up and dance.” Before the credit crisis, there was an attitude that if everyone else was profiting immensely from a particular business or product, your firm should be in that business regardless of your historical risk profile. That attitude should be replaced with a conservative approach that regularly assesses risk, prices it and guides decisions as to be in or out of certain businesses. This can only come from the top.
A Balance of Power between Those Paying for Services and Those Providing Them Is Healthy
The best outcomes occur when both parties have equal power and an equal amount to gain or lose. In the world of compensation, an employee’s power is ultimately derived from the ability to leave for more money. This power is enhanced by the fact that there are few frictional costs to moving - jobs are largely fungible across organizations; there is little reputational risk from changing jobs; deferred awards are typically bought out; and jobs are concentrated in specific geographies making it unnecessary to physically move.
On the other hand, in order to successfully move, the employee must be able to readily transfer his/her skills, book of business and reputation. The destination will also have to have the reputation, products and infrastructure to allow the individual to replicate his/her prior level of production. The current employer needs to determine whether or not the individual is critical to the retention of business and if the individual’s pay demands will reduce margins below acceptable levels. Using this perspective, the best assurance an employer has of holding power in the compensation debate is to institutionalize relationships, build bench strength and clearly understand the individual’s prominence in generating profits.
They can also retain employees by having great products, support systems and keeping pay at competitive levels. If there is parity of power, compensation will be reasonable, but not inflated. The real challenge is tailoring pay in order to ensure that you are not overpaying for production when the production relies on franchise value and not underpaying when the individual talent is driving the production.
The Government Is Incapable of Passing Legislation to Ensure Prudent Decision Making
History is littered with examples of governments’ failures to regulate pay in a way that achieves sensible outcomes (e.g., 162(m), 280(g)). The latest example is the compensation restrictions associated with participation in the TARP program. These regulations have encouraged companies to raise fixed compensation to ridiculous levels thereby reducing both performance-based incentives and the retentive value of equity grants.
Without belaboring the point, simplistic regulations transfer the direction of creative energy from growing the business to finding ways to work around wrong-minded regulations. The bottom line is that we do not want the government as a partner in the design of pay plans. In order to avoid more government involvement in compensation decisions, Boards of Directors should fully exercise their responsibility to ensure that compensation plans do not encourage excessive risk and that pay levels are reasonable in light of true performance. To do the latter, Boards will have to truly understand the nuances of bank performance.
Equity–In All Its Forms–Is an Excellent Way to Pay People over the Long-Term
Depending on where we are in the economic and political cycle, stock options and/or restricted stock have been publicly vilified or praised as successful tools to align shareholder and executive interests. I believe the latter is true over any extended period of time. Paying executives a significant portion of their compensation in equity and encouraging them to hold that equity will, over time, provide motivation to do what is in the long-term interests of the shareholder and will reward them appropriately if they are successful. Failure should have the opposite result.
The problem occurs when we introduce timing into the equations—grants made at high (or low) valuations could result in undesirable outcomes. This problem is mitigated if organizations stay the course and make grants with long vesting cycles every year. It also requires that executives remain with the organization for a long period of time (see below).
Rewarding Those That Stay at the Expense of Those That Leave Is a Winning Strategy
If we want to ensure that we have the right people making the right decisions and that those people have a vested interest in the long-term success of the enterprise, we should systematically over-reward those that stay and disadvantage those employees that leave. The best way to achieve that outcome is to provide for long-term wealth accumulation for those that stay and perform over an extended period of time. This was historically achieved through the use of “career” shares (stock held until retirement) and a generous defined benefit pension plan. For a number of reasons, those programs have fallen out of favor.
Unfortunately, the concept of employee/employer loyalty has diminished as well. One suggestion is to implement a series of programs to encourage/reward the long-term granting and holding of equity. For example, enhance grant guidelines for those that hold all of their after-tax stock grants. Another concept is to match shares for those employees who buy and hold shares. It is true that these actions will result in a grossly undiversified portfolio—but that is true for any owner (and we want our employees to feel and act like owners).
Pay for Performance Relationships Require a Real Understanding of True Performance and the Measurement of Risk over a Realistic Time Frame
Decisions that are made by employees do not all neatly come to fruition in a single calendar year. Because of that, performance cannot always be determined in a calendar year cycle. Employees that have a significant impact on earnings and balance sheet risk should have compensation levels and structures that fully reflect the nature and duration of the risk and return associated with their activities.
The first component is a salary that reflects the true value of the job. The next component is an annual incentive to reflect success of short-term activities (many things can be measured in an annual time frame). The new component is a mid-term incentive where value is posted to an account each year. The amount posted would reflect performance in the current period where the final outcome is not yet determined. Over the subsequent years the account will debited or credited based on actual results. Up to one-third of the account balance can be distributed in any given year. The final piece is the long-term equity component as described above.
Performance Measurement Cannot Be Abdicated to a Simple Formula – a More Comprehensive Review Looking at Multiple Factors against Important Benchmarks Is Needed
Many boards have asked for clear and concise measures to evaluate the success of executive management. They would prefer that the incentive pool be a simple formula reflecting results versus a benchmark. This approach would enable them to avoid messy discussions with management as to why the bonus pool is down. Similarly, they would not have to answer to shareholders as to why they decided what they did. Therefore, many institutions ended up with funding formulas that were based on ROE and EPS growth. The Compensation Committees could then simply check the math on the formula and call it a day.
Unfortunately, it is not that simple. First, performance cannot be adequately measured by one or two financial metrics. It is also important to consider the benchmark—is it budget, last year’s actual results or results relative to peers? Each of these benchmarks has advantages and disadvantages, but taken together they tell a rich story. Couple a robust set of benchmarks with measures that capture the full set of metrics, including operational efficiency, strength of balance sheet, credit quality, shareholder return and risk management (see McLagan Alert “The Brave New World of Executive Compensation”).
This full set of benchmarks and metrics may be confusing, but it is the only way to get the full performance picture. It is the Compensation Committee’s responsibility to understand and evaluate these measures if they are truly going to measure performance. No short cuts are allowed.
In Order to Align Pay and Performance, Pay Determination Must Be Extended to Match the Performance Cycle
While deferrals (typically in equity) have existed for a number of years, as currently designed, they are rather blunt instruments. If you stick around to the end of the performance period (typically three years) you get the stock. More performance-based vesting would go a long way to aligning long-term pay and performance.
The critical point of consideration is the type of performance that is actually being measured: personal, business unit or firm-wide. The devil, as always, is in the details. Measuring firm-wide performance is the easiest and, consequently, the most common performance metric used.
Recently, a few firms have introduced plans that adjust deferred amounts based on business unit performance. They are intriguing plans, but require a considerable amount of administrative support. The only individually adjusted deferrals I am aware of are for stock brokers, whose individual performance is easily quantified.
Organizations may need to take small steps to get a fully functioning pay for performance system. For now, a first step in the right direction (e.g., adding performance vesting conditions) is critically important in the eyes of regulators, the public and most, assuredly, Boards of Directors. But first steps are exactly that—the start of a journey which will include tying pay to real performance over the full performance cycle.
Let’s hope we don’t ever have a repeat of 2009—and one way to do that is to learn from what happened.